Diagnostics

14 pages, 247 KB

Open AccessReview

The Role of Synthetic Data and Generative AI in Breast Imaging: Promise, Pitfalls, and Pathways Forward

by Filippo Pesapane, Lucrezia D’Amelio, Luca Nicosia, Carmen Mallardi, Anna Bozzini, Lorenza Meneghetti, Gianpaolo Carrafiello, Enrico Cassano and Sonia Santicchia

Diagnostics 2025, 15(23), 2996; https://doi.org/10.3390/diagnostics15232996 - 25 Nov 2025

Abstract

Artificial intelligence is reshaping breast imaging, yet progress is constrained by data scarcity, privacy restrictions, and uneven representation. This narrative review synthesizes evidence (2020–April 2025) on synthetic data and generative AI—principally GANs and diffusion models—in mammography and related modalities. We examine how synthetic [...] Read more.

Artificial intelligence is reshaping breast imaging, yet progress is constrained by data scarcity, privacy restrictions, and uneven representation. This narrative review synthesizes evidence (2020–April 2025) on synthetic data and generative AI—principally GANs and diffusion models—in mammography and related modalities. We examine how synthetic images enable data augmentation, class balancing, external validation, and simulation-based training; summarize reported gains in detection performance; and assess their potential to mitigate or, if misapplied, amplify bias across subgroups (age, density, ethnicity). We analyze threats to validity, including enriched cohorts, distribution shift, and unverifiable realism, and address medico-legal exposure, image provenance, and deepfake risks. Finally, we outline task-specific validation and reporting practices, equity auditing across density and demographics, and governance pathways aligned with EU/US regulatory expectations. Synthetic data and generative AI can enhance performance, training, and data sharing; however, responsible clinical adoption requires rigorous validation, transparency on failure modes, tamper-evident provenance, and shared accountability models. Full article

(This article belongs to the Special Issue Deep Learning in Biomedical Signal Analysis)

8 pages, 1171 KB

Open AccessInteresting Images

Uterine Angiomyolipoma Presenting as a Rapidly Growing Uterine Mass in a Postmenopausal Woman

by Dae Hyun Song, Hyo Jung An and Jong Chul Baek

Diagnostics 2025, 15(23), 2995; https://doi.org/10.3390/diagnostics15232995 - 25 Nov 2025

Abstract

Uterine angiomyolipoma (AML) is an exceptionally rare mesenchymal tumor of the perivascular epithelioid cell tumor (PEComa) family. Most cases are benign and exhibit a triphasic histologic pattern. Although extragenital PEComas typically show strong, diffuse HMB-45 reactivity, uterine AMLs/PEComas often exhibit weak or negative [...] Read more.

Uterine angiomyolipoma (AML) is an exceptionally rare mesenchymal tumor of the perivascular epithelioid cell tumor (PEComa) family. Most cases are benign and exhibit a triphasic histologic pattern. Although extragenital PEComas typically show strong, diffuse HMB-45 reactivity, uterine AMLs/PEComas often exhibit weak or negative staining, thereby introducing diagnostic uncertainty. We describe a rare case of uterine AML with diffuse nuclear atypia in a postmenopausal woman, which mimicked a degenerative leiomyoma or leiomyosarcoma. A 49-year-old postmenopausal woman presented with the rapid enlargement of a uterine mass that had been followed for four years as a presumed leiomyoma. Imaging revealed a well-circumscribed uterine mass with heterogeneous enhancement, cystic degeneration, and restricted diffusion on MRI. A total hysterectomy was performed. Grossly, the tumor measured 8 cm. Microscopically, it consisted of pleomorphic epithelioid cells (70%), mature adipose tissue (20%), and thick-walled vessels. Immunohistochemistry revealed diffuse smooth muscle actin (SMA) positivity, while Human Melanoma Black (HMB)-45 and Melan-A were negative. Only one mitosis per 50 HPF was identified, with no atypical mitoses or necrosis, and the Ki-67 index was low (<5%). The patient has remained disease-free for 56 months post-surgery. This case represents the first documented HMB-45-negative uterine angiomyolipoma with diffuse nuclear atypia, characterized by a low mitotic index, low Ki-67 proliferation rate, and a benign 56-month follow-up. It broadens the morphologic spectrum of uterine AML, demonstrating that diffuse nuclear atypia can occur in HMB-45-negative tumors with benign behavior, and that atypia alone should not be interpreted as evidence of malignancy. Recognition of this rare variant is essential to avoid misdiagnosing it as leiomyosarcoma. Full article

(This article belongs to the Section Clinical Diagnosis and Prognosis)

► Show Figures

Figure 1

22 pages, 7579 KB

Open AccessHypothesis

Codify and Localize Lesions on a Coronary Acoustic Map: Scientific Rationale, Trial Design and Artificial Intelligence Algorithm Protocols

by Thach Nguyen, Khiem Ngo, Hoang Anh Tien, Dzung T. Ho, Chinh D. Nguyen, Loc T. Vu, Mihas Kodenchery, Huynh Hung, Vinh X. Huynh, Aravinda Nanjundappa and Michael Gibson

Diagnostics 2025, 15(23), 2994; https://doi.org/10.3390/diagnostics15232994 - 25 Nov 2025

Abstract

In coronary artery disease (CAD), the initiation, progression, and regression of atherosclerosis remain incompletely understood, limiting the effectiveness of specific diagnostic and personalized medicine management strategies based on current imaging and assessment methods. In this scientific rationale and study design analysis, the framework [...] Read more.

