Tomography, Volume 12, Issue 9 (September 2026) – 14 articles

This narrative review synthesizes a comprehensive multimodal characterization framework for atrial fibrillation (AF), tracing its progression from patient-specific anatomical reconstruction to electrophysiological phenotyping and standardized spatial mapping. The literature was identified through searches of PubMed, Web of Science, and Google Scholar, covering publications from January 2010 to June 2026, with additional studies identified from relevant references. Anatomical characterization leverages clinical imaging modalities alongside advanced deep learning models to execute precise whole-chamber, subregional, and tissue-level modeling. Electrophysiological profiling spans multi-scale modalities, including 12-lead electrocardiography (ECG), body surface potential mapping (BSPM), electrocardiographic imaging (ECGI), and electroanatomical mapping (EAM), complemented by wearable sensors for longitudinal rhythm surveillance. To bridge heterogeneous datasets across subjects and modalities, standardized coordinate systems and multimodal registration enable reproducible data integration. These integrated data parameterize patient-specific computational models to advance mechanistic insight, risk stratification, and personalized AF management. However, broad clinical translation remains constrained by imaging variability, reconstruction noise, registration uncertainty, and limited prospective multicenter validation. Future progress hinges on moving beyond technical accuracy toward prospective trials evaluating algorithm-guided clinical utility. Full article

Interventional radiology (IR) has reshaped contemporary management strategies for selected gynecological and obstetric disorders by introducing minimally invasive, organ-sparing therapeutic alternatives to conventional surgical approaches. This review examines the current applications of IR in the treatment of symptomatic uterine fibroids, pelvic congestion syndrome, placenta accreta spectrum, cesarean scar pregnancy, and selected gynecological malignancies requiring palliative or preoperative embolization. In addition, selected vascular conditions in pregnancy, including visceral and cerebral aneurysms, are discussed due to their high-risk clinical profile and growing relevance for endovascular management. These entities were chosen because of their substantial clinical relevance, risk of severe complications, and the expanding body of evidence supporting endovascular management. Techniques such as uterine artery embolization and ovarian vein embolization are associated with high technical success and meaningful symptom reduction, while simultaneously decreasing hospitalization time, perioperative morbidity, and convalescence compared with traditional surgery. In complex obstetric scenarios—including placenta accreta spectrum and splenic artery aneurysm—prophylactic balloon occlusion and endovascular embolization may significantly limit life-threatening hemorrhage and improve maternal prognosis in appropriately selected patients when implemented within a multidisciplinary setting; for splenic artery aneurysm in particular, this evidence currently derives from case reports and small case series rather than comparative trials. Imaging modalities, including ultrasonography, computed tomography, magnetic resonance imaging, and angiography, are integral to diagnosis, patient qualification, procedural planning, and post-intervention monitoring. Although existing data generally confirm the safety and efficacy of interventional techniques, important limitations persist, including heterogeneous study populations, relatively small cohorts, and insufficient long-term reproductive follow-up. Greater standardization of clinical indications and procedural protocols, ongoing technological advancement, and well-designed multicenter trials are required to consolidate the evidence base. In gynecologic oncology, embolization techniques may provide effective hemorrhage control and facilitate surgical treatment in selected patients. In addition, uterine artery embolization has emerged as an important uterus-preserving adjunct in the management of cesarean scar pregnancy, although long-term reproductive (fertility) outcome data remain limited. IR is becoming an increasingly important component of modern gynecological care, supporting individualized, uterus-preserving, and less invasive therapeutic strategies; fertility preservation specifically is an important goal that, for several of the indications discussed, remains incompletely proven by long-term reproductive-outcome data. Full article

Background: Accurate breast lesion classification and localization from mammograms remain challenging because lesions vary in size, morphology, density, contrast, and boundary clarity. This study proposes MSF-Swin-ICFNet, an image-derived Multi-Scale Swin Transformer framework with Image-Derived Cross-Scale Fusion (ICF) for joint benign–malignant classification and lightweight lesion localization. Methods: The proposed framework extracts hierarchical representations from four Swin Transformer stages and aligns, adaptively recalibrates, and fuses these multi-scale features into a shared representation for classification and lesion-mask prediction. The model was evaluated on CBIS-DDSM, comprising 891 patients and 1592 mammograms, and RTM, comprising 573 patients and 10,070 mammograms. Patient-level stratified partitioning was used to ensure that mammograms and associated annotations from the same patient remained within a single data partition. Classification was evaluated using accuracy, precision, recall, F1-score, and AUC-ROC, while lesion localization was assessed using the Dice coefficient. Results: MSF-Swin-ICFNet achieved strong mammogram-level classification performance on the patient-disjoint test cohorts. On CBIS-DDSM, the model achieved 0.949 accuracy, 0.931 F1-score, and 0.985 AUC-ROC. On RTM, it achieved 0.968 accuracy, 0.956 F1-score, and 0.992 AUC-ROC. The corresponding Dice coefficients for lesion localization were 0.786 on CBIS-DDSM and 0.878 on RTM. Ablation experiments further supported the contribution of cross-scale fusion, scale recalibration, feature refinement, and the boundary-aware localization head. Conclusions: MSF-Swin-ICFNet demonstrates promising retrospective performance for image-based breast lesion classification with complementary lesion localization. The results support the technical potential of adaptive hierarchical feature fusion for mammographic analysis. However, the model is not intended as a standalone diagnostic system, and independent multicentre validation and radiologist reader studies are required before clinical deployment. Full article

