Multimodal Imaging of Choroidal Tumors: What Ultrasonography Still Uniquely Provides

Abstract

Choroidal tumors encompass a heterogeneous group of lesions ranging from benign melanocytic nevi and vascular tumors to malignant melanomas and metastatic lesions. Their evaluation increasingly relies on multimodal imaging combining techniques that interrogate different physical and biological properties of the lesion and surrounding tissues. This narrative review examines the principles, clinical contributions, and limitations of optical and acoustic imaging in the assessment of choroidal tumors. Modern optical modalities—including spectral-domain and swept-source optical coherence tomography (OCT), OCT angiography (OCT-A), fundus autofluorescence, ultra-widefield imaging, and scanning laser ophthalmoscope (SLO)-based techniques—provide high-resolution information on lesion morphology where optically accessible, as well as on associated retinal pigment epithelium alterations, subretinal fluid, photoreceptor changes, fluorophore-related signals, and vascular flow or permeability. However, many of these findings reflect responses of the retina, RPE, or choroidal vasculature adjacent to the tumor rather than intrinsic tumor characteristics. Ultrasonography provides different information based on acoustic backscatter and attenuation, including lesion dimensions, internal reflectivity and echo patterns, and remains particularly important for larger or peripheral lesions and when optical visualization is limited by media opacity. Its interpretation and measurement are nevertheless operator- and scan-dependent, with limitations related to spatial resolution, insonation geometry, boundary definition, and reproducibility, particularly for small lesions. Representative examples of choroidal nevi, melanomas, circumscribed choroidal hemangiomas, metastases, and selected pseudotumoral lesions illustrate how multimodal findings contribute to differential diagnosis, longitudinal assessment, and clinical decision-making. Emerging techniques such as OCT-A and Retro mode imaging provide additional structural or flow-related information, although their tumor-specific diagnostic performance remains incompletely validated. Multimodal imaging can substantially narrow the differential diagnosis but cannot always establish tissue diagnosis, and biopsy or systemic investigation may remain necessary in selected cases. Accurate evaluation of choroidal tumors therefore depends on integrating the distinct information and limitations of each modality with clinical findings and longitudinal evolution.

1. Introduction

1.1. Choroidal Tumors and the Evolution of Multimodal Imaging

Choroidal tumors represent a relatively uncommon but clinically critical group of disorders in which both visual function and, occasionally, patient survival may be compromised [1,2]. The diagnosis of a choroidal mass rarely relies on a single examination. Instead, it emerges from a combination of structural, vascular, optical, and acoustic findings obtained through multimodal imaging [3,4]. In many cases, longitudinal evolution further contributes to diagnostic confirmation, particularly when differentiating benign choroidal nevi from small melanomas [5,6].

Over the past two decades, retinal imaging has undergone major technological evolutions. Spectral-domain optical coherence tomography (SD-OCT), swept-source OCT (SS-OCT), OCT angiography (OCT-A), ultra-widefield imaging, and scanning laser ophthalmoscope (SLO)-based platforms have substantially expanded the evaluation of retinal and choroidal diseases [7,8,9,10,11]. These modalities provide high-resolution visualization of the retina, retinal pigment epithelium (RPE), choroid, and retinal or choroidal vasculature. In ocular oncology, OCT-based imaging has progressively complemented and, in selected situations, challenged ultrasonography for the evaluation of small choroidal lesions and the characterization of subtle retinal and RPE alterations associated with choroidal tumors [12,13,14].

Despite these advances, ultrasonography remains essential in the evaluation of choroidal tumors. Unlike purely optical imaging techniques, ultrasonography provides information not only on lesion thickness and shape but also on internal reflectivity, internal echo patterns, and acoustic attenuation characteristics [15,16]. These acoustic features contribute to the differential diagnosis between melanomas, hemangiomas, metastases, and other choroidal lesions [17,18]. Ultrasonography also remains essential when optical visualization is limited by media opacity [19].

Other imaging modalities, including color Doppler imaging (CDI) and orbital magnetic resonance imaging (MRI), may provide valuable complementary information in selected clinical situations. Color Doppler imaging enables assessment of intratumoral and retrobulbar blood flow, whereas dedicated ocular MRI offers high soft-tissue contrast and may improve the evaluation of ciliary body involvement, optic nerve invasion, extrascleral extension, and treatment planning, particularly in complex or posteriorly located lesions [20]. Nevertheless, these techniques are generally reserved for selected cases and often require specialized expertise and multidisciplinary interpretation [21,22]. Because the present review focuses on multimodal imaging techniques routinely used in ophthalmic practice, detailed discussion of Doppler ultrasonography and MRI falls beyond its scope.

1.2. Why No Single Modality Is Sufficient

Importantly, multimodal imaging does not simply provide redundant information. Different imaging techniques evaluate distinct biological and physical properties of choroidal tumors. Some modalities primarily characterize the intrinsic structural or acoustic properties of the lesion itself, whereas others mainly assess secondary retinal, retinal pigment epithelium, choroidal, or vascular responses induced by the tumor.

Optical imaging depends on light reflection, scattering, fluorescence, and absorption properties, whereas ultrasonography relies on acoustic impedance differences and sound-wave attenuation within tissues [17,18]. These fundamental physical differences explain why certain lesions exhibit characteristic optical or angiographic signatures, while others are more effectively characterized through their acoustic behavior. Optical imaging techniques, particularly OCT and OCT-A, provide high-resolution visualization of retinal architecture, subretinal fluid, photoreceptor alterations, lipofuscin deposition, and retinal or choroidal vascular remodeling [13]. In contrast, ultrasonography provides assessment of internal reflectivity patterns, internal echo patterns, and acoustic attenuation.

In parallel with OCT and angiographic techniques, newer SLO-based imaging modalities such as Retro mode imaging have attracted interest for the evaluation of retinal and choroidal lesions [23]. By using indirect retroillumination through laterally displaced or annular apertures, Retro mode imaging enhances the perception of retinal and subretinal topography and may improve delineation of some pigmented choroidal lesions [24].

1.3. Purpose of the Review

The purpose of this review is to discuss the complementary roles of ultrasonography and modern optical imaging techniques in the characterization of choroidal tumors. Particular attention is given to the physical principles underlying image formation, the multimodal evaluation of melanomas, hemangiomas, nevi, and metastases, and the secondary retinal and choroidal responses associated with these lesions. Emphasis is also placed on the specific diagnostic information that ultrasonography continues to uniquely provide despite major advances in retinal imaging technologies.

Rather than providing an exhaustive catalog of all choroidal tumors and pseudotumoral conditions, this narrative review focuses on representative lesions that illustrate the complementary information provided by optical imaging and ultrasonography, while selected mimickers are discussed to highlight important diagnostic pitfalls.

1.4. Literature Search and Review Approach

This article was designed as a narrative review and expert perspective rather than a systematic review. Relevant literature was identified primarily through searches of PubMed/MEDLINE, supplemented by reference lists of selected articles and additional publications known to the authors. Searches were performed during manuscript preparation and updated in August 2026. No publication date restriction was applied, because both seminal historical publications and recent developments in multimodal imaging were considered relevant to the scope of the review. The search focused on publications addressing multimodal imaging of choroidal tumors, ocular ultrasonography, OCT and OCT angiography, fundus autofluorescence, angiography, infrared and Retro mode imaging, and the differential diagnosis of choroidal lesions. Search terms included combinations of “choroidal tumor”, “choroidal melanoma”, “choroidal nevus”, “choroidal hemangioma”, “choroidal metastasis”, “ocular ultrasonography”, “A-scan”, “B-scan”, “OCT”, “OCT angiography”, “fundus autofluorescence”, “Retro mode”, and “multimodal imaging”. Priority was given to clinically relevant original studies, major reviews, and seminal publications describing the physical principles or characteristic imaging features discussed in this article. Additional recent publications were included when they provided relevant information on emerging imaging modalities or tumor-associated retinal and choroidal changes. Because this is a narrative review, study selection was not based on predefined systematic-review eligibility criteria, and the literature cited should not be considered an exhaustive survey of all publications in the field.

Clinical images included in this review were selected from the authors’ clinical archives for illustrative purposes and were fully anonymized before inclusion in the manuscript. Written informed consent had been obtained from the patients, authorizing the use of anonymized clinical data and imaging for research and publication purposes. Images or schematic illustrations derived or adapted from previously published or publicly available sources are identified individually in the corresponding figure legends, with citation of the source.

2. Light and Ultrasound: Different Physical Principles

2.1. Optical Imaging and Tissue Reflectance

Most retinal imaging techniques are based on the interaction between light and ocular tissues. Color fundus photography, scanning laser ophthalmoscopy (SLO), fundus autofluorescence, OCT, and OCT angiography all rely on reflection, scattering, absorption, or fluorescence phenomena occurring within ocular tissues and at tissue interfaces (Figure 1) [25,26].

Light is a transverse electromagnetic wave that does not require a material medium for propagation. Optical reflectivity depends primarily on differences in refractive indices between adjacent tissues [26]. In retinal imaging, these interfaces include structures such as the internal limiting membrane, photoreceptor layers, retinal pigment epithelium (RPE), Bruch’s membrane, and the choroid [27].