In coronary artery disease (CAD), the initiation, progression, and regression of atherosclerosis remain incompletely understood, limiting the effectiveness of specific diagnostic and personalized medicine management strategies based on current imaging and assessment methods. In this scientific rationale and study design analysis, the framework conceptualizes the cardiovascular system as an integrated hydraulic network of pumps and pipes, advancing a shift from static imaging of luminal stenosis toward dynamic assessment of coronary flow. Grounded in fluid mechanics and acoustic principles, this analysis establishes a scientific rationale for an angiographic investigation of hemodynamic disturbances that compromise endothelial integrity in coronary arteries. The first section examines injury arising from repetitive flexion and extension of coronary segments driven by left ventricular contraction, most prominent at the transition from diastole to systole. The second section evaluates the hypothetical effects of thickened boundary layers and intimal injury caused by oxygen deprivation along the proximal portion of the outer curvature of side branches. The third section explores the hypothetical role of recirculating flow in accelerating lesion development at these sites. The fourth section presents an acoustic-based diagnostic framework for assessing the hypothetical impact of retrograde pressure-wave propagation associated with water-hammer phenomena. Collectively, these mechanisms establish the systematic codification and spatial delineation of coronary lesions as represented on the coronary acoustic map. Building on these insights, the present analysis proposes a clinical trial framework integrating AI-driven algorithmic protocols to rigorously assess the diagnostic performance and predictive accuracy of the coronary acoustic map. Full article

(This article belongs to the Special Issue Innovative Solutions to Cardiovascular Dilemmas: From Diagnosis to Therapy, Second Edition)

► Show Figures

Figure 1

13 pages, 16473 KB

Open AccessArticle

Logarithmic Scaling of Loss Functions for Enhanced Self-Supervised Accelerated MRI Reconstruction

by Jaejin Cho

Diagnostics 2025, 15(23), 2993; https://doi.org/10.3390/diagnostics15232993 - 25 Nov 2025

Abstract

Background/Objectives: Magnetic resonance imaging (MRI) is a widely used non-invasive imaging modality that provides high-fidelity soft-tissue contrast without ionizing radiation. However, acquiring high-resolution MRI scans is time-consuming, necessitating accelerated acquisition and reconstruction methods. Recently, self-supervised learning approaches have been introduced for reconstructing undersampled [...] Read more.

Background/Objectives: Magnetic resonance imaging (MRI) is a widely used non-invasive imaging modality that provides high-fidelity soft-tissue contrast without ionizing radiation. However, acquiring high-resolution MRI scans is time-consuming, necessitating accelerated acquisition and reconstruction methods. Recently, self-supervised learning approaches have been introduced for reconstructing undersampled MRI data without external fully sampled ground truth. Methods: In this work, we propose a logarithmic scaled scheme for conventional loss functions (e.g.,

ℓ_{1}

,

ℓ_{2}

) to enhance self-supervised MRI reconstruction. Standard self-supervised methods typically compute loss in the k-space domain, which tends to overemphasize low spatial frequencies while under-representing high-frequency information. Our method introduces a logarithmic scaling to adaptively rescale residuals, emphasizing high-frequency contributions and improving perceptual quality. Results: Experiments on public datasets demonstrate consistent quantitative improvements when the proposed log-scaled loss is applied within a self-supervised MRI reconstruction framework. Conclusions: The proposed approach improves reconstruction fidelity and perceptual quality while remaining lightweight, architecture-agnostic, and readily integrable into existing self-supervised MRI reconstruction pipelines. Full article

(This article belongs to the Special Issue 3rd Edition: AI/ML-Based Medical Image Processing and Analysis)

► Show Figures

Figure 1

13 pages, 519 KB

Open AccessReview

Modeling Working Memory in Neurodegeneration: A Focus on EEG Methods

by Yuliya Komarova, Alexander Zakharov, Mariya Sergeeva, Natalia Romanchuk, Tatyana Vladimirova and Igor Shirolapov

Diagnostics 2025, 15(23), 2992; https://doi.org/10.3390/diagnostics15232992 - 25 Nov 2025

Abstract

Working memory is one of the most vulnerable cognitive domains in neurodegenerative diseases. According to the World Health Organization, around 55 million people worldwide were living with dementia in 2021, a number projected to exceed 150 million by 2050. Impairments in working memory [...] Read more.

Working memory is one of the most vulnerable cognitive domains in neurodegenerative diseases. According to the World Health Organization, around 55 million people worldwide were living with dementia in 2021, a number projected to exceed 150 million by 2050. Impairments in working memory occur in 80–90% of patients with Alzheimer’s disease, 40–60% with Parkinson’s disease, and about 50% with frontotemporal dementia. These deficits include reduced information capacity, slower response times, increased errors in manipulation, and difficulties in maintaining information, making them sensitive indicators of progressive decline. This review aims to systematize current approaches to modeling working memory phenotypes using electroencephalography (EEG). It highlights experimental paradigms applied to probe working memory, methods of EEG signal processing and analysis, and the integration of machine learning and neural network models. Particular emphasis is placed on studies achieving high diagnostic accuracy, with classification rates of 85–90% when distinguishing patients with neurodegeneration from healthy participants. Limitations of existing methods, especially EEG variability, are considered. The review concludes by outlining future directions: integration of multimodal EEG data, application of artificial intelligence, and development of digital cognitive biomarkers for hybrid models capable of predicting cognitive decline and advancing clinical translation. Full article

(This article belongs to the Section Medical Imaging and Theranostics)

► Show Figures

Figure 1

18 pages, 5768 KB

Open AccessReview

Diagnostic Advances and Public Health Challenges for Monkeypox Virus: Clade-Specific Insight and the Urgent Need for Rapid Testing in Africa

by Caroline N. Sambo, Amanda Skepu, Nolwandle P. Nxumalo and Ketlareng L. Polori

Diagnostics 2025, 15(23), 2991; https://doi.org/10.3390/diagnostics15232991 - 25 Nov 2025

Abstract

Background: Monkeypox (MPX), caused by the Monkeypox virus (MPOX) of the Orthopoxvirus genus, has re-emerged as a significant global health threat. Once confined to Central and West Africa, the 2022–2025 multi-country outbreaks, predominantly caused by Clade IIb, demonstrated sustained human-to-human transmission and global [...] Read more.