Background/Objectives: Radiologists usually read a chest radiograph by comparing it with an earlier one, yet most deep learning models are trained and tested on single images. This study measures how much a longitudinal pair of radiographs improves finding-specific change detection on MIMIC-CXR and tests whether the gain can be explained by encoder pretraining. Methods: One frontal pair per patient was built from the earliest and latest studies within 180 days (14,043 pairs), and paired CheXpert labels defined four transition classes: absent-to-absent, onset, resolved, and persistent. A Siamese DenseNet-121 with two symmetric, weight-sharing branches fused the two images through concatenation and feature difference, and it was compared with a matched baseline that received only the final image. Evaluation used patient-grouped five-fold cross-validation over five seeds and three leakage-free encoder regimes. Results: On two matched binary tasks that hold the final image fixed—resolved versus absent (1→0 vs. 0→0) and onset versus persistent (0→1 vs. 1→1)—the single-image baseline performed at chance (AUROC 0.45 to 0.60), while the paired model scored well above it (paired difference +0.18 to +0.37 for resolution and +0.12 to +0.34 for onset; all significant by patient-level bootstrap after multiple-comparison correction), reaching a resolved-versus-absent AUROC of up to 0.87 (support devices). The benefit was stable across seeds, encoder regimes, projection changes, and follow-up intervals, and the CheXpert-derived transitions agreed with radiologist annotations (Cohen’s κ up to 0.97). Conclusions: A prior radiograph adds value whenever the final image alone underdetermines the transition, and they support pair-based designs for longitudinal change analysis. Full article

Background/Objectives: N-terminal pro-B-type natriuretic peptide (NT-proBNP) is an established heart failure biomarker, but its associations with cardiac magnetic resonance (CMR) parameters in patients with reduced ejection fraction remain unclear. This study examined the associations between serum NT-proBNP concentrations and CMR-derived functional, structural, and tissue characterization parameters in patients with HFrEF and evaluated which imaging parameters remained associated with NT-proBNP after multivariable adjustment. Methods: This retrospective single-center study included 92 consecutive patients with HFrEF who underwent CMR and NT-proBNP measurement within a 24 h window, with a median absolute interval of 6 h (range, 1–15 h) between the two assessments. CMR analysis included left ventricular ejection fraction (LVEF), left ventricular and left atrial (LA) cavity diameters, myocardial mass, extracellular volume fraction (ECV), the late gadolinium enhancement (LGE) burden, and the T2 signal intensity (T2-SI) ratio. Spearman correlation, univariate linear regression, and multivariable linear regression analyses were performed using log-transformed NT-proBNP values. Results: Log NT-proBNP demonstrated significant correlations with LVEF (r_s = −0.359, p < 0.001), T2-SI ratio (r_s = −0.332, p = 0.001), ECV (r_s = 0.383, p < 0.001), and LA diameter (r_s = 0.347, p < 0.001). No statistically significant correlations were observed between log-transformed NT-proBNP and LGE burden, LV cavity diameter, myocardial mass, LV wall thickness, age, or the recorded comorbidities. In multivariable analysis, LVEF, T2-SI ratio, and LA diameter remained associated with NT-proBNP after adjustment for the available covariates. Conclusions: Lower LVEF, a lower T2-SI ratio, and a larger LA diameter were associated with higher NT-proBNP levels in patients with HFrEF. The T2-SI ratio and LA diameter should be considered exploratory imaging markers requiring prospective validation. Full article