The wavelength of incident light strongly influences tissue penetration and image quality. Shorter wavelengths are more strongly scattered, whereas longer wavelengths penetrate more deeply into biological tissues. This principle explains why swept-source OCT (approximately 1050 nm) provides improved visualization of the choroid compared with spectral-domain OCT operating near 840–870 nm (Figure 2) [28].

In OCT, image formation relies on low-coherence interferometry and the analysis of backscattered photons [27]. Structures with high optical backscattering generate hyperreflective signals, whereas optical shadowing may occur beneath pigmented lesions, hemorrhage, subretinal fluid, or dense fibrotic tissue because of attenuation of the incident and reflected light beams.

Importantly, optical imaging frequently evaluates not only the lesion itself but also the biological response of adjacent tissues. In choroidal tumors, multimodal optical imaging may reveal subretinal fluid, “shaggy photoreceptors,” lipofuscin accumulation, retinal pigment epithelium dysfunction, and retinal vascular leakage [8,9,13]. These secondary retinal and RPE alterations often contribute more to malignancy assessment than the intrinsic optical appearance of the tumor itself.

2.2. Ultrasonography and Acoustic Reflectivity

In contrast to light, ultrasound consists of longitudinal mechanical waves that require a material medium for propagation. Ultrasonographic imaging depends on differences in acoustic impedance between tissues [29]. At each tissue interface, part of the incident acoustic energy is reflected or backscattered toward the transducer, while another part is transmitted, scattered in other directions, or absorbed within the tissue (Figure 3) [15].

The echoes returning to the transducer generate the ultrasonographic signal. In ocular oncology, internal reflectivity refers to the amplitude and distribution of echoes arising from within a lesion and therefore depends on the number, size, acoustic contrast, and spatial organization of internal scattering interfaces. Acoustic attenuation, in contrast, describes the progressive loss of acoustic energy as the wave propagates through tissue and results principally from absorption and scattering [15]. Internal reflectivity and attenuation are therefore related but distinct acoustic properties and should not be considered interchangeable.

Ultrasound frequency strongly influences both spatial resolution and tissue penetration. Higher-frequency probes, such as 20 MHz probes, provide improved spatial resolution but greater attenuation and therefore reduced penetration compared with lower-frequency probes. Consequently, probe selection depends on lesion size, location, and the structures being evaluated. These distinctions are particularly relevant in choroidal tumors (Figure 4 and Figure 5).

Choroidal melanomas characteristically demonstrate low-to-medium internal reflectivity, reflecting relatively limited internal acoustic backscatter. On B-scan imaging, relatively limited internal backscatter may produce acoustic hollowness, defined as a relative paucity of internal echoes within the lesion. Standardized A-scan, in which echo amplitudes are evaluated under standardized acquisition and calibration conditions, typically demonstrates decreasing echo amplitudes from the anterior portion of the lesion toward its base. This progressive decline is represented by the κ angle, with a larger κ angle indicating a more pronounced reduction in echo amplitude through the lesion [30].

These findings are associated with the relatively uniform acoustic organization of many melanomas, although they should not be interpreted as a direct surrogate for histologic cellular homogeneity.

In contrast, circumscribed choroidal hemangiomas usually demonstrate high internal reflectivity, with relatively strong echoes persisting throughout the lesion [30]. Their vascular architecture provides numerous acoustic interfaces capable of generating internal backscatter. Standardized A-scan therefore typically shows high-amplitude echoes with a smaller κ angle and comparatively limited decline in echo amplitude toward the lesion base. Choroidal metastases more often exhibit intermediate or heterogeneous internal reflectivity, consistent with their variable internal architecture and stromal composition.

Importantly, low internal reflectivity, acoustic attenuation, and acoustic hollowness describe different aspects of the ultrasonographic signal. Likewise, posterior acoustic shadowing should not be equated with low internal reflectivity. In highly reflective calcified lesions, marked posterior shadowing results predominantly from strong reflection and attenuation of the incident acoustic beam at or within the lesion. In melanomas, by contrast, acoustic hollowness reflects relatively low internal backscatter, while the progressive decrease in echo amplitude with depth reflects attenuation during propagation through the tumor. Together, these acoustic features provide information about lesion organization, but they do not constitute a direct histologic characterization.

Regarding the presentation of ultrasonographic images in this review, images included in composite multimodal figures were selected to illustrate characteristic lesion morphology and, where appropriate, were cropped or enlarged to facilitate comparison with corresponding optical imaging findings. Such image presentation was performed for illustrative purposes only and does not represent a standardized quantitative acquisition or measurement protocol. Unless otherwise specified, the high-resolution B-scan images shown in these figures were acquired with the AVISO system (Quantel Medical, Cournon-d’Auvergne, France) using a 20-MHz probe.

2.3. Why These Physical Differences Matter Clinically

The different physical principles underlying optical and acoustic imaging determine the type of information provided by each modality. OCT can directly visualize accessible portions of choroidal lesions while providing high-resolution assessment of associated retinal, RPE, and subretinal changes. OCT-A and angiographic techniques provide flow-related or dye-based vascular information, whereas fundus autofluorescence depicts the distribution of endogenous fluorophores and associated RPE alterations.

Ultrasonography provides lesion dimensions and acoustic characteristics, including internal reflectivity, echo patterns, and attenuation, and remains applicable when optical visualization is limited by media opacity or lesion thickness [17]. These differences explain why multimodal evaluation may combine information that cannot be obtained from a single imaging technique. The principal characteristics, limitations, reproducibility issues, and clinical contributions of the imaging modalities discussed in this review are summarized in

Table 1. Comparative characteristics of the main multimodal imaging techniques used in the evaluation of choroidal tumors, including the physical basis of the detected signal, principal limitations and artifacts, reproducibility issues, and contribution to clinical assessment. Optical and acoustic modalities provide distinct but complementary information and differ in spatial resolution, tissue penetration, field of view, susceptibility to media opacity and artifacts, and measurement reproducibility.

Imaging Modality	What is Measured or Detected	Main Limitations, Artifacts, and Reproducibility Issues	Principal Contribution to Clinical Assessment
Color fundus photography/ultra-widefield imaging	Reflected visible-light signal documenting lesion color, pigmentation, surface features, margins, and associated retinal changes.	Two-dimensional representation; color and contrast depend on illumination, acquisition system, media clarity, and pigmentation. Peripheral distortion and differences between imaging systems may affect apparent lesion dimensions and longitudinal comparisons.	Baseline documentation of lesion appearance and margins; assessment of pigmentation, orange pigment, hemorrhage, and associated retinal changes; photographic documentation of growth during follow-up [31,32].
Near-infrared reflectance/multicolor imaging	Reflected and scattered light at near-infrared or multiple wavelengths, providing contrast related to retinal, RPE, and choroidal structures.	Signal depends on pigmentation, tissue depth, wavelength, and device-specific acquisition and processing. Reflectivity differences are not specific for tissue composition.	May improve visualization of lesion boundaries and subtle reflectivity differences and complements color photography and OCT for lesion localization and multimodal documentation [33,34].
Fundus autofluorescence (FAF)	Endogenous fluorophore-derived signal, predominantly related to lipofuscin and melanolipofuscin within the RPE, and influenced by absorbing structures.	Provides indirect rather than direct information on tissue biology. Signal is influenced by fluorophore composition, masking, pigmentation, media opacity, excitation wavelength, and acquisition system.	Documents tumor-associated RPE alterations and orange pigment; provides indirect information regarding RPE status and contributes to multimodal risk assessment of melanocytic lesions [3,35].
Fluorescein angiography (FA)	Time-dependent fluorescence of intravenously administered fluorescein, demonstrating vascular filling, leakage, staining, pooling, and blockage.	Invasive and time-dependent acquisition; late leakage may obscure lesion boundaries. Limited penetration through pigment, hemorrhage, and RPE reduces visualization of deeper choroidal structures.	Characterizes leakage and secondary retinal vascular or exudative changes; contributes to differential diagnosis and assessment of lesion-associated retinal consequences [34].
Indocyanine green angiography (ICGA)	Near-infrared fluorescence of intravenously administered indocyanine green, providing information on choroidal vascular filling and lesion-associated vascular patterns.	Invasive and time-dependent acquisition; interpretation depends on acquisition phase and lesion characteristics. Provides vascular rather than direct histologic information.	Characterizes intrinsic choroidal vascular patterns and may substantially narrow the differential diagnosis, particularly for vascular lesions such as circumscribed choroidal hemangioma [34,36].
Structural OCT/EDI-OCT/swept-source OCT	Optical backscatter providing high-resolution cross-sectional and en face structural information from the retina, RPE, choroid, and optically accessible portions of choroidal lesions.	Signal attenuation and shadowing from pigmentation or lesion thickness; segmentation artifacts; incomplete penetration through large or highly attenuating lesions. Measurements depend on scan orientation and visibility of lesion boundaries.	High-resolution assessment of lesion contour and secondary retinal/RPE changes; sensitive detection of shallow subretinal fluid, photoreceptor alterations, and hyperreflective deposits; useful for characterization and longitudinal monitoring of small lesions [8,12].
OCT angiography (OCT-A)	Motion-contrast signal generated predominantly by moving erythrocytes within predefined segmentation slabs, providing flow-related vascular information without intravenous dye.	Segmentation and projection artifacts, shadowing, motion artifacts, slab selection, and flow-detection thresholds may alter apparent vascular patterns. Quantitative measurements are device-, segmentation-, and processing-dependent, and absence of detectable signal does not necessarily indicate absence of vessels.	Provides noninvasive assessment of flow-related vascular organization and may complement structural OCT in lesion characterization. Quantitative differences have been reported among lesion types, but validated tumor-specific diagnostic thresholds remain limited [37,38].
Retro mode imaging	Indirectly scattered near-infrared light collected through laterally displaced (DL/DR) or annular (RA) apertures, generating shadow-based topographic contrast.	Platform-specific and predominantly qualitative. Apparent pseudo-relief depends on aperture geometry and illumination direction and does not represent true three-dimensional topography. Tumor-specific evidence remains limited and validated diagnostic thresholds are lacking.	May enhance visualization of lesion borders, surface contour, and associated topographic or pigmentary alterations. At present, it should be considered a complementary rather than decision-defining modality in ocular oncology [23,24].
B-scan ultrasonography	Spatial distribution and amplitude of returning ultrasound echoes, providing information on lesion configuration, dimensions, internal echogenicity, and acoustic attenuation within the acquired scanning plane.	Operator- and orientation-dependent. Oblique insonation, boundary selection, inclusion of retina, sclera, or adjacent subretinal fluid, acoustic shadowing, and limited resolution for small lesions may affect measurements. Conventional two-dimensional examinations sample selected planes rather than the entire lesion volume.	Measures lesion thickness and configuration; evaluates lesions when optical visualization is limited by opaque media; assesses peripheral or larger lesions and possible extrascleral extension; contributes to differential diagnosis, treatment planning, and longitudinal follow-up [17,39].
Standardized A-scan ultrasonography	Echo amplitude as a function of tissue depth under standardized acquisition conditions, providing an internal acoustic reflectivity profile.	Requires standardized acquisition and calibration and remains dependent on probe alignment and insonation angle. Interpretation may be influenced by lesion size and internal heterogeneity.	Provides acoustic tissue characterization complementary to B-scan morphology. Internal reflectivity patterns may contribute to differentiation among melanoma, hemangioma, metastasis, and pseudotumoral lesions when interpreted within the complete multimodal examination [18,30].

