Search Results (603)

Background: Irritable Bowel Syndrome (IBS) is a heterogeneous gastrointestinal disorder characterized by complex brain–gut interactions. Patient Similarity Networks (PSNs) offer a novel approach for exploring this heterogeneity and identifying clinically relevant patient subgroups. Methods: We analyzed data from 78 participants (49 IBS patients and 29 healthy controls) with 36 brain morphometric measures (FreeSurfer v7.4.1) and 6 measures of cognitive functions (5 RBANS domain indices plus a Total Scale score). PSNs were constructed using multiple similarity measures (Euclidean, cosine, correlation-based) with Gaussian kernel transformation. We performed community detection (Louvain algorithm), centrality analyses, feature importance analysis, and correlations with symptom severity. Statistical validation included bootstrap confidence intervals and permutation testing. Results: The PSN comprised 78 nodes connected by 469 edges, with four communities detected. These communities did not significantly correspond to diagnostic groups (Adjusted Rand Index = 0.011, permutation $p=0.212$), indicating IBS patients and healthy controls were intermixed. However, each community exhibited distinct neurobiological profiles: Community 1 (oldest, preserved cognition) showed elevated intracranial volume but reduced subcortical gray matter; Community 2 (youngest, most severe IBS symptoms) had elevated cortical volumes but reduced white matter; Community 3 (most balanced IBS/HC ratio, mildest IBS symptoms) showed the largest subcortical volumes; Community 4 (lowest cognitive performance across multiple domains) displayed the lowest RBANS scores alongside high IBS prevalence. Top network features included subcortical structures, corpus callosum, and cognitive indices (Language, Attention). Conclusions: PSN identifies brain–cognition communities that cut across diagnostic categories, with distinct feature profiles suggesting different hypothesis-generating neurobiological patterns within IBS that may inform personalized treatment strategies. Full article

Background/Objectives: Early detection of hepatocellular carcinoma (HCC), particularly small lesions (<2 cm), which is crucial for curative treatment, remains challenging with conventional liver dynamic computed tomography (LDCT). We aimed to develop a deep learning algorithm to generate synthetic CT during hepatic arteriography (CTHA) from non-invasive LDCT and evaluate its lesion detection performance. Methods: A cycle-consistent generative adversarial network with an attention module [Unsupervised Generative Attentional Networks with Adaptive Layer-Instance Normalization (U-GAT-IT)] was trained using paired LDCT and CTHA images from 277 patients. The model was validated using internal (68 patients, 139 lesions) and external sets from two independent centers (87 patients, 117 lesions). Two radiologists assessed detection performance using a 5-point scale and the detection rate. Results: Synthetic CTHA significantly improved the detection of sub-centimeter (<1 cm) HCCs compared with LDCT in the internal set (69.6% vs. 47.8%, p < 0.05). This improvement was robust in the external set; synthetic CTHA detected a greater number of small lesions than LDCT. Quantitative metrics (structural similarity index measure and peak signal-to-noise ratio) indicated high structural fidelity. Conclusions: Deep-learning–based synthetic CTHA significantly enhanced the detection of small HCCs compared with standard LDCT, offering a non-invasive alternative with high detection sensitivity, which was validated across multicentric data. Full article

Background and Clinical Significance: Giant sphenoid wing meningiomas are generally viewed as skull base masses that compress frontal centers and their respective pathways gradually enough to cause a dysexecutive–apathetic syndrome, which can mimic primary neurodegenerative disease. The aim of this report is to illustrate how bedside phenotyping and multimodal imaging can disclose similar clinical presentations as surgically treatable network lesions. Case Presentation: An independent, right-handed older female developed an incremental, two-year decline of her ability to perform executive functions, extreme apathy, lack of instrumental functioning, and a frontal-based gait disturbance, culminating in a first generalized seizure and a newly acquired left-sided upper extremity pyramidal sign. Standardized neuropsychological evaluation revealed a predominant frontal-based dysexecutive profile with intact core language skills, similar to behavioral-variant frontotemporal dementia (bvFTD). MRI demonstrated a large, right fronto-temporo-basal extra-axial tumor attached to the sphenoid wing with homogeneous postcontrast enhancement, significant vasogenic edema within the frontal projection pathways, and a marked midline displacement of structures with an open venous pathway. With the use of a skull-base flattening pterional craniotomy with early devascularization followed by staged internal debulking, arachnoid preserving dissection, and conservative venous preservation, the surgeon accomplished a Simpson Grade I resection. Sequential improvements in the patient’s frontal “re-awakening” were demonstrated through postoperative improvements on standardized stroke, cognitive and functional assessment scales that correlated well with persistent decompression and symmetric ventricles on follow-up images. Conclusions: This case illustrates the possibility of a non-dominant sphenoid wing meningioma resulting in a pseudo-degenerative frontal syndrome and its potential for reversal if recognized as a network lesion and treated with tailored, venous-sparing skull-base surgery. Contrast-enhanced imaging and routine frontal testing in atypical “dementia” presentations may aid in identifying additional patients with potentially surgically remediable cases. Full article