Background/Objectives: Compression syndromes of the stylohyoid complex (SHC) encompass neural, arterial, and venous presentations extending beyond classic Eagle syndrome. Symptomatic disease is increasingly recognised as a dynamic neurovascular conflict determined not by styloid process (SP) length alone, but by three-dimensional orientation, vessel and nerve proximity, and positional musculoskeletal interactions. This review synthesises that literature and proposes a phenotype-based framework integrating neural, arterial, venous, and mixed presentations. Methods: This is a narrative, pathophysiologically organised review rather than a systematic or scoping review. Landmark anatomical, clinical, and cohort studies were synthesised alongside systematic reviews, meta-analyses, osteological data, and selected case reports illustrating mechanisms not captured at cohort level. Evidence strength was interpreted by study design, distinguishing pooled estimates from retrospective series and proposed mechanisms. Results: The SHC and adjacent C1 transverse process form a narrow corridor containing the internal jugular vein, carotid arteries, and lower cranial nerves. SP elongation is common, but symptomatic compression depends more on orientation, SP-to-C1 distance, vessel proximity, and dynamic positional factors than on length. Three-dimensional CT angiography/venography is the anatomical cornerstone for defining morphology and vessel relationships, whereas catheter venography with manometry remains the haemodynamic reference standard in selected venous cases. Jugular narrowing is frequent in asymptomatic individuals and is not diagnostic without clinical and haemodynamic correlation. Surgical decompression appears favourable in selected patients, whereas primary endovascular stenting, evaluated only in single-centre observational series, carries a substantial complication profile and should generally be reserved for refractory or combined presentations. Conclusions: SHC compression syndromes are best understood as overlapping neural, arterial, venous, and mixed phenotypes within a dynamic neurovascular corridor. A phenotype-based classification and a cautious “styloidectomy-first” strategy for osseous venous compression are proposed as practical frameworks, but require prospective validation. Full article

Background: Lung surveillance in cystic fibrosis (CF) relies on chest radiography (CR) despite cumulative radiation. Low-field MRI may be a radiation-free alternative for children and young adults. Methods: We prospectively compared 0.55 T MRI and CR in 28 same-day examinations of 22 people with CF (mean age 13 ± 5 years). Three raters scored the disease with the MRI CF score and a modified Chrispin–Norman score, expanded from two to three zones per hemithorax to match the six MRI lobes. Analyses used per-examination values averaged across the raters (Wilcoxon tests, intraclass correlation coefficients, ICC). Results: MRI yielded higher global scores (median 8.2 vs. 6.7, p = 0.002) and higher centrilobular opacity scores (p = 0.005, adjusted p = 0.03). Air trapping was also higher on MRI (p = 0.025) but not after correction for multiple comparisons (adjusted p = 0.12). The other categories did not differ. Both correlated inversely with FEV1 (p < 0.001), without a significant modality-by-FEV1 interaction (p = 0.08). Interobserver agreement was higher for MRI (ICC = 0.93) than CR (0.82), difference 0.12 (95% bootstrap CI 0.06 to 0.18). The findings held in a per-participant sensitivity analysis. Conclusions: 0.55 T MRI is a feasible, radiation-free alternative to CR in CF, yielding higher modality-specific scores and more consistent interobserver agreement. Because the instruments were only category-mapped and no CT reference was available, higher scores do not establish superior lesion detection. Full article

Background/Objectives: The volume computed tomography (CT) dose index (CTDIvol) is a scanner output, not an organ dose, and cannot express how tube-current modulation varies along a patient. An organ-specific weighted CTDIvol addressing this has been reported before, in single-institution cohorts and often from inputs routine archives do not retain. New here is not the quantity but what an open, multi-vendor operationalisation reveals: whether its inputs survive archive curation and what the fallback costs when they do not. Methods: Forty abdominal CT series, ten per manufacturer, were drawn from the Cancer Imaging Archive and twelve organs segmented with TotalSegmentator at inference. Of 480 requested organ–series combinations, 455 were produced. A rule-based acquisition-constancy criterion admitted 39 series. Results: Modulation weights spanned 0.59 to 1.69, so the index departs from the whole-scan CTDIvol by up to 70% within one acquisition. A recorded CTDIvol survived in 29 of 40 archived headers and was reconstructable in 5 and unavailable in 6, availability differing markedly between manufacturers. Forcing that reconstruction on series that did retain a value agreed to within 12% on three scanner models and diverged by 58% and 84% on two others. Estimated organ mass was broadly consistent with International Commission on Radiological Protection (ICRP) Publication 89 for liver and kidneys. Conclusions: This index is not an absorbed dose; the implementation is open. Full article