.

3. Multimodal Imaging of Tumor-Associated Tissue Responses

3.1. Retinal Pigment Epithelium Distress and Retinal Response

One of the major contributions of modern retinal imaging in ocular oncology has been the ability to visualize subtle secondary retinal and retinal pigment epithelium (RPE) alterations associated with choroidal tumors [8]. Importantly, many imaging findings considered suggestive of malignancy do not correspond directly to the tumor itself but rather to the biological response induced within adjacent tissues [6].

In melanocytic lesions, chronic or progressive dysfunction of the RPE may produce drusen formation, pigmentary alterations, lipofuscin accumulation, subretinal fluid, and photoreceptor abnormalities [8,40]. Optical coherence tomography (OCT) frequently reveals shallow subretinal fluid that may remain clinically inapparent, together with elongation of photoreceptor outer segments, disruption of the ellipsoid zone, and the characteristic appearance of “shaggy photoreceptors” [8]. These findings reflect retinal and RPE alterations associated with the lesion and may contribute more strongly to malignancy assessment than the intrinsic morphology of the tumor itself [6].

Fundus autofluorescence (FAF) provides a fluorophore-derived signal that may indirectly reflect the functional and metabolic status of the RPE, particularly through changes related to lipofuscin and melanolipofuscin accumulation [35]. Clinically visible orange pigment overlying a melanocytic lesion is frequently associated with increased autofluorescence and represents an established risk feature for growth of a choroidal nevus [40]. However, FAF detects the distribution of endogenous fluorophore-related signals and does not by itself establish a unique cellular or molecular source for the clinically observed orange pigment. Conversely, reduced autofluorescence may occur in areas of chronic RPE atrophy or loss.

Fluorescein angiography provides information on both tumor-associated vascular phenomena and secondary retinal and RPE alterations. Progressive leakage and pinpoint hyperfluorescent spots (“pinpoints”) may reflect altered vascular permeability and RPE dysfunction, whereas pooling corresponds to dye accumulation within fluid-containing spaces such as a serous retinal detachment. Progressive tumor hyperfluorescence or staining may additionally reflect vascular and interstitial characteristics of the lesion itself. These angiographic patterns should therefore not be interpreted exclusively as secondary manifestations of RPE dysfunction [37,40].

These observations illustrate an important concept in multimodal ocular oncology: many modern imaging techniques evaluate the consequences of the lesion on surrounding tissues rather than directly characterizing the tumor itself [8].

3.2. Choroidal Vascular and Tissue Changes Associated with Choroidal Tumors

Recent advances in enhanced-depth imaging OCT, OCT angiography (OCT-A), infrared imaging, and multimodal retinal imaging have improved visualization of structural and vascular changes associated with choroidal tumors, including alterations involving the overlying retinal pigment epithelium, outer retina, and choroidal vascular architecture (Figure 6, Figure 7 and Figure 8).

In choroidal nevi, structural OCT may reveal localized hyperreflective alterations overlying or adjacent to the lesion, while OCT-A may demonstrate spatial variations in flow-related signals. Three-dimensional structural OCT studies have also illustrated complex spatial relationships between small melanocytic choroidal tumors and larger choroidal vessels, including persistent intratumoral vascular channels, displacement or encasement of vessels, and focal vascular obliteration [41,42]. These structural observations should not be interpreted as direct evidence of choriocapillaris remodeling or altered blood flow.

Figure 8 summarizes a conceptual framework for interpreting the spatial relationship between structural OCT and OCT-A findings. Increased structural OCT reflectivity may result from altered optical backscattering related to changes in tissue interfaces or organization, whereas localized OCT-A signal abnormalities may be associated with differences in vascular position or detectable blood flow. Whether these findings represent tissue compression, vascular displacement, choriocapillaris reorganization, or other biological processes cannot be determined from cross-sectional imaging alone. These mechanisms should therefore be regarded as hypotheses requiring longitudinal and histologic validation.

Quantitative OCT-A analyses have reported lower vascular density, vessel length, and vascular complexity in choroidal melanomas compared with benign nevi [38]. These cross-sectional differences represent associations with lesion type and should not be interpreted as evidence of progressive microvascular change during tumor evolution. OCT-A measurements are also influenced by segmentation accuracy, slab selection, projection artifacts, shadowing, flow-detection thresholds, and other acquisition- and processing-related factors that may affect apparent vascular density and complexity.

In contrast, circumscribed choroidal hemangiomas may exhibit a more continuous vascular transition with the surrounding choroid. Characteristic intrinsic vascular patterns have been described on indocyanine green angiography and OCT-A, including “bag of worms” and “spaghetti-like” appearances [36].

Structural OCT and OCT-A therefore provide complementary but fundamentally different information: structural OCT primarily depicts tissue reflectivity and anatomical organization, whereas OCT-A detects flow-related signals within defined segmentation slabs. Apparent spatial concordance or discordance between structural and OCT-A abnormalities may provide clues to tumor–tissue interactions, but their biological interpretation remains limited by the cross-sectional nature of most available studies and by modality-specific imaging artifacts. Longitudinal studies integrating structural OCT, OCT-A, and other imaging modalities will be required to determine whether these observations reflect stable lesion-associated features or dynamic changes during tumor evolution.

3.3. Emerging Role of Retro Mode Imaging

Retro mode imaging is an SLO-based technique available on selected multimodal imaging platforms [23,24]. It uses infrared illumination combined with non-confocal detection to enhance contrast from indirectly scattered light. Laterally displaced apertures (DL or DR modes) produce directional shadowing and a pseudo-relief effect that emphasizes contour and elevation, whereas the annular Ring Aperture (RA) mode provides more symmetric contrast without the directional pseudo-relief characteristic of DL/DR imaging [24]. These images do not represent true topographic reconstructions.

In ocular oncology, tumor-specific evidence remains limited. Retro mode imaging has been investigated primarily in choroidal nevi, in which it may improve qualitative visualization of lesion borders and surface topography [23]. Individual reports have also illustrated its use in retinal cavernous hemangioma [43], choroidal osteoma [44], and sclerochoroidal calcification [45]. The available evidence is therefore preliminary, largely qualitative and platform-specific, and does not establish diagnostic sensitivity, specificity, or incremental diagnostic value over conventional multimodal imaging.