Background: Monkeypox (MPX), caused by the Monkeypox virus (MPOX) of the Orthopoxvirus genus, has re-emerged as a significant global health threat. Once confined to Central and West Africa, the 2022–2025 multi-country outbreaks, predominantly caused by Clade IIb, demonstrated sustained human-to-human transmission and global spread. Objective: This review summarizes current knowledge on MPX virology, epidemiology, clinical presentation, and diagnostic technologies, with a focus on innovations supporting rapid and field-deployable detection in resource-limited settings. Methods: The recent literature (2019–2025), including peer-reviewed studies, WHO and Africa CDC reports, and clinical guidelines, was critically reviewed. Data were synthesized to outline key developments in diagnostic methodologies and surveillance approaches. Results: MPX comprises two genetic clades: Clade I (Congo Basin) and Clade II (West African), which differ in virulence and transmission. Clade IIb is associated with sexual and close-contact transmission during recent outbreaks. Clinical manifestations have shifted from classic disseminated rash to localized anogenital lesions and atypical or subclinical infections. RT-PCR remains the diagnostic gold standard, while emerging assays such as loop-mediated isothermal amplification (LAMP), recombinase polymerase amplification (RPA), and CRISPR/Cas-based platforms show promise for rapid point-of-care (POC) testing. Complementary serological tools, including ELISA and lateral flow assays, enhance surveillance and immune profiling. Conclusions: The resurgence of MPX highlights the urgent need for accessible, sensitive, and specific diagnostic platforms to strengthen surveillance and outbreak control, especially in endemic and resource-constrained regions. Full article

(This article belongs to the Section Diagnostic Microbiology and Infectious Disease)

► Show Figures

Figure 1

9 pages, 1608 KB

Open AccessCase Report

Tibial Nerve Palsy Secondary to Spontaneous Isolated Popliteus Muscle Rupture and Localized Compartment Syndrome

by Sophie Jolliet, Yves Harder and Sébastien Durand

Diagnostics 2025, 15(23), 2990; https://doi.org/10.3390/diagnostics15232990 - 25 Nov 2025

Abstract

We report the case of a 68-year-old man who developed sudden pain at the right calf, followed by progressive tibial nerve palsy. Magnetic resonance imaging (MRI) demonstrated localized signal abnormalities and swelling of the popliteus muscle compressing the tibial nerve. A medial surgical [...] Read more.

We report the case of a 68-year-old man who developed sudden pain at the right calf, followed by progressive tibial nerve palsy. Magnetic resonance imaging (MRI) demonstrated localized signal abnormalities and swelling of the popliteus muscle compressing the tibial nerve. A medial surgical approach in supine position provided direct access to both the popliteus muscle and the tibial nerve while minimizing operative risk. Intraoperative findings confirmed necrosis of the popliteus muscle, which was subsequently supported by histopathology. Surgical decompression consisted of debridement of the necrotic tissue, associated with the release of the tibial nerve at the soleus arch. This case highlights a dual mechanism of tibial nerve compression: 1. swelling of the popliteus muscle following spontaneous rupture exerting a direct mass effect and 2. concomitant localized compartment syndrome. This dual mechanism led to significant transient tibial nerve palsy, which was successfully reversed following surgical decompression, with sustained recovery at ten months. Full article

(This article belongs to the Section Clinical Diagnosis and Prognosis)

► Show Figures

Figure 1

11 pages, 4202 KB

Open AccessArticle

Pattern of Mandibular Bone Invasion as a Prognostic Factor

by Richard Pink, Jaroslav Michálek, Zdeněk Dvořák, Peter Tvrdý, Lenka Šašková, Michal Herman, Petr Heinz, Markéta Hermanová, Jana Zapletalová and Michal Mozoľa

Diagnostics 2025, 15(23), 2989; https://doi.org/10.3390/diagnostics15232989 - 25 Nov 2025

Abstract

Background/Objectives: Mandibular bone invasion is a common and clinically relevant feature of OSCC, particularly in tumors of the lower alveolus, floor of mouth, and retromolar trigone. The prognostic value of the pattern of invasion—rather than its mere presence—remains insufficiently defined. Therefore, we evaluated [...] Read more.

Background/Objectives: Mandibular bone invasion is a common and clinically relevant feature of OSCC, particularly in tumors of the lower alveolus, floor of mouth, and retromolar trigone. The prognostic value of the pattern of invasion—rather than its mere presence—remains insufficiently defined. Therefore, we evaluated the prognostic relevance of erosive vs. infiltrative mandibular invasion and the diagnostic reliability of preoperative CT. Methods: This retrospective, single-center observational cohort study included 83 patients with OSCC involving or adjacent to the mandible who underwent surgical resection at the Department of Oral and Maxillofacial Surgery, University Hospital Olomouc (2008–2018). Bone invasion type was classified histopathologically as erosive or infiltrative. Survival outcomes were analyzed using Kaplan–Meier and Cox regression methods. Correlation between radiologic and histologic findings was assessed using Cohen’s kappa statistics. Results: Mandibular invasion was confirmed in 50.6% of cases, of which roughly two-thirds were infiltrative. DSS differed across invasion groups (log-rank p = 0.012): infiltrative had a median DSS of 14.5 months (95% CI 0.0–32.8), no invasion had 54.2 months (CI not estimable), while erosive did not reach the median (fewer than half experienced the event). In the adjusted model (covariates: invasion type, ENE, grade, margins), infiltrative vs. no invasion was associated with worse DSS (aHR 1.93, 95% CI 1.02–3.64; p = 0.042); for OS, erosive vs. no invasion showed a protective association (aHR 0.39, 95% CI 0.16–0.96; p = 0.041). Positive/close margins were independently adverse across endpoints (e.g., DSS aHR 3.30, 95% CI 1.74–6.22). CT–histology agreement for bone invasion was κ = 0.45, indicating moderate agreement. Conclusions: In this retrospective single-center cohort, the pattern of mandibular bone invasion was associated with survival: infiltrative invasion aligned with worse outcomes, whereas erosive behaved similarly to no invasion, particularly for OS. Prospective, multicenter validation is warranted before routine incorporation into risk stratification or treatment selection. Full article

(This article belongs to the Special Issue Advanced Diagnosis and Integrated Management of Bone and Soft Tissue Tumors)

► Show Figures

Figure 1

20 pages, 4846 KB

Open AccessCase Report

Diagnosis of Isolated Saccular Dysfunction Using Trapezius cVEMP: A Detailed Vestibular Assessment

by Mădălina Georgescu, Oana Irina Popa, Horațiu Ștefănescu, Violeta Necula, Alma Maniu, Irina Enache and Andrei Osman

Diagnostics 2025, 15(23), 2988; https://doi.org/10.3390/diagnostics15232988 - 25 Nov 2025

Abstract

Background and Clinical Significance: Vestibular disorders include a wide range of conditions with overlapping symptoms such as dizziness, vertigo and imbalance, often offering diagnostic challenges when distinguishing between peripheral and central etiology. Accurate differentiation is essential for establishing effective treatment plans. In [...] Read more.