Background/Objective: Radiomics offers a powerful, non-invasive approach for extracting quantitative features to predict lesion phenotypes. This study performs a radiomic characterization of breast tissue using synchrotron radiation breast computed tomography (SR-bCT). Method: Four mastectomy samples were imaged at the Australian Synchrotron (ANSTO) using five energies (25, 32, 35, 40, and 60 keV). From 10 slices per energy for each sample, 1546 regions of interest (ROIs) were extracted across four tissue subtypes: fatty, glandular, fibrous, and microcalcified. Using Pyradiomics, 93 features were initially calculated and then reduced to 34, eliminating highly correlated variables to reduce redundancies. Seven linear regression models and a discriminant analysis evaluated subtype tissue separation and radiomic characterization across individuals and combined samples and energies with a good fit (0.81 ≤ R ≤ 0.98). Results: The study identified six possible robust exploratory candidate features independent of sample variability and energy levels, whose mean values are significantly different among the four tissue subtypes (clusters, p < 0.001). The selected candidate features were 90th Percentile, Kurtosis, Skewness, GLCM IDM, GLDM LDE, and NGTDM Coarseness. These metrics successfully differentiated all tissue pairs (microcalcifications/fat, microcalcifications/gland, microcalcifications/fiber, fat/gland, fat/fiber and gland/fiber), with p < 0.05 inside the most general regression model and p < 0.0001 with the linear discriminant separation in clusters. Conclusions: Findings indicate that some first-order and second-order texture metrics reflecting global dependencies remain stable across experimental conditions. Conversely, fine-texture metrics are highly sensitive to sample energy changes, limiting their generalizability. These results align with successful biomarkers in mammography and validate the potential of radiomics in SR-bCT characterization. Full article

Background/Objectives: Sacroiliac joint (SIJ) anatomical variants and spinopelvic alignment have each been studied extensively, yet in separate studies, and whether the two are related remains unknown. Exploiting a single-acquisition CT protocol that captures the entire spinal axis together with the pelvis, we examined the association between SIJ anatomical variants and spinopelvic parameters. Methods: In this retrospective single-center study, 239 trauma patients who had undergone craniocervical-to-pelvic CT were analyzed. Each sacroiliac joint was scored separately for six SIJ variants (Prassopoulos classification), and at the patient level, a variant was considered present when identified on at least one side. Lumbosacral transitional vertebrae (LSTV) were also assessed, and pelvic incidence (PI), sacral slope (SS), pelvic tilt, and the spinosacral angle (SSA) were measured. Associations were tested by univariable comparison and multivariable logistic regression, and interobserver reliability was quantified with the intraclass correlation coefficient and Cohen’s kappa. Results: SIJ variants were common, occurring in 62.8% of left and 65.3% of right joints and in 68.2% of patients; these prevalences derive from a single-center trauma cohort and may not reflect the general population. The presence of any variant, considered collectively, was unrelated to the CT-derived spinopelvic parameters. When the six variants were examined individually, however, one of them—the iliosacral complex—was associated with higher SS, PI, and SSA (all p ≤ 0.016), although only the association with SSA persisted after correction for the false discovery rate. On multivariable analysis, female sex (adjusted OR 6.35), LSTV (adjusted OR 3.02), and the SSA (adjusted OR 1.05 per degree) were independently associated with the iliosacral complex. Interobserver agreement was excellent for the quantitative parameters and perfect for the iliosacral complex (κ = 1.000). Conclusions: Among the six variants examined, only the iliosacral complex was associated with sacropelvic morphology, being independently related to the spinosacral angle, female sex, and a lumbosacral transitional vertebra. Given the modest effect sizes and the retrospective, single-center, supine-CT design, the iliosacral complex is best regarded as an associated anatomical feature rather than a determinant of spinopelvic morphology, and prospective studies, ideally with upright imaging, are needed to establish whether this association has any functional consequence. Full article