Compared with conventional near-infrared reflectance, Retro mode primarily modifies the geometry of light collection to enhance contour-related contrast rather than providing a distinct tissue-specific signal. Multicolor imaging combines reflectance information obtained at different wavelengths, whereas ultra-widefield photography provides substantially greater spatial coverage, particularly for peripheral lesions. Retro mode should therefore be regarded as a complementary visualization technique rather than an established diagnostic modality for choroidal tumors.

4. Choroidal Melanoma

4.1. Clinical and Multimodal Characteristics

Choroidal melanoma is the most common primary intraocular malignant tumor in adults and typically presents as a pigmented dome-shaped or collar-button choroidal lesion associated with secondary retinal pigment epithelium (RPE) alterations, subretinal fluid, lipofuscin deposition, and secondary retinal changes (Figure 9, Figure 10 and Figure 11) [2,8]. Multimodal imaging plays a central role in lesion characterization by combining structural, vascular, optical, and acoustic information [13].

Color fundus photography remains essential for documenting tumor pigmentation, lesion margins, clinically visible orange pigment, and associated exudative changes. Fundus autofluorescence is particularly useful for demonstrating fluorophore-related signal alterations associated with orange pigment, which frequently appears hyperautofluorescent (Figure 11) [40]. The presence of orange pigment is an established risk feature associated with subsequent growth of melanocytic choroidal lesions and should be interpreted together with other clinical and multimodal risk factors [6].

Fluorescein angiography demonstrates progressive tumor staining and may reveal pinpoint hyperfluorescent spots corresponding to focal RPE decompensation overlying the lesion (Figure 9) [17]. In some melanomas, angiography also demonstrates heterogeneous fluorescence patterns resulting from combined masking effects, intrinsic vascularization, and secondary leakage phenomena. However, fluorescein angiography primarily reflects secondary retinal and RPE alterations rather than the intrinsic tumor architecture itself.

Structural OCT provides high-resolution visualization of the retina and outer retinal layers overlying the melanoma. Typical findings include dome-shaped choroidal elevation, subretinal fluid, outer retinal disruption, shaggy photoreceptors, and focal RPE irregularities (Figure 9 and Figure 11) [8,13]. OCT is particularly sensitive for detecting subtle exudative changes that may not yet be clinically apparent. Enhanced-depth imaging and swept-source OCT further improve visualization of the anterior tumor contour and associated retinal alterations, although posterior tumor margins often remain incompletely visualized in thicker lesions because of signal attenuation [13].

In the melanoma illustrated in Figure 9, Retro mode imaging complements conventional multimodal imaging by emphasizing surface topography and pigment-related changes through indirect retroillumination. These observations are consistent with the topographic contrast generated by Retro mode imaging [24], although its specific diagnostic value in choroidal melanoma remains insufficiently documented.

These findings further illustrate how different optical imaging modalities emphasize distinct structural and optical characteristics of the same lesion.

Overall, multimodal imaging demonstrates that choroidal melanoma may be associated with changes involving the retina, RPE, subretinal space, and adjacent choroidal structures. The combination of pigmentary changes, subretinal fluid, autofluorescence abnormalities, angiographic leakage, and OCT alterations contributes to diagnostic assessment and risk stratification [6,13].

4.2. Ultrasonographic Characteristics

Despite major advances in retinal imaging, ultrasonography remains a cornerstone in the diagnosis and characterization of choroidal melanoma (Figure 10) [17,18]. Unlike optical imaging modalities, ultrasonography provides information regarding lesion dimensions and intrinsic acoustic characteristics, including internal reflectivity, acoustic attenuation, and internal echo patterns [15,16].

On B-scan ultrasonography, choroidal melanomas classically appear as dome-shaped or mushroom-shaped lesions with relatively low internal echogenicity and an acoustically hollow appearance [17,30]. Associated findings may include choroidal excavation, extrascleral extension, retinal detachment, or orbital extension in advanced lesions. Standardized A-scan ultrasonography characteristically demonstrates low-to-medium internal reflectivity, with a progressive decrease in echo amplitude from the anterior portion of the lesion toward its base and a distinct posterior scleral spike [18].

These acoustic features are related to the internal organization of the tumor but should not be regarded as direct surrogates for histologic architecture. Compared with highly reflective vascular lesions such as circumscribed choroidal hemangiomas, many melanomas generate relatively limited internal acoustic backscatter. On standardized A-scan, the progressive decline in echo amplitude toward the tumor base contributes to the characteristic large κ angle (Figure 4, Figure 5 and Figure 10) [30].

Ultrasonography also remains essential for tumor thickness assessment, which is critical for diagnosis, treatment planning, therapeutic follow-up, and prognostic evaluation [17]. However, measurement reproducibility depends on appropriate probe orientation and consistent identification of the anterior and posterior tumor boundaries. Potential sources of variability include inadvertent inclusion of the overlying retina or underlying sclera, associated subretinal fluid, posterior acoustic shadowing, oblique insonation, and interobserver differences in caliper placement [39]. These limitations are particularly relevant for small lesions and during longitudinal follow-up, where small differences in boundary selection or scan orientation may simulate or obscure true tumor growth. Ultrasonography can assess the full lesion thickness and its extent within the acquired scanning planes, including posterior tumor contours that may remain inaccessible to OCT, and retains a major advantage when optical imaging is limited by media opacity, dense pigmentation, extensive subretinal fluid, or OCT signal attenuation [19].

4.3. Longitudinal Evolution and Risk Assessment

Longitudinal multimodal imaging plays a major role in distinguishing benign choroidal nevi from small melanomas and in identifying lesions at risk of malignant transformation [5,6]. While isolated imaging findings may remain nonspecific, the new appearance or increase in subretinal fluid, orange pigment, tumor growth, increasing thickness, or evolving vascular abnormalities may raise concern for malignant evolution [6].

Several clinical and imaging features have been associated with increased melanoma risk, including lesion thickness, orange pigment, subretinal fluid, symptoms, proximity to the optic disc, ultrasonographic hollowness, and documented growth over time [2,6]. OCT has considerably improved the detection of shallow subretinal fluid and subtle outer retinal alterations, whereas fundus autofluorescence sensitively detects fluorophore-related signal abnormalities associated with clinically visible orange pigment [8,40].

Longitudinal multimodal imaging may document evolving secondary features such as changes in subretinal fluid, autofluorescence patterns, angiographic leakage, or OCT abnormalities. Such changes should be interpreted together with documented lesion growth and other established risk features rather than considered independent evidence of malignant progression [13]. Because no single imaging feature is fully specific for malignant transformation, the diagnosis of small choroidal melanoma relies on the convergence of clinical examination, multimodal imaging findings, ultrasonographic features, and temporal evolution [2,5].

5. Circumscribed Choroidal Hemangioma

5.1. Clinical Appearance and Retinal Consequences

Circumscribed choroidal hemangioma is a benign vascular tumor typically presenting as an orange-red elevated choroidal lesion with indistinct margins located in the posterior pole (Figure 12) [46,47]. Although histologically benign, the lesion may induce significant visual impairment because of chronic exudation, subretinal fluid (SRF) accumulation, and secondary retinal degeneration [47].

Color fundus photography usually demonstrates a relatively homogeneous orange-red mass whose coloration reflects the vascular nature of the lesion. Chronic exudation may be associated with shallow exudative retinal detachment, pigmentary changes, and progressive alterations of the outer retina and retinal pigment epithelium (RPE). Visual symptoms are therefore often related less to the intrinsic tumor itself than to the secondary retinal consequences of persistent SRF [46].

Fundus autofluorescence frequently reveals irregular hyperautofluorescent changes overlying the lesion and areas of chronic SRF (Figure 12). These abnormalities should not be attributed solely to RPE dysfunction. Persistent separation of the outer retina from the RPE may impair normal phagocytosis of shed photoreceptor outer segments, resulting in accumulation of autofluorescent material within the outer retina and subretinal space; macrophages containing phagocytosed outer-segment-derived material may also contribute to focal hyperautofluorescent deposits. With increasing chronicity, progressive photoreceptor and RPE atrophy may subsequently result in reduced autofluorescence [48].

Structural OCT demonstrates a smooth dome-shaped choroidal elevation associated with SRF, elongation of photoreceptor outer segments, outer retinal disruption, and variable RPE irregularities [36]. In chronic cases, retinal thinning and persistent photoreceptor damage may remain visible even after successful tumor treatment and may account for irreversible visual loss [36,47]. Multimodal imaging therefore contributes not only to tumor characterization but also to assessment of secondary retinal injury and visual prognosis.

5.2. Angiographic Dynamics

Angiographic behavior represents one of the most characteristic imaging features of circumscribed choroidal hemangioma [36,46]. The observed patterns reflect both the intrinsic vascular architecture of the tumor and the different intravascular and extravascular behavior of fluorescein and indocyanine green.

On fluorescein angiography, large intrinsic tumor vessels may become visible during the early phases, followed by progressive diffuse hyperfluorescence as fluorescein accumulates within the extravascular tissue of the tumor (Figure 12) [49]. This tumor staining should be distinguished from secondary leakage into the retina or subretinal space, which may become prominent in the late phases. Progressive dye diffusion and associated exudative changes may subsequently obscure the intrinsic vascular architecture [46,49].