Background and Clinical Significance: Vestibular disorders include a wide range of conditions with overlapping symptoms such as dizziness, vertigo and imbalance, often offering diagnostic challenges when distinguishing between peripheral and central etiology. Accurate differentiation is essential for establishing effective treatment plans. In rare or atypical cases with subtle findings, comprehensive diagnostic tools—such as extended vestibular tests and structured questionnaires like the Dizziness Handicap Inventory (DHI)—are critical for diagnosis and monitoring patient recovery. Case Presentation: A 35-year-old female presented with chronic imbalance and motion-induced dizziness persisting for four years. The patient had a surgical history of right-sided functional neck dissection for a parotid tumor. A comprehensive audiovestibular evaluation was performed, including pure tone audiometry (PTA), tympanometry, videonystagmography (VNG), cervical vestibular evoked myogenic potentials (cVEMP), ocular vestibular evoked myogenic potentials (oVEMP), video head impulse testing (vHIT), computerized dynamic posturography (CDP), and magnetic resonance imaging (MRI). The Dizziness Handicap Index (DHI) was administered at baseline and post-treatment to monitor subjective symptom changes. Objective testing revealed marked right–left amplitude asymmetry on cVEMP, which were recorded from the trapezius muscle due to prior neck dissection surgery, indicating isolated right-sided saccular hypofunction. Following targeted vestibular rehabilitation and pharmacologic treatment, the 3-month reassessment demonstrated resolution of symptoms and a reduction in DHI scores from 24 to 6. Conclusions: Comprehensive vestibular testing, performed in a single diagnostic session, enabled the accurate identification of isolated right-sided saccular hypofunction in this complex post-surgical case. Combining cVEMP, CDP, and DHI assessment provided a complete functional profile, guided targeted rehabilitation, and allowed objective monitoring of recovery. Full article

(This article belongs to the Special Issue Research Updates in Vestibular Dysfunction: Diagnostic Breakthroughs)

► Show Figures

Figure 1

12 pages, 8082 KB

Open AccessArticle

Anatomical Insights into the Lateral Meniscus and Anterolateral Ligament: A Cadaveric Study

by João Lobo, Joana Almeida, José Fernandes, Hélio Alves, André Rodrigues Pinho, Maria Dulce Madeira, Levi Fernandes, Ana Esteves and Pedro Alberto Pereira

Diagnostics 2025, 15(23), 2987; https://doi.org/10.3390/diagnostics15232987 - 24 Nov 2025

Abstract

Background/Objectives: This study aims to describe in detail the previously reported close relationship between the anterolateral ligament (ALL) and the lateral meniscus. Few previous studies identified and characterized this relation. This study further characterizes the anatomical relation between the ALL and the lateral [...] Read more.

Background/Objectives: This study aims to describe in detail the previously reported close relationship between the anterolateral ligament (ALL) and the lateral meniscus. Few previous studies identified and characterized this relation. This study further characterizes the anatomical relation between the ALL and the lateral meniscus through meticulous cadaveric dissection. Methods: A total of 31 cadaver knees were dissected. The ALL relation with the lateral meniscus was explored using a specific dissection protocol that involved removing the central pivot (cruciate ligaments) and medial structures of the knee to enhance visualization of the anterolateral complex. The zone and attachment pattern of the ALL on the lateral meniscus were recorded. Results: The ALL was found in all 31 dissected knees and in all cases has an attachment to the lateral meniscus. It was attached in zone 2b of the lateral meniscus in 97% of cases. The median anteroposterior length of attachment of the ALL on the lateral meniscus was 6 mm (25th and 75th percentiles of 5–7 mm). Almost 80% (77.4%) of ALL attachments on the lateral meniscus were full thickness or bipolar (superior and inferior margins). In the remaining knees, the ALL was fixed only in the upper part (4 knees, 12.9%) or only in the lower part (3 knees, 9.7%) of the lateral meniscus. Conclusions: The ALL has an attachment to the lateral meniscus in all studied knees, with its most prevalent site at zone 2b. The most frequent types of ALL attachment on the lateral meniscus were full thickness or bipolar. These anatomic insights support targeted anterolateral augmentation and meniscal preservation to optimize clinical results. Full article

(This article belongs to the Special Issue Advances in Anatomy—Third Edition)

► Show Figures

Figure 1

20 pages, 2812 KB

Open AccessArticle

A Novel Deep Learning Framework for Liver Fibrosis Staging and Etiology Diagnosis Using Integrated Liver–Spleen Elastography

by Kai Yang, Fei Chen, Aiping Tian, Long Deng and Xiaorong Mao

Diagnostics 2025, 15(23), 2986; https://doi.org/10.3390/diagnostics15232986 - 24 Nov 2025

Abstract

Objectives: Liver fibrosis staging and etiology diagnosis are critical for patient management, but non-invasive methods remain challenging. This study aims to evaluate the performance of radiomics models using 2D shear wave elastography (2D-SWE) of the liver and spleen for liver fibrosis staging and [...] Read more.