Background: Mucopolysaccharidosis type IIIC (MPS IIIC) is a rare lysosomal storage disorder caused by pathogenic variants in the HGSNAT gene. Data from large patient cohorts remain scarce, particularly in Latin America. Methods: We retrospectively analyzed clinical, biochemical, and genetic data from patients diagnosed with MPS IIIC through the MPS Brazil Network. Diagnosis was based on reduced activity of acetyl-CoA:α-glucosaminide N-acetyltransferase (HGSNAT), elevated urinary glycosaminoglycans (uGAGs), and/or molecular genetics tests. Results: A total of 101 patients were confirmed with MPS IIIC, representing one of the largest cohorts worldwide. Females accounted for 60% of cases. The mean age at symptom onset was 5.4 ± 3.9 years, while the mean age at diagnosis was 11.7 ± 6.9 years, reflecting a 6-year diagnostic delay. Most patients initially presented with developmental delay (82%) and facial dysmorphism (80%), whereas behavioral manifestations were less frequently identified (25%), suggesting a milder phenotype than previously reported. Genetic information was available for 28% of patients, showing recurrent alleles (c.372-2A>G, c.252dupT) and several novel mutations, which expand the mutational spectrum of the disease. Genotype–phenotype similarities with Portuguese, Italian, and Chinese cases suggest shared ancestry contributions. Regional differences included earlier diagnoses in the North of Brazil and high consanguinity rates in the Northeast region. Conclusions: This study describes the largest Brazilian cohort of MPS IIIC, documenting novel variants and regional heterogeneity. Findings highlight diagnostic delays, ancestry influences, and the urgent need for disease-modifying therapies. Full article

Background/Objectives: Uveal melanoma (UVM), the leading primary intraocular cancer in adults, is driven by GNAQ/GNA11 mutations, encoding the active forms of Gαq proteins. While local treatments like surgery or radiation can control primary tumors, nearly half of patients die from metastasis. Our aim was identifying potential pathways involved in resistance to targeted therapy in UVM. Methods: Here, we screened 100 pathway-activating mutant complementary DNAs by lentiviral overexpression to identify those that enhance the survival of cancer cells in the presence of clinically relevant targeted therapies, using BAP1 wild-type UVM cells and validated the most significant results in BAP1-mutant cells. Results: This revealed JAK/STAT activation, overexpression of anti-apoptotic BCL2/BCL-XL, and dysregulated PI3K/mTOR or Hippo pathways as escape routes under MEK-ERK or FAK inhibition. Bioinformatic analysis of UVM transcriptome in TCGA further showed that high expression of the hallmark PI3K/AKT/mTOR pathway and IL6/JAK/STAT signaling correlates with poor prognosis. A similar correlation was shown by YAP and anti-apoptotic signatures. The analysis of individual representative genes from these signatures revealed that MTOR, BCL2L1 (BCL-XL), and TEAD4 gene expression are linked to poorer survival, underscoring the potential clinical impact of these adaptive pathways. Proliferation and apoptosis assay demonstrated that aberrant activation of AKT and YAP promotes resistance to FAK and MEK inhibitors. Conclusions: These findings support the adaptability of UVM lesions and suggest rational combination therapies targeting both primary GNAQ/GNA11-driven oncogenic signals and their compensatory networks as a more effective, personalized treatment approach for advanced UVM. Full article