ICGA provides more direct visualization of the intrinsic tumor vasculature because ICG is largely protein-bound and therefore leaks more slowly from choroidal vessels than fluorescein [50]. Circumscribed choroidal hemangiomas typically demonstrate early filling of intrinsic vascular channels, sometimes producing characteristic “bag of worms” or “spaghetti-like” vascular patterns, followed by increasing tumor fluorescence during the first minutes [36,49]. In the late phase, tumor fluorescence progressively decreases relative to the surrounding choroid, producing the characteristic “washout” appearance (Figure 12) [49].

These angiographic dynamics illustrate the vascular nature of circumscribed choroidal hemangioma and the complementary information provided by the two dyes. Fluorescein angiography emphasizes progressive extravascular staining and secondary leakage, whereas ICGA better delineates intrinsic vascular channels and their temporal filling and clearance. The abundance of vascular interfaces within the lesion may also contribute to the characteristically high internal reflectivity observed on ultrasonography [30].

5.3. OCT and OCT-A Findings

OCT has substantially improved the evaluation of circumscribed choroidal hemangioma by allowing precise analysis of the associated retinal and choroidal changes [36,47]. Structural OCT typically demonstrates a smooth choroidal elevation with gradual contour transition toward the surrounding choroid, in contrast to the more abrupt profile often observed in melanomas (Figure 13) [36].

Subretinal fluid is common and may extend well beyond the clinically visible tumor margins. Chronic SRF frequently produces secondary retinal abnormalities, including elongation of photoreceptor outer segments, cystoid retinal degeneration, ellipsoid zone disruption, and diffuse RPE alterations [47]. These retinal consequences often correlate more closely with visual function than tumor thickness itself.

Enhanced-depth imaging OCT and swept-source OCT improve visualization of the lesion contour and choroidal architecture [36]. OCT-A may additionally demonstrate alterations of the intrinsic vascular network and spatial relationships between the tumor vasculature and the surrounding choroidal circulation [36]. Compared with melanomas, circumscribed choroidal hemangiomas may show a more continuous vascular transition with the adjacent choroid.

5.4. Ultrasonographic Characteristics

Ultrasonography remains highly valuable for the diagnosis of circumscribed choroidal hemangioma, particularly when differentiating the lesion from amelanotic melanoma, metastasis, or posterior scleritis [17,30].

On B-scan ultrasonography, circumscribed choroidal hemangioma usually appears as a smooth dome-shaped choroidal mass with high internal echogenicity and relatively homogeneous internal echoes (Figure 13) [30].

On standardized A-scan ultrasonography, the lesion characteristically demonstrates high-amplitude internal echoes that remain relatively preserved toward the posterior tumor boundary. In contrast to choroidal melanoma, the κ angle is typically small, reflecting the relatively limited decline in echo amplitude through the lesion (Figure 4 and Figure 5) [18,30]. The numerous vascular and stromal interfaces within circumscribed choroidal hemangiomas are considered to contribute to this abundant internal acoustic backscatter.

These ultrasonographic characteristics provide an acoustic signature related to the internal organization of the lesion rather than a direct representation of its histologic microarchitecture. The numerous vascular and stromal interfaces within the tumor are considered to contribute to its strong internal backscatter and high internal reflectivity [30]. Acoustic attenuation during propagation represents a distinct phenomenon and should not be inferred solely from the amplitude of the internal echoes [15].

Ultrasonography also provides assessment of tumor thickness and remains valuable for therapeutic follow-up after photodynamic therapy, laser treatment, or radiotherapy [46]. Reduction in subretinal fluid may occur rapidly after treatment, whereas regression of tumor thickness is often slower and incomplete [47].

6. Choroidal Metastases and Other Differential Diagnoses

6.1. Choroidal Metastases

Choroidal metastases represent the most common intraocular malignant tumors overall and typically originate from breast carcinoma in women and lung carcinoma in men [51,52]. Because of the rich vascular supply of the posterior choroid, metastatic lesions show a marked predilection for the posterior pole, where they may rapidly produce visual symptoms through exudative retinal detachment and macular involvement (Figure 14) [53].

Clinically, choroidal metastases usually appear as creamy-yellow, ill-defined choroidal lesions with relatively shallow elevation [52]. In contrast to melanomas, pigmentation is generally limited or absent. Multifocality and bilaterality may also suggest metastatic disease [54].

Fluorescein angiography often demonstrates heterogeneous and irregular tumor staining with progressive leakage during the late phases [51]. Extensive dye diffusion into the subretinal space contributes to blurring of lesion margins and reflects the marked permeability of the abnormal tumor vasculature (Figure 14). The associated exudative retinal detachment may become particularly conspicuous during late angiographic phases.

Structural OCT typically demonstrates an irregular (“lumpy-bumpy”) anterior choroidal contour associated with underlying choroidal thickening, prominent subretinal fluid, diffuse outer retinal dysfunction, and retinal pigment epithelium alterations. The characteristic “lumpy-bumpy” configuration, although not entirely specific, has been reported more frequently in choroidal metastases and choroidal lymphomas than in choroidal melanomas and may provide an additional clue in the multimodal differential diagnosis [51,55].

Ultrasonographically, choroidal metastases generally demonstrate moderate-to-high internal reflectivity with greater heterogeneity than melanomas [30,51]. Acoustic attenuation is usually intermediate because of the heterogeneous histologic organization and variable stromal composition of metastatic tumors (Figure 4 and Figure 5). Unlike the homogeneous acoustic profile of melanomas, metastatic lesions often produce irregular internal echo patterns reflecting necrosis, glandular organization, fibrosis, or variable vascularity [17].

Overall, the multimodal imaging appearance of choroidal metastases reflects their infiltrative vascular behavior, marked exudative tendency, and heterogeneous histologic architecture [51].

6.2. Choroidal Nevus

Choroidal nevus is the most common benign melanocytic lesion of the choroid and represents an important diagnostic challenge because some nevi may evolve into melanoma over time [5,6]. Most choroidal nevi remain stable throughout life and demonstrate characteristic signs of chronicity, including overlying drusen, retinal pigment epithelium alterations, and localized chorioretinal changes [6].

Fundus photography often demonstrates a flat or minimally elevated pigmented lesion with relatively stable margins. The presence of drusen generally suggests chronicity and longstanding lesion stability, whereas orange pigment, subretinal fluid, or documented growth raise concern for malignant transformation [6,40].

Structural OCT has substantially improved the evaluation of choroidal nevi by allowing detailed analysis of the overlying retinal and RPE changes [8,13]. Typical findings include shallow choroidal thickening, drusen-related RPE irregularities, localized outer retinal alterations, and focal hyperreflective changes anterior to the lesion (Figure 6, Figure 7 and Figure 8).

In some cases, hyperreflective OCT alterations may extend beyond the clinically visible lesion margins (Figure 7 and Figure 8). The biological basis of these changes remains uncertain, and their possible relationship to alterations in tissue organization or tumor–tissue interactions should be regarded as hypothetical.

Infrared and Retro mode imaging may further enhance visualization of lesion borders, surface topography, and subtle contour abnormalities associated with choroidal nevi (Figure 6 and Figure 7) [23,24].

In the case illustrated in Figure 6, Ring Aperture Retro mode imaging emphasizes focal pigmentary and RPE-associated changes overlying the lesion.

Three-dimensional structural OCT studies have demonstrated complex spatial relationships between small melanocytic choroidal tumors and the surrounding choroidal vasculature, including persistent intratumoral vascular channels and displacement, encasement, or focal obliteration of larger choroidal vessels [41,42]. These observations provide morphological evidence of interactions between melanocytic lesions and adjacent choroidal structures but should not be interpreted as direct evidence of altered blood flow or choriocapillaris remodeling.

Importantly, the diagnosis of choroidal nevus remains largely based on long-term stability. Multimodal imaging therefore plays a central role not only in lesion characterization but also in longitudinal surveillance and early detection of malignant transformation [5,6].

6.3. Choroidal and Pseudotumoral Mimickers

Several infiltrative, inflammatory, calcific, and hemorrhagic conditions may mimic choroidal tumors clinically and radiologically. Multimodal imaging is therefore essential for distinguishing neoplastic lesions from non-neoplastic pseudotumoral processes and for recognizing lesions whose apparent choroidal origin may be misleading [8,31].

Choroidal lymphoma is a rare infiltrative malignancy that may mimic inflammatory choroidal disease, metastasis, or amelanotic melanoma. Choroid-focused OCT typically demonstrates diffuse or placoid choroidal thickening with relatively homogeneous hyporeflective infiltration, loss of the normal vascular architecture, and variable secondary RPE alterations and subretinal fluid [56]. The choroidal surface may acquire a characteristic undulating configuration, previously described as “placid”, “rippled”, or “seasick” according to lesion thickness [8]. ICGA frequently demonstrates multifocal hypofluorescent areas extending beyond clinically visible involvement, while ultrasonography can demonstrate choroidal thickening and may help identify extrascleral extension [56]. This predominantly choroidal pattern should be distinguished from primary vitreoretinal lymphoma, in which OCT more typically demonstrates vitreous cells, RPE nodularity, outer retinal hyperreflectivity, and sub-RPE or subretinal infiltrates [57].