Objectives: Liver fibrosis staging and etiology diagnosis are critical for patient management, but non-invasive methods remain challenging. This study aims to evaluate the performance of radiomics models using 2D shear wave elastography (2D-SWE) of the liver and spleen for liver fibrosis staging and etiology differentiation, comparing them with serum biomarkers and conventional ultrasound. Methods: A retrospective analysis was conducted on 198 patients with liver fibrosis confirmed by biopsy. Radiomics features were extracted from the liver and spleen grayscale and 2D-SWE images. Machine learning (ML) and transfer learning (TL) models were established for fibrosis staging and etiology diagnosis. Model performance was evaluated according to receiver operating characteristic (ROC) curves. Results: For fibrosis staging, 2D-SWE-based models outperformed grayscale and serum biomarkers. The combined liver–spleen TL model achieved exceptional validation performance (AUCs 0.99 for S4, 0.98 for ≥S3, 1.00 for ≥S2). For etiology diagnosis, the liver 2D-SWE TL model and the combined liver–spleen TL model achieved AUCs of 0.97 and 0.94, respectively, significantly outperforming ML models in terms of AUC. Conclusions: Integrating liver and spleen 2D-SWE radiomics with TL significantly improves non-invasive liver fibrosis staging and etiology diagnosis, offering superior accuracy over conventional methods. This approach holds promise for clinical application, though further validation is needed. Full article

(This article belongs to the Special Issue The Role of AI in Ultrasound)

► Show Figures

Figure 1

15 pages, 420 KB

Open AccessArticle

Complementary Role of Ultrasound and Clinical Features in Assessing Carpal Tunnel Syndrome Severity: A Cross-Sectional Study

by Daniela Nicoleta Popescu, Claudiu Costinel Popescu, Oana Morari, Natalia Blidaru, Alina Dima, Ioana Adriana Catanoiu, Alice Rakoczy, Ioana Otobic, Magda Ileana Parvu, Catalin Codreanu and Luminita Enache

Diagnostics 2025, 15(23), 2985; https://doi.org/10.3390/diagnostics15232985 - 24 Nov 2025

Abstract

Background/Objectives: The goal of this study was to assess the correlation between ultrasound measurements and nerve conduction study (NCS)-defined carpal tunnel syndrome (CTS) severity and to explore clinical and demographic factors associated with CTS severity in a sample of Romanian patients. Methods [...] Read more.

Background/Objectives: The goal of this study was to assess the correlation between ultrasound measurements and nerve conduction study (NCS)-defined carpal tunnel syndrome (CTS) severity and to explore clinical and demographic factors associated with CTS severity in a sample of Romanian patients. Methods: We prospectively evaluated consecutive patients with clinically diagnosed CTS. All patients underwent standardized clinical assessment, ultrasonographic examination of the median nerve, and NCS. CTS severity was graded electrophysiologically (three-level scale), and associations with demographic, clinical, and ultrasound parameters were examined using univariate analyses and multivariable generalized estimating equation (GEE) models to account for within-patient clustering. Results: Among 193 CTS hands (100 patients, mean age 58 years, 93% female), electrophysiological severity correlated significantly with several ultrasound and clinical parameters. In multivariable GEE models, the presence of nocturnal symptoms, sensory loss, thenar weakness/atrophy, male sex, larger maximal median nerve cross-sectional area (mCSA), and impaired median nerve mobility were independent predictors of higher NCS-defined severity. Pseudo-R² increased from 0.04 in the core clinical model to 0.25 when ultrasound parameters were included, indicating improved model performance. Conclusions: Ultrasound parameters, particularly mCSA and median nerve mobility, together with clinical features, such as nocturnal symptoms, sensory loss, and thenar weakness, are independently associated with NCS-defined CTS severity. These findings support the complementary role of ultrasound alongside NCS in severity grading and highlight its potential to guide timely diagnosis and management. Full article

(This article belongs to the Section Medical Imaging and Theranostics)

► Show Figures

Figure 1

20 pages, 2271 KB

Open AccessSystematic Review

Prevalence of Radial Artery Variants and Their Relationship with Clinical Considerations of the Antebrachial Region: Systematic Review and Meta-Analysis

by Juan Sanchis-Gimeno, Jessica Paola Loaiza-Giraldo, Yael Alruiz, Maximiliano Vergara, Maria Fernanda Navia, Camila Roman, Alejandra Suazo-Santibañez, Pablo Nova-Baeza, Mathias Orellana-Donoso, Gustavo Oyanedel-Amaro, Macarena Rodriguez-Luengo, Alejandro Bruna-Mejias, Juan José Valenzuela-Fuenzalida, Jose E. León-Rojas and Guinevere Granite

Diagnostics 2025, 15(23), 2984; https://doi.org/10.3390/diagnostics15232984 - 24 Nov 2025

Abstract

Background: The radial artery (RA) is one of the terminal branches of the brachial artery, extending along the lateral forearm, crossing the anatomical snuffbox, and contributing to the palmar arches. Anatomical variations in the RA are of great clinical relevance due to their [...] Read more.