Background/Objectives: Impacted mandibular third molars are associated with distal caries of the adjacent second molar and pericoronal follicle enlargement. This study evaluated the independent associations of both Winter angulation and Pell–Gregory classification with these complications on panoramic radiographs. Methods: This retrospective radiographic study evaluated 1491 impacted mandibular third molars from 1487 patients aged 18 years or older, reported in accordance with the STROBE guidelines for observational studies. Impaction patterns were classified using both the Winter and Pell–Gregory systems. Follicular space exceeding 2.5 mm was defined as follicle enlargement, measured digitally on panoramic images. Distal caries was defined as a radiolucent lesion on the distal surface of the adjacent second molar. Binary logistic regression (univariate and multiple models) was used to assess associations. Results: Among the 1487 patients (745 females, 742 males; mean age 25.21 ± 4.15 years), neither Winter angulation nor Pell–Gregory classification was associated with distal caries in univariate or multivariable analysis (all p > 0.05). Pericoronal follicle enlargement was significantly associated with deeper impaction: Pell–Gregory Class I carried lower risk than Class III (OR 0.296; 95% CI 0.184–0.478; p < 0.001), and Class A carried lower risk than Class C (OR 0.254; 95% CI 0.161–0.400; p < 0.001). Younger age independently predicted follicle enlargement (OR 0.964/year; p = 0.028). Winter angulation was not independently associated with follicle enlargement (all p > 0.05). Conclusions: Neither Winter nor Pell–Gregory classification predicted distal second molar caries on panoramic radiography. Deeper Pell–Gregory impaction positions and younger age were independently associated with pericoronal follicle enlargement, supporting early radiographic surveillance in young patients with deep impactions. Full article

Objectives: To assess the clinical viability of compressed sensing magnetization, we prepared a rapid acquisition gradient echo (CS-MPRAGE) for brain volume measurement by comparing its scan time and image quality with standard MPRAGE. Methods: In this retrospective study, we analyzed a total of 40 morphologically normal MRIs from relatively young subjects (mean age 27.03 ± 5.02, range 19~40, F:M = 33:7) who underwent both CS-MPRAGE and standard MPRAGE. MRI was performed with a 3T MR scanner using a 64-channel head coil. Volumetric data analysis was conducted using commercial AI-powered quantification software. The signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR_{gray matter/white matter}) were measured and compared. Qualitative analysis based on overall image quality, deep gray matter delineation, artifacts, and gray–white matter differentiation was rated on a five-point visual scoring system. Statistical analyses included paired t-tests, intraclass correlation coefficients (ICCs) and Pearson correlation. Results: CS-MPRAGE had a significantly shorter scan time than standard MPRAGE (1:46 vs. 5:20 min; p < 0.001). Whole-brain volume was similar (standard-MPRAGE, 1176.89 ± 105.36; CS-MPRAGE, 1181.14 ± 105.62 mL; r = 0.988; ICC, 0.9754–0.9931). Gray matter and regional lobe volumes were comparable, while the white matter volume was slightly larger with CS-MPRAGE. CNR of the gray–white matter was significantly higher with CS-MPRAGE (left, 30.45 ± 11.21 vs. 49.53 ± 32.71; right, 30.73 ± 11.97 vs. 46.89 ± 30.40; p = 0.0001) while SNR were similar across key brain regions. Conclusions: CS-MPRAGE provides high-quality 3D images and reliable volume data with significantly reduced acquisition time and comparable image quality. Full article

Background/Objectives: Beam-hardening artifacts from the skull base reduce image quality in routine non-contrast head CT, particularly in the posterior fossa. Virtual monoenergetic images (VMIs) reconstructed from dual-layer spectral CT have been shown to reduce these artifacts, but previous studies have generally been limited by small cohorts or evaluation of only a limited range of monoenergetic reconstructions. The aim of this study was to comprehensively evaluate image quality and posterior fossa artifact reduction across the monoenergetic spectrum in a large cohort of normal head CT examinations. Methods: Consecutive adult patients undergoing non-contrast head CT on a dual-layer spectral CT system were retrospectively included if no intracranial pathology was identified on clinical interpretation. Regions of interest were placed in predefined posterior fossa and supratentorial locations in conventional images and automatically propagated to VMIs reconstructed from 40 to 200 keV. Image noise, signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), attenuation, and a posterior fossa artifact index were quantified. Two neuroradiologists independently performed qualitative image quality assessment. Results: A total of 188 examinations were included in the quantitative analysis and 40 in the qualitative assessment. Compared with conventional images, VMIs reconstructed at ≥50 keV demonstrated significantly reduced image noise in the posterior fossa together with a higher SNR, while the CNR was improved between 40 and 80 keV. Qualitative assessment likewise demonstrated superior overall image quality and reduced artifact severity, with the highest ratings generally obtained for reconstructions around 60 keV. Improvements became marginal above approximately 80 keV. Conclusions: Virtual monoenergetic reconstructions from dual-layer spectral CT improve posterior fossa image quality by reducing beam-hardening artifacts while maintaining favorable image noise and tissue contrast. Intermediate-energy reconstructions (approximately 50–70 keV) provide the most favorable balance between beam-hardening artifact reduction, image noise, and tissue contrast, supporting their use as an image optimization strategy for non-contrast head CT. Whether these technical improvements translate into improved diagnostic performance should be evaluated in future studies including patients with posterior fossa pathology. Full article