Calcific lesions provide another important source of pseudotumoral appearances. Sclerochoroidal calcification is a benign, predominantly scleral process typically encountered in older individuals and may be mistaken for choroidal osteoma, metastasis, nevus, melanoma, or lymphoma [45,58]. EDI-OCT demonstrates the scleral origin of the lesion, compression and thinning of the overlying choroid, and characteristic “mountain-like” configurations ranging from flat or rolling to rocky-rolling and table-mountain patterns [58]. Ultrasonography confirms the highly reflective calcified nature of the lesion, whereas Retro mode imaging can enhance visualization of its surface elevation and topography [45].

Choroidal osteoma, in contrast, is a benign choroidal tumor composed of mature bone, typically occurring in younger patients. Multimodal imaging demonstrates a calcified choroidal lesion with high acoustic reflectivity on ultrasonography and characteristic hyperreflective lamellar or spongiform intralesional structures on OCT, corresponding to its cancellous bone architecture [59,60]. Progressive decalcification, outer retinal and RPE alterations, subretinal fluid, and secondary choroidal neovascularization may substantially modify its appearance over time [60]. Thus, establishing whether a calcified-appearing lesion originates primarily from the sclera or from the choroid is an important contribution of multimodal imaging.

Nodular posterior scleritis deserves separate consideration from diffuse posterior scleritis because it can closely simulate an amelanotic choroidal mass. It may present as a pale or yellow-orange elevated lesion associated with chorioretinal folds and exudative retinal detachment [61,62]. On B-scan ultrasonography, the lesion is generally highly echogenic and may be associated with posterior scleral thickening and fluid in the sub-Tenon space, producing the classical T-sign; absence of choroidal excavation provides an additional clue against melanoma [61,62]. Clinical inflammatory signs may nevertheless be subtle or absent. Rapid reduction in lesion thickness and associated subretinal fluid following anti-inflammatory treatment provides a further important longitudinal diagnostic clue [61,62].

Granulomatous choroidal lesions, particularly those associated with tuberculosis or sarcoidosis, may similarly present as elevated amelanotic masses and mimic melanoma or metastasis. Their OCT, angiographic, and ultrasonographic appearances can overlap with those of neoplastic lesions, making systemic context, inflammatory findings, and longitudinal evolution important components of the diagnostic process [63,64].

Finally, peripheral exudative hemorrhagic chorioretinopathy (PEHCR) is an important hemorrhagic pseudomelanoma, particularly in elderly patients. Peripheral subretinal or sub-RPE hemorrhage, exudation, and hemorrhagic mass-like elevation may closely simulate choroidal melanoma [65]. Wide-field examination and angiography help demonstrate the peripheral and frequently extensive nature of the process, while ultrasonography is particularly useful when hemorrhage limits optical visualization. The absence of choroidal excavation on B-scan, together with the characteristic peripheral hemorrhagic and exudative distribution, may provide useful clues in the differential diagnosis with melanoma [65].

These differential diagnoses illustrate the importance of integrating multimodal imaging findings rather than relying on a single modality. In particular, discrepancies between the apparent location or morphology of a lesion on fundus examination and its structural origin on OCT or ultrasonography may themselves provide important diagnostic information. Ultrasonography remains especially valuable when optical imaging cannot reliably determine whether an apparent choroidal mass represents a solid neoplasm, diffuse infiltration, scleral inflammation, calcification, or hemorrhagic material.

The principal multimodal imaging characteristics and differential clues of representative choroidal tumors and pseudotumoral lesions are summarized in

Table 2. Principal multimodal imaging characteristics and differential diagnostic clues of representative choroidal tumors and pseudotumoral lesions. The table summarizes the complementary contributions of ultrasonography, optical coherence tomography, fundus autofluorescence, and angiographic imaging to lesion characterization and differential diagnosis. Imaging findings should be interpreted in combination with clinical presentation and longitudinal evolution, as individual features are not necessarily specific to a single lesion type. Abbreviations: CNV, choroidal neovascularization; FA, fluorescein angiography; FAF, fundus autofluorescence; ICGA, indocyanine green angiography; OCT, optical coherence tomography; PEHCR, peripheral exudative hemorrhagic chorioretinopathy; PVRL, primary vitreoretinal lymphoma; RPE, retinal pigment epithelium; SRF, subretinal fluid; κ, kappa angle.

Lesion	Ultrasonography (B-Scan/A-Scan)	OCT	FAF/Angiography	Key Differential Clues and Pitfalls	Principal Contribution of Multimodal Imaging
Choroidal nevus	Flat/slightly elevated; small lesions often poorly characterized; no typical melanoma acoustic pattern	Shallow choroidal thickening; drusen/RPE and outer retinal changes; ±SRF	FAF: RPE changes, drusen, lipofuscin; angiography usually nonspecific	Stability, drusen, chronic RPE changes favor nevus; growth, orange pigment, SRF, increasing thickness raise suspicion	Longitudinal surveillance for malignant transformation
Choroidal melanoma	Dome- or mushroom-shaped; acoustic hollowness; low-to-medium internal reflectivity; progressive decline in A-scan echo amplitude with large κ angle; thickness measurement	Choroidal elevation; RPE/outer retinal disruption; SRF; lipofuscin-associated changes	FAF: increased autofluorescence frequently associated with clinically visible orange pigment; FA/ICGA: intrinsic vascularity, leakage and variable masking	Growth, orange pigment, SRF and increasing thickness favor melanoma; amelanotic lesions may mimic metastasis, hemangioma or inflammatory lesions	Intrinsic acoustic characterization and thickness assessment, integrated with optical markers of tumor activity
Circumscribed choroidal hemangioma	Smooth dome-shaped mass; high, relatively homogeneous internal echoes; high A-scan amplitudes with limited decline and small κ angle	Smooth choroidal elevation; compression of choroidal structures; ±SRF and secondary outer retinal/RPE changes	FAF: variable with chronic RPE changes/SRF; FA: progressive staining; ICGA: early filling with characteristic late washout	Orange-red appearance and high internal reflectivity favor hemangioma; may mimic amelanotic melanoma or metastasis	Acoustic differentiation from other amelanotic masses, combined with vascular characterization and assessment of retinal consequences
Choroidal metastasis	Usually shallow/dome-shaped; moderate-to-high, often heterogeneous internal reflectivity; variable A-scan pattern	Irregular “lumpy-bumpy” choroidal surface; prominent SRF; RPE and outer retinal alterations	FAF: variable RPE-related changes; FA: heterogeneous staining with late leakage; ICGA often hypofluorescent	Creamy-yellow appearance; multifocality/bilaterality; lumpy-bumpy OCT surface; may mimic amelanotic melanoma or lymphoma	Recognition of an infiltrative/exudative pattern and differentiation from melanoma, integrating morphology, vascular behavior, and acoustic heterogeneity
Choroidal lymphoma	Diffuse or nodular choroidal thickening; variable internal reflectivity; may demonstrate extrascleral extension	Diffuse/placoid hyporeflective choroidal infiltration; loss of normal vascular architecture; “placid”, “rippled” or “seasick” surface; ±SRF/RPE changes	FAF: variable RPE alterations; ICGA: multifocal hypofluorescent areas, often beyond clinically visible involvement	Creamy-yellow, multifocal/bilateral infiltration; distinguish from metastasis, inflammatory disease and PVRL	Define the extent and predominantly choroidal pattern of infiltration, integrating OCT/ICGA with ultrasonographic assessment
Sclerochoroidal calcification	Highly reflective calcified lesion with marked posterior acoustic shadowing	Scleral origin; compression/thinning of overlying choroid; “mountain-like” configurations	FAF: variable; angiography usually nonspecific; Retro mode enhances surface elevation/topography	Typically yellow-white and superotemporal; may mimic osteoma or amelanotic tumor; scleral rather than choroidal origin is a key clue	Establish calcific nature and scleral origin, integrating ultrasound with OCT topography
Choroidal osteoma	Highly reflective calcified lesion with posterior acoustic shadowing	Choroidal origin; hyperreflective lamellar/spongiform intralesional structures; outer retinal/RPE changes with decalcification; ±SRF/CNV	FAF varies with calcification/RPE status; FA/ICGA may delineate vascular and decalcified areas	Typically young patient; juxtapapillary/macular location; distinguish from sclerochoroidal calcification and other amelanotic masses	Confirm calcific nature and define choroidal architecture, while OCT monitors decalcification and secondary retinal/CNV changes
Nodular posterior scleritis	Highly echogenic mass; posterior scleral thickening; ±sub-Tenon fluid/T-sign; absence of choroidal excavation	Focal sclerochoroidal elevation; choroidal/retinal folds; ±prominent SRF	FAF: variable; FA: inflammatory leakage; ICGA may show hypofluorescence	May closely mimic amelanotic melanoma; pain/inflammation may be absent; rapid response to anti-inflammatory therapy is a useful clue	Identify scleral/inflammatory origin and distinguish a pseudotumoral process from a solid choroidal mass
Granulomatous choroidal lesions	Focal choroidal mass/thickening; variable internal reflectivity; A-scan findings nonspecific	Elevated choroidal lesion; variable internal reflectivity; compression of choriocapillaris; ±SRF and outer retinal/RPE changes	FAF: variable; FA/ICGA: inflammatory staining/leakage and hypofluorescent areas	Tuberculosis/sarcoidosis may mimic amelanotic melanoma or metastasis; systemic/inflammatory context and evolution are critical	Differentiate inflammatory infiltration from neoplasia through multimodal and longitudinal convergence
PEHCR	Peripheral mass-like elevation; variable internal reflectivity; absence of choroidal excavation; no specific A-scan pattern	Peripheral subretinal/sub-RPE hemorrhagic elevation; ±SRF/exudation; OCT often limited by peripheral location/hemorrhage	FAF: variable; FA/ICGA may show blockage by hemorrhage and peripheral vascular abnormalities	Major pseudomelanoma in elderly patients; hemorrhage may conceal underlying structures; characteristic peripheral hemorrhagic/exudative distribution	Exclude an underlying solid choroidal mass when optical visualization is limited by hemorrhage

.