Background: The radial artery (RA) is one of the terminal branches of the brachial artery, extending along the lateral forearm, crossing the anatomical snuffbox, and contributing to the palmar arches. Anatomical variations in the RA are of great clinical relevance due to their implications in procedures such as transradial catheterization, arterial cannulation, and bypass grafting. These variants may alter the course, branching pattern, or origin of the vessel, potentially increasing procedural complexity and the risk of iatrogenic injury. In critically ill patients and in surgical or interventional settings, accurate identification of RA anatomy is essential. The objective of this study was to systematically identify and describe RA variants reported in the scientific literature and to analyze their clinical relevance. Methods: A systematic search was conducted across six electronic databases: Medline, Scopus, Web of Science, Google Scholar, Cumulative Index to Nursing and Allied Health Literature (CINAHL), and Latin American and Caribbean Literature in Health Sciences (LILACS), covering publications up to July 2025. Eligible studies included anatomical, radiological, and surgical investigations reporting RA variants. Study quality was evaluated using the Assessment of Quality in Anatomical Studies (AQUA) tool. Quantitative synthesis was performed using a random-effects model to estimate the pooled prevalence of RA variants and subgroup differences. Twenty-three studies met the inclusion criteria, and eleven were included in the meta-analysis, encompassing a total of 6320 participants. Results: Radial artery variants were categorized into three main types: variations in origin, course, and branching pattern. The pooled global prevalence of RA variants was 12% (95% CI: 6–18%), with substantial heterogeneity (I² = 97.7%). Higher prevalence was found in imaging-based studies (14%) compared with donor-based studies (12%). Sex-based subgroup analysis revealed a higher prevalence in females (18%; CI: 9–28%) compared with males (3%; CI: 3–4%), with moderate heterogeneity (I² = 61.3%). Regionally, European populations demonstrated a higher prevalence (20%) than Asian populations (11%), both showing high heterogeneity (I² > 98%). Notably, only one study from the Americas and none from Africa or Oceania were identified, representing a major geographical limitation in the available data. The findings of this study highlight the considerable variability in RA anatomy across populations. Such variations hold significant clinical importance, particularly in the context of transradial interventions, arterial cannulation, and reconstructive procedures where vascular integrity is critical. The high degree of heterogeneity observed may reflect differences in population genetics, sample size, and imaging or dissection methodologies. The limited representation of certain regions underscores the need for further anatomical and radiological studies to obtain a more comprehensive understanding of global RA variability. Preoperative or pre-procedural imaging using Doppler ultrasonography or computed tomography angiography is recommended to identify anomalous patterns and minimize iatrogenic complications. Conclusions: Radial artery variants are frequent and diverse. Their recognition is fundamental for the safety and success of invasive and surgical procedures in the upper limb. A standardized approach to vascular evaluation, particularly through preoperative imaging, is essential to improve procedural outcomes and reduce the risk of arterial injury in clinical practice. Full article

(This article belongs to the Section Clinical Diagnosis and Prognosis)

► Show Figures

Figure 1

13 pages, 545 KB

Open AccessArticle

Factors Influencing Stroke Severity Based on Collateral Circulation, Clinical Markers and Machine Learning

by Jia-Lang Xu

Diagnostics 2025, 15(23), 2983; https://doi.org/10.3390/diagnostics15232983 - 24 Nov 2025

Abstract

Background/Objectives: Stroke is a serious neurological disorder that significantly affects patients’ quality of life and overall health. The severity of a stroke can vary widely and is influenced by multiple factors, such as clinical presentation, diagnostic findings, and the site of onset. This [...] Read more.

Background/Objectives: Stroke is a serious neurological disorder that significantly affects patients’ quality of life and overall health. The severity of a stroke can vary widely and is influenced by multiple factors, such as clinical presentation, diagnostic findings, and the site of onset. This study aimed to identify and analyze key variables that contribute to stroke severity, with a particular focus on the role of collateral circulation. Methods: This study analyzed clinical, imaging, and biochemical variables—ipsilateral collateral flow on MRA, MRI unilateral–bilateral stroke, systolic blood pressure (SBP), fasting plasma glucose (FPG), and blood urea nitrogen (BUN). Group differences used chi-square and Mann–Whitney U tests. Class imbalance was addressed with SMOTE; Logistic Regression, Random Forest, XGBoost, and SVM were cross-validated, reporting accuracy, precision, recall, and F1 with 95% CIs. Results: Reduced or absent ipsilateral collateral flow and unilateral–bilateral stroke were strongly associated with greater severity (p < 0.001). SBP was significant (p = 0.034), FPG was significant (p = 0.023), and BUN was borderline (p = 0.059). SMOTE improved prediction: Random Forest achieved accuracy 83.3% (CI: 79.1–87.6) and F1 84.0% (CI: 79.1–88.9); XGBoost reached accuracy 80.2% (CI: 71.5–89.0) and F1 81.4% (CI: 73.8–89.0). Logistic Regression improved to F1 70.8% (CI: 55.4–86.2), whereas SVM declined to accuracy 52.2% (CI: 37.5–67.0). Conclusions: Collateral status and unilateral–bilateral stroke are key determinants of severity; SBP and FPG add prognostic value, with BUN borderline. Tree-based ensembles trained on SMOTE-balanced data provide the most reliable predictions for risk stratification. These findings suggest that future work may focus on integrating such predictive models into Clinical Decision Support Systems (CDSSs) to enhance early risk identification, strengthen CDSSs, and enable more personalized care planning for stroke patients. Full article

(This article belongs to the Special Issue Artificial Intelligence in Diagnostics: From Algorithms to Clinical Impact)

► Show Figures

Figure 1

13 pages, 4834 KB

Open AccessArticle

Validation of Body Surface Area Equations for Estimating Fat-Free Mass by Dual X-Ray Absorptiometry in a Regional Chilean Sample Aged 4 to 85 Years

by Marco Cossio-Bolaños, Rubén Vidal Espinoza, Jose Sulla-Torres, Camilo Urra-Albornoz, Lucila Sanchez-Macedo, Miguel de Arruda, Fernando Alvear-Vasquez, Evandro Lazari and Rossana Gomez-Campos

Diagnostics 2025, 15(23), 2982; https://doi.org/10.3390/diagnostics15232982 - 24 Nov 2025

Abstract

Background/Objectives: Body surface area (BSA) is an important metric that represents human dimensionality and could provide a more accurate representation of body composition. The objectives were (a) to verify the validity of a set of equations based on BSA to estimate lean body [...] Read more.

Background/Objectives: Body surface area (BSA) is an important metric that represents human dimensionality and could provide a more accurate representation of body composition. The objectives were (a) to verify the validity of a set of equations based on BSA to estimate lean body mass (LBM), using dual X-ray absorptiometry (DXA) as a reference method and (b) to propose reference values of BSA by anthropometry and LBM by DXA in a regional sample of Chile aged 4 to 85 years. Methods: A descriptive cross-sectional study was performed. The sample size was 5493 participants. Weight and height were measured. BSA was calculated using seven equations. LBM was assessed by DXA. Results: Only three BSA equations (Dubois–Dubois, 1916, Fujimoto, Watanabe, 1969, and Mattar, 1981) best explained LBM. The explanatory power for males was R² = 83 to 84%, and that for females was R² = 69%. The standard error of estimation (SEE) of the three equations showed acceptable values in both sexes. These values ranged from 0.049 to 0.080 in males and from 0.035 to 0.088 in females. The Bland–Altman concordance analysis showed adequate limits of agreement. In men, they ranged from −0.092 to 0.069 m². In females, they ranged from −0.064 to 0.084 m². Reference values for BSA and LBM were constructed using percentiles. Conclusions: This study demonstrated the validity of three equations for estimating LBM in a Chilean sample aged between 4 and 85 years. These results show consistent behavior and acceptable accuracy, especially in the Mattar equation for all ages. However, the Dubois & Dubois and Fujimoto equations could also be an alternative in females. Reference values were generated for BSA and LBM according to age and sex. The results suggest their applicability and usefulness in clinical and public health contexts. Full article

(This article belongs to the Special Issue Recent Advances in Diagnosis and Management of Musculoskeletal Disorders)

► Show Figures

Figure 1

17 pages, 4026 KB

Open AccessArticle

DuXplore: A Dual-Hierarchical Deep Learning Model for Prognostic Prediction of Hepatocellular Carcinoma in Digital Pathology

by Haotian Zhang, Mengling Liu, Xinshen Zhao, Yichen Zhang and Li Sui

Diagnostics 2025, 15(23), 2981; https://doi.org/10.3390/diagnostics15232981 - 24 Nov 2025

Abstract

Background: Spatial heterogeneity in tumor tissue has been linked to patient prognosis. To exploit both structural and semantic cues in whole slide images (WSIs), we propose Dual eXplanatory Framework (DuXplore), a dual-branch deep learning framework that integrates tissue architecture and cellular morphology [...] Read more.

Background: Spatial heterogeneity in tumor tissue has been linked to patient prognosis. To exploit both structural and semantic cues in whole slide images (WSIs), we propose Dual eXplanatory Framework (DuXplore), a dual-branch deep learning framework that integrates tissue architecture and cellular morphology for hepatocellular carcinoma (HCC) prognosis. Method: At the macroscopic level, DuXplore constructs a multi-channel tissue organization probability map (MTOP) to represent the spatial layout of eight tissue categories within the WSIs. At the microscopic level, a feature-guided Fused Structure Tensor (FST) based on tissue composition is employed to extract representative cell morphology patches. Accordingly, MTOP representations are modeled by Macro-Net, while FST-guided patches are modeled by Micro-Net. Each branch produces a 32-dimensional prognostic embedding, which are fused and passed through a multi-layer perceptron with a Cox proportional hazards head to generate patient-level risk predictions. To further elucidate the distinct contributions of the two branches, we conducted model-agnostic interpretability analyses, including occlusion sensitivity mapping (OSM) on MTOP and nuclear morphometrics from CellProfiler on high- versus low-risk tiles. Result: DuXplore achieves promising performance with C-indices of 0.764 on the public Cancer Genome Atlas (TCGA) dataset and 0.713 on the Eastern Hepatobiliary HCC (EHBH) cohort from our clinical center, along with significant patient risk stratification (log-rank p < 0.001). OSM highlighted necrosis and central fibrosis as high-risk and marginal fibrosis as protective; these patterns were corroborated by multivariable Cox using reproducible structural parameters (N-ratio, FIB-center, FIB-edge). Micro-level analysis revealed that higher nuclear staining intensity, increased texture irregularity (GLCM features), and greater morphological heterogeneity characterize high-risk tiles, aligning with pathological understanding. Conclusions: DuXplore advances prognostic modeling by coupling structure-aware micro-sampling with macro architectural encoding, delivering robust, generalizable survival prediction and biologically plausible explanations. While validated on HCC WSIs, broader multi-center, multi-omics studies are warranted to refine sampling scales and enhance clinical translation. Full article

(This article belongs to the Section Machine Learning and Artificial Intelligence in Diagnostics)

► Show Figures

Figure 1

11 pages, 2219 KB

Open AccessArticle

Comparison of Echocardiography and Invasive Transseptal Catheterization for Assessing Transvalvular Gradient in Patients with Surgical Aortic Valve Prostheses: Fact or Myth?

by Ahmet Hakan Ates, Ahmet Kivrak, Ugur Canpolat, Mert Dogan, Gul Sinem Kılıc, Can Menemencioglu, Ugur Nadir Karakulak, Ergun Barıs Kaya, Mehmet Levent Sahiner and Kudret Aytemir

Diagnostics 2025, 15(23), 2980; https://doi.org/10.3390/diagnostics15232980 - 24 Nov 2025

Abstract

Background: Echocardiography is the primary assessment tool for follow-up in patients with aortic valve prostheses. However, there are concerns regarding the consistency between echocardiographic and invasive transvalvular gradients (TVGs). This study utilized both noninvasive and invasive methods to compare the TVGs in aortic [...] Read more.

Background: Echocardiography is the primary assessment tool for follow-up in patients with aortic valve prostheses. However, there are concerns regarding the consistency between echocardiographic and invasive transvalvular gradients (TVGs). This study utilized both noninvasive and invasive methods to compare the TVGs in aortic valve prostheses. Methods: The study included fourteen patients who had previously undergone surgical aortic valve replacement [metallic (n = 12) and bioprosthetic (n = 2)]. All patients had moderate-to-severe TVGs, which were measured during follow-up echocardiography, and they underwent invasive transseptal catheterization. Invasive and echocardiographic TVGs were measured and compared. Results: The median interval between index valvular surgery and invasive TVG measurement was 6.7 (2.5–11.5) years. The median interval between echocardiographic and invasive TVG measurements was 7.2 (2–19) days. Only 12 (85.7%) patients were symptomatic during echocardiographic assessment. Maximum TVGs obtained by echocardiography were higher than invasive peak-to-peak TVGs (77.0 ± 13.1 vs. 47.5 ± 21.7 mmHg, p < 0.05). There was a significant negative correlation between the echocardiography-based aortic valve area and the effective orifice area index with the catheter-based peak-to-peak aortic gradient (r = −0.64, p = 0.014 and r = −0.63, p = 0.015). Six patients (42.9%) who revealed severe catheter-based peak-to-peak aortic gradient underwent redo aortic valve surgeries. The cut-off value of EOAI of <0.50 cm²/m² was found to be a predictor of severe catheter-based peak-to-peak aortic gradient. Conclusions: In our preliminary cohort study, the TVGs of aortic valvular prostheses measured by echocardiography were significantly greater than those measured by invasive transseptal catheterization. During follow-up, invasive confirmation of echocardiographic moderate-to-severe TVGs in selected patients with surgical aortic valvular prostheses may be considered. Full article