7. Longitudinal Multimodal Imaging

7.1. Temporal Evolution of Choroidal Lesions

Longitudinal imaging is an important component of choroidal tumor assessment because stability, growth, and changes in associated retinal or RPE findings may provide information that cannot be obtained from a single examination [5,6]. Serial imaging may document changes in lesion dimensions, subretinal fluid, orange pigment, outer retinal integrity, or angiographic features, which should be interpreted together with established clinical and multimodal risk factors rather than as isolated indicators of malignant progression [6,13].

Structural OCT is particularly useful for monitoring shallow subretinal fluid and retinal or RPE alterations, while fundus autofluorescence may document changes in lipofuscin-related signals and angiography may show evolving leakage or vascular filling patterns [8,40]. Longitudinal changes may therefore concern both the lesion itself and its effects on surrounding retinal and choroidal tissues.

This distinction is also relevant for functional assessment. In circumscribed choroidal hemangioma, visual outcome may depend more on chronic retinal damage than on tumor dimensions alone, whereas in metastatic lesions the evolution of subretinal fluid and foveal involvement may substantially influence visual function [36,47,51].

Longitudinal imaging is also important after treatment. Regression patterns following radiotherapy, proton beam therapy, photodynamic therapy, or systemic oncologic treatment vary according to tumor type [17,46]. Tumor thickness may decrease slowly or incompletely, whereas retinal exudation and subretinal fluid may regress earlier. Serial B-scan ultrasonography is therefore particularly useful for documenting changes in tumor thickness during observation and after treatment, provided that acquisition planes and measurement boundaries are reproduced as consistently as possible. Multimodal follow-up complements these dimensional measurements by documenting both tumor response and secondary retinal changes [13,31].

7.2. Complementarity of Imaging Modalities

Multimodal imaging of choroidal tumors combines techniques that interrogate different properties of the lesion and surrounding tissues [3,13]. Optical modalities provide complementary information based on reflection, scattering, absorption, fluorescence, or flow-related signals [11]. Structural OCT may directly delineate the anterior contour and accessible internal structure of a choroidal lesion while providing high-resolution assessment of associated retinal, RPE, and subretinal changes. Fundus autofluorescence depicts the distribution of endogenous fluorophores and associated RPE alterations, whereas angiographic techniques characterize vascular filling, permeability, and flow-related features [8,51].

Ultrasonography provides a different set of information based on acoustic impedance, backscatter, and sound-wave attenuation [15,16]. Lesion thickness, internal reflectivity, echo patterns, and acoustic attenuation contribute to characterization of lesion dimensions and acoustic properties, including when optical visualization is incomplete [17,18,30].

Multimodal interpretation therefore relies on the relationship between these different signals rather than on the assumption that one modality provides a complete representation of the lesion. Concordant or discordant findings may help refine the differential diagnosis, provided that each finding is interpreted according to the physical basis and limitations of the corresponding technique [11,15].

7.3. Limits and Future Directions

Despite considerable progress in retinal imaging, important limitations remain in the multimodal evaluation of choroidal tumors [13,51]. Optical imaging techniques are inherently constrained by light penetration, media opacity, pigmentation, and signal attenuation within elevated lesions [11]. Thick melanomas, dense pigmentation, hemorrhage, or extensive exudative retinal detachment may substantially limit OCT visualization of deeper tumor structures [8]. Similarly, OCT angiography remains technically challenging in highly elevated lesions because of limited penetration into deeper choroidal tissues [38].

Interpretation of multimodal imaging findings also remains partly qualitative and observer-dependent. Many OCT-A observations concerning tumor-associated differences in vascular density, organization, or flow-related signals remain exploratory and require further validation [38].

Similarly, Retro mode imaging remains relatively underexplored in ocular oncology, with available evidence being largely qualitative and without standardized interpretation criteria or validated incremental diagnostic value [23].

Future developments will likely involve increasingly quantitative multimodal approaches integrating structural OCT, OCT-A, ultrasonography, autofluorescence, and artificial intelligence-assisted image analysis [4,13]. Automated segmentation, vascular quantification, texture analysis, radiomics, and multimodal data integration may improve reproducibility and assist lesion characterization, longitudinal assessment, and treatment-response evaluation.

7.4. Clinical Implications of Multimodal Imaging

Beyond lesion characterization and differential diagnosis, multimodal imaging contributes to several clinically relevant decision points by determining whether additional surveillance, systemic investigation, cross-sectional imaging, or tissue confirmation should be considered. Importantly, multimodal imaging can substantially narrow the differential diagnosis and increase diagnostic confidence but cannot always establish tissue diagnosis. Selected clinical scenarios illustrating how imaging findings may guide further diagnostic evaluation are summarized in

Table 3. Contribution of multimodal imaging to selected clinical decision points in the evaluation of choroidal lesions. The table summarizes how complementary imaging findings may guide diagnostic assessment and indicate when additional investigations should be considered. It is intended as an imaging-oriented framework rather than a therapeutic management algorithm. Multimodal imaging can substantially narrow the differential diagnosis and increase diagnostic confidence but cannot always establish tissue diagnosis. Individual imaging findings should therefore be interpreted in conjunction with clinical history, longitudinal evolution, and, when appropriate, systemic evaluation, additional cross-sectional imaging, or histopathologic confirmation. FA: fluorescein angiography; FAF: fundus autofluorescence; ICGA: indocyanine green angiography; MRI: magnetic resonance imaging; OCT: optical coherence tomography; PEHCR: peripheral exudative hemorrhagic chorioretinopathy; SRF: subretinal fluid; TFSOM-DIM: To Find Small Ocular Melanoma Doing Imaging; UBM: ultrasound biomicroscopy; US: ultrasonography.