(This article belongs to the Section Medical Imaging and Theranostics)

► Show Figures

Figure 1

21 pages, 686 KB

Open AccessArticle

Dyslipidemia Assessed in Pediatric Patients: Validation of LDL-C Assessed by Friedewald Formula, Direct Assessment, and Sampson–NIH Formula

by Joanna Wawer, Agnieszka Chojęta, Jakub Swadźba, Mariola Janiszewska, Michał Chojęta, Genowefa Anna Wawer, Ewelina Grywalska and Anna Milaniuk

Diagnostics 2025, 15(23), 2979; https://doi.org/10.3390/diagnostics15232979 - 24 Nov 2025

Abstract

Background: The epidemic increase in obesity, metabolic syndrome, cardiac disease, or hypertension is associated with lipid deregulation. Studies suggest a strong link between elevated levels of plasma cholesterol and the premature formation of atherosclerotic plaques. Primary prevention of early clinical manifestations of [...] Read more.

Background: The epidemic increase in obesity, metabolic syndrome, cardiac disease, or hypertension is associated with lipid deregulation. Studies suggest a strong link between elevated levels of plasma cholesterol and the premature formation of atherosclerotic plaques. Primary prevention of early clinical manifestations of atherosclerosis allows slowing or preventing the development of several health problems later in adult life. Objectives: The purpose of this study was the validation of LDL-C measured by the Friedewald formula, direct method, and Sampson–NIH Formula. The results of the three methods used to assess LDL-C were compared to check whether the three measurements of LDL-C yielded different results. Methods: The study was conducted in a large cohort of in-patients aged 8 months to 18 years. Lipid profile parameters were determined. Indirect methods for dyslipidemia diagnosis were compared against direct LDL measurement. Incorrect and missed diagnoses were analyzed. To measure the central tendency, a statistical analysis of distributions of numerical variables was used. Differences between categorical variables were assessed. The agreement between pairs of competing methods in estimating LDL concentration was assessed via Bland–Altman analysis. Results: In total, 1982 pediatric patients underwent lipid profile assessment. Significant differences in lipid parameters between boys and girls were observed. TG, TC, and HDL levels were higher in boys. LDL-C as measured by the Friedewald formula and direct methods showed significant differences. Comparison of the direct methods with the Sampson–NIH indicated that the Sampson–NIH formula underestimates LDL values. Conclusions: The analysis revealed differences between the methods used to assess dyslipidemia. A systematic underestimation of LDL concentrations determined by the indirect methods was found. Small differences between the Friedewald and Sampson–NIH methods were observed. Although both indirect methods underestimate LDL levels compared to the direct method, the differences between them are small, though still detectable. Full article

(This article belongs to the Section Clinical Diagnosis and Prognosis)

► Show Figures

Figure 1

20 pages, 575 KB

Open AccessReview

Possible Diagnostic and Therapeutic Applications of Bioprinting for Bone Regeneration in Maxillofacial Surgery

by Lorenzo Marini, Alessandro Tel, Marco Zeppieri, Luca Michelutti, Massimo Robiony, Caterina Gagliano, Fabiana D’Esposito, Matteo Capobianco, Tamara Ius and Marieme Khouyyi

Diagnostics 2025, 15(23), 2978; https://doi.org/10.3390/diagnostics15232978 - 24 Nov 2025

Abstract

Background: The integration of 3D bioprinting, biomaterials science, and cellular biology presents a viable strategy for maxillofacial bone regeneration, overcoming the constraints of traditional graft techniques. This review offers a thorough examination of the present condition, obstacles, uses, and future outlook of [...] Read more.

Background: The integration of 3D bioprinting, biomaterials science, and cellular biology presents a viable strategy for maxillofacial bone regeneration, overcoming the constraints of traditional graft techniques. This review offers a thorough examination of the present condition, obstacles, uses, and future outlook of 3D bioprinting technology in maxillofacial bone regeneration. An essential understanding has been attained by analyzing the technological constraints, specifically in vascularization and neuro-integration, and by delineating the vital translational pathway from benchtop models to clinical application. We have examined several bioprinting techniques—namely extrusion, inkjet, and laser-assisted methods—and the requisite bioinks, emphasizing their physicochemical and biological features vital for osteogenesis. Significant clinical applications, including the treatment of trauma-induced abnormalities and the reconstruction of oncology-related resections, have been emphasized. This review highlights the urgent necessity for established regulatory frameworks and refined printing settings to guarantee effective, functional, and durable bone substitutes, providing a distinct pathway for future research and clinical implementation in this specialized surgical domain. Aim: The purpose of this review was to present a general overview of the current clinical and diagnostic applications of bioprinting in bone tissue engineering for the reconstruction of bone defects. Methods: A search of major scientific databases, including PubMed, Science Direct, Embase, and Cochrane, was conducted. Articles published within the last 10 years that analyze the possible applications of bioprinting in bone tissue fabrication were included. Results: Several bioinks, based on hydrogels and stem cells, can enable the fabrication of such tissues using this technology. This review reports on the processes adopted for the bioprinting of bone tissue, the bioinks used, and cell cultivation methods. Conclusions: Bioprinting represents a promising solution for bone regeneration with potential applications that could revolutionize current surgical practices, despite the many challenges that future research will face. Full article

(This article belongs to the Section Clinical Diagnosis and Prognosis)

► Show Figures

Figure 1

Journal Description

Diagnostics

Latest Articles

Journal Menu

Journal Browser

Highly Accessed Articles

Latest Books

E-Mail Alert

News

Topics

Conferences

Special Issues

Topical Collections

Further Information

Guidelines

MDPI Initiatives

Follow MDPI