Clinical Scenario	Key Multimodal Imaging Findings	Specific Contribution of Ultrasonography	When Additional Evaluation May Be Required	Clinical Implication/Next Diagnostic Step
Small melanocytic lesion: nevus vs. melanoma	Documented growth and multimodal risk features increase concern for melanoma. TFSOM-DIM identifies factors associated with subsequent growth: thickness > 2 mm (US), subretinal fluid (OCT), symptomatic visual loss, orange pigment (FAF), acoustic hollowness (US), and diameter > 5 mm (fundus photography). These are risk factors for growth/transformation and should not be regarded individually as diagnostic criteria for melanoma [3,6].	Measures lesion thickness and basal extent within acquired planes and assesses internal acoustic characteristics. Serial examinations may document growth, but small differences require consistent probe orientation and boundary selection.	Indeterminate lesions may require closer multimodal surveillance or referral to an ocular oncology center. Biopsy may be considered when the clinical diagnosis remains uncertain and histopathologic confirmation is expected to influence management [66,67].	Integrate longitudinal behavior with multimodal risk features rather than relying on any single imaging sign. Increasing risk features or documented growth increases the need for specialist evaluation.
Amelanotic melanoma vs. choroidal metastasis or inflammatory granuloma	Color, OCT, FAF, FA/ICGA, and clinical context may narrow the differential diagnosis. Metastases often show an irregular or “lumpy-bumpy” anterior contour and prominent SRF; granulomas may show inflammatory-associated features; amelanotic melanoma may retain melanoma-associated acoustic and structural characteristics. No individual imaging sign is fully specific [51,63,64].	Confirms the presence and dimensions of a solid lesion and provides internal reflectivity and echo-pattern information that may help distinguish competing diagnoses [17].	Systemic oncologic evaluation is required when metastasis is suspected, particularly without a known primary malignancy. Infectious/inflammatory investigation is appropriate when granuloma is suspected. Biopsy may remain necessary when imaging and systemic evaluation do not establish a sufficiently confident diagnosis [67].	Multimodal imaging substantially narrows the differential diagnosis but does not always establish tissue diagnosis; clinical history and targeted systemic evaluation remain essential.
Circumscribed choroidal hemangioma vs. amelanotic melanoma	Hemangioma typically demonstrates a smooth dome-shaped lesion, characteristic vascular behavior on ICGA, and relatively homogeneous structural features. OCT defines associated SRF and retinal effects. Multimodal concordance is more informative than any isolated sign [36].	Hemangioma generally shows high internal reflectivity and relatively homogeneous internal echoes, contrasting with the lower reflectivity and acoustic hollowness often associated with melanoma; overlap remains possible [17,30].	Additional angiographic characterization, particularly ICGA, may be useful when clinical and ultrasonographic findings are equivocal [17].	Concordant optical, angiographic, and acoustic findings can substantially increase diagnostic confidence and reduce the risk of misclassifying a benign vascular tumor as an amelanotic melanoma.
Hemorrhagic pseudotumor/PEHCR vs. choroidal melanoma	Peripheral hemorrhage, exudation, pigment epithelial detachment, and subretinal or sub-RPE blood may obscure the underlying anatomy and mimic a solid choroidal mass. Evolution on serial imaging may be diagnostically informative [65].	Particularly important when hemorrhage limits optical visualization; helps determine whether a persistent solid mass is present beneath or adjacent to the hemorrhage [65].	Follow-up after partial resolution of hemorrhage may clarify lesion anatomy. Persistent or atypical mass-like findings require further investigation, and biopsy may occasionally be necessary when melanoma cannot otherwise be excluded.	Avoid both premature tumor treatment and premature reassurance: hemorrhagic appearance alone does not establish a benign pseudotumoral diagnosis.
Choroidal lesion obscured by opaque media	Fundus photography, FAF, angiography, and OCT may be unavailable or nondiagnostic because of cataract, vitreous hemorrhage, or other media opacity.	Becomes the principal ocular imaging technique for detecting a mass, defining its location, measuring thickness and extent within acquired planes, and assessing acoustic characteristics [19].	MRI may complement ultrasonography when local extension cannot be adequately characterized or when ciliary body, extrascleral, or extraocular extension is suspected [67].	Ultrasonography may determine whether a true intraocular mass is present and whether further oncologic evaluation is required despite absence of direct optical visualization.
Suspected melanoma requiring assessment of local extent or unresolved tissue diagnosis	Multimodal ocular imaging establishes lesion morphology and associated retinal/choroidal changes but may not fully define extraocular extension or determine histologic identity in atypical lesions.	Tumor dimensions influence staging and treatment planning; ultrasonography/UBM is particularly important for thickness and anterior/ciliary body involvement [67].	Orbital MRI is indicated when ciliary body, extrascleral, or extraocular extension requires further assessment. Biopsy is not mandatory for a clinically typical uveal melanoma but is recommended when the clinical diagnosis remains uncertain; biopsy may also provide prognostic molecular information in selected confirmed melanomas [67].	Recognize the limits of imaging: additional cross-sectional imaging or tissue sampling should be considered when the unanswered diagnostic or staging question could alter clinical management.

.

8. What Ultrasonography Still Uniquely Provides

8.1. Intrinsic Acoustic Tissue Characterization

One of the major strengths of ultrasonography in ocular oncology is its ability to characterize intrinsic acoustic properties of a lesion rather than primarily its secondary retinal consequences [17,18]. Unlike optical imaging modalities, which interrogate light reflection, scattering, absorption, fluorescence, and flow-related signals, ultrasonography provides information derived from the interaction of acoustic waves with the lesion itself, including internal reflectivity, internal echo patterns, and acoustic attenuation [15,16]. These parameters provide an acoustic signature related to the internal organization of the lesion, although they should not be interpreted as a direct representation of its histologic architecture.

Internal reflectivity corresponds to the amplitude and distribution of echoes arising from within the lesion and depends on the number, acoustic contrast, size, and spatial organization of internal scattering interfaces [18]. Choroidal melanomas typically demonstrate low-to-medium internal reflectivity, consistent with relatively limited internal acoustic backscatter [17,30]. In contrast, circumscribed choroidal hemangiomas typically generate abundant internal backscatter and demonstrate high internal reflectivity, a finding considered to be related to their numerous vascular and stromal interfaces [30]. Metastatic lesions more often exhibit intermediate or heterogeneous reflectivity patterns, consistent with their variable internal organization and stromal composition [51].

Acoustic attenuation represents a distinct property and corresponds to the progressive loss of acoustic energy during propagation through tissue as a result of absorption and scattering [15]. It should therefore be distinguished from internal reflectivity and from the descriptive B-scan appearance of acoustic hollowness. In choroidal melanoma, relatively low internal backscatter may produce an acoustically hollow appearance on B-scan, while standardized A-scan may demonstrate a progressive decline in echo amplitude toward the tumor base, resulting in a characteristic large κ angle [18,30]. Circumscribed choroidal hemangiomas, by comparison, typically demonstrate high-amplitude internal echoes that remain relatively preserved throughout the lesion, resulting in a smaller κ angle [30].

Importantly, these acoustic characteristics provide information about lesion organization that cannot be directly inferred from OCT, autofluorescence, angiography, or other optical imaging modalities alone [17]. They therefore complement the structural, vascular, fluorophore-related, and secondary tissue information provided by optical imaging, without constituting a direct histologic characterization [13].

8.2. Independence from Optical Transparency

Another major advantage of ultrasonography is its relative independence from optical transparency [17,19]. Because ultrasound propagation does not depend on visible light transmission through ocular media, lesion visualization remains possible even in the presence of dense cataract, vitreous hemorrhage, corneal opacity, hyphema, or severe retinal detachment [15].

This property is particularly important in ocular oncology because media opacity may obscure clinically significant lesions or prevent adequate OCT visualization [13]. In some cases, ultrasonography may reveal a large choroidal melanoma hidden behind dense vitreous hemorrhage or extensive hemorrhagic retinal detachment [17]. Similarly, peripheral tumors located beyond the field of conventional OCT acquisition may remain readily accessible to ultrasonographic examination [19].

Large tumors also frequently exceed the penetration capabilities of optical imaging techniques. Although enhanced-depth imaging OCT and swept-source OCT have considerably improved choroidal visualization, posterior tumor margins often remain incompletely visualized in thick melanomas because of signal attenuation and limited penetration depth [8,13]. Ultrasonography, by contrast, can assess the full lesion thickness and its extent within the acquired scanning planes, including posterior contours and possible extrascleral extension, largely independently of pigmentation, media opacity, or lesion thickness [17,30].

Thus, while modern retinal imaging technologies provide high-resolution visualization of superficial retinal and choroidal structures, ultrasonography retains clear advantages whenever optical access becomes limited or lesion dimensions exceed OCT capabilities [4,15].

8.3. Persistent Role and Limitations of Ultrasonography

Despite advances in OCT and other optical imaging techniques, ultrasonography continues to provide clinically relevant information that is not directly reproduced by optical modalities. In addition to lesion thickness and extent within the acquired scanning planes, internal reflectivity, echo patterns, and acoustic attenuation may contribute to the differential diagnosis of choroidal lesions, particularly when optical penetration is limited or media opacity prevents direct visualization [17,19,30].

Ultrasonography nevertheless has important limitations. Routine two-dimensional examinations sample selected scanning planes, and thickness measurements depend on probe orientation, boundary identification, and operator technique. Small lesions may approach the limits of ultrasonographic resolution, while subretinal fluid, adjacent retinal or scleral structures, posterior shadowing, and oblique insonation may affect measurement reproducibility [39]. These limitations support interpretation of ultrasonographic findings in conjunction with clinical examination and other imaging modalities, particularly during longitudinal assessment of small lesions.

9. Conclusions

Multimodal imaging provides complementary information for the characterization and differential diagnosis of choroidal tumors because each modality interrogates different tissue properties. Optical techniques can directly characterize accessible portions of choroidal lesions while providing high-resolution assessment of associated retinal, RPE, subretinal, vascular, and fluorophore-related changes. Ultrasonography provides dimensional and acoustic information through lesion thickness, internal reflectivity, echo patterns, and acoustic attenuation, including when optical visualization is limited. These acoustic characteristics may reflect aspects of internal lesion organization but should not be interpreted as direct surrogates for histologic architecture.

The diagnostic contribution of multimodal imaging therefore depends on integrating findings according to the physical basis and limitations of each technique. Structural, vascular, fluorophore-related, topographic, and acoustic findings may narrow the differential diagnosis and increase diagnostic confidence, but imaging cannot always establish tissue diagnosis. Clinical history, longitudinal evolution, systemic evaluation, additional cross-sectional imaging, and, in selected indeterminate cases, histopathologic confirmation remain necessary components of the diagnostic pathway.

Future developments are likely to increase the quantitative integration of OCT-based imaging, ultrasonography, radiomics, and artificial intelligence-assisted analysis. Their clinical value will depend not only on improved image acquisition and analysis but also on determining which imaging features provide reproducible information that meaningfully contributes to diagnosis, longitudinal assessment, and clinical decision-making.