Background: Myelodysplastic syndromes (MDSs) and acute myeloid leukemia (AML) are clonal hematological disorders that share molecular origins but present with distinct clinical features. MicroRNAs (miRNAs) are key post-transcriptional regulators, and their altered expression may reflect biological shifts contributing to disease progression. Methods: Expression levels of miR-155, miR-181, miR-221, miR-222, and miR-223 were analyzed by RT-qPCR in bone marrow samples from 37 MDS patients, 20 AML patients, and 7 controls. Group comparisons were performed using ANOVA (with Benjamini–Hochberg correction) and Tukey post hoc testing. Diagnostic performance and network behavior were evaluated using ROC analysis, Pearson correlation matrices, and principal component analysis (PCA). Results: miR-155, miR-181, and miR-223 were upregulated in AML, whereas miR-221 and miR-222 were downregulated. miR-222 showed the highest diagnostic accuracy (AUC ~0.87 for both AML vs. control and MDS vs. control). Its expression was significantly higher in high IPSS-R MDS cases (p = 0.046), with a similar upward tendency for miR-221 (p = 0.054). Progressive loss of coordinated miRNA expression was observed from controls to MDS and AML. PCA supported these findings by showing separation mainly driven by miR-222 and miR-155. Conclusions: Combined miRNA profiling highlights miR-222 and, to a lesser extent miR-155, as consistent indicators of myeloid disease transformation. While further validation in larger and genetically stratified cohorts is warranted, these findings support the potential contribution of miRNA signatures to diagnostic evaluation and risk stratification in MDS and AML, in line with precision hematology approaches. Full article

Background and Objectives: Anxiety is prevalent in patients with major depressive disorder (MDD) and bipolar disorder (BD). Understanding its interaction with mood disorders may provide deeper insight into symptom clustering, severity, and interventions. We compared the networks of MDD and BD using the Beck Anxiety Inventory (BAI) to identify central symptoms and interconnections. Materials and Methods: This cross-sectional study involved 815 individuals with MDD (n = 332) and BD (n = 483) who had clinically significant anxiety symptoms (BAI score > 8). Network analysis identified anxiety symptom clusters. Network centrality, stability, and comparison tests assessed the structural differences and global strength variations between the groups. Results: Both the MDD and BD networks showed strong interconnections among several BAI items and demonstrated stable centrality measures. Core symptoms with high centrality included “Losing control”, “Choking”, “Breathing”, “Unsteady”, and “Shaky” in both MDD and BD. Although no significant differences were found in the overall network structures between MDD and BD, the global strength of the network differed significantly, with MDD exhibiting modestly higher overall anxiety network connectivity than BD. Conclusions: Network clusters revealed aspects of both cognitive and somatic symptoms of anxiety. Although overall structures were similar between the groups, the MDD group showed stronger interconnections for central anxiety symptoms. Targeting central anxiety symptoms can enhance prevention and intervention strategies for mood disorders and improve clinical outcomes. Full article

Acellular dermal matrices (ADMs) are widely used in soft-tissue reconstruction, yet the optimal timing for split-thickness skin grafting (STSG) remains unsettled. We conducted a single-center retrospective cohort study (January 2023–August 2025) of adults undergoing ADM-based reconstruction with Integra^® Double Layer (IDL), Integra^® Single Layer (ISL), or Nevelia^®. Primary endpoints included length of stay (LOS), STSG requirement and timing, and in-hospital complications; secondary endpoints included spontaneous epithelialization. Prespecified adjusted analyses (linear/logistic models) controlled for age, sex, etiology, anatomical site, diabetes/PAOD, smoking, wound size (when available), wound contamination, and matrix type. Histology and immunohistochemistry (H&E, Masson trichrome, CD105, D2-40) assessed matrix integration and vascular/lymphatic maturation. Seventy-five patients were included (IDL n = 40; ISL n = 20; Nevelia n = 15). On multivariable analysis, matrix type was not an independent predictor of LOS (ISL vs. IDL β = +2.84 days, 95% CI −17.34 to +23.02; Nevelia vs. IDL β = −4.49 days, 95% CI −16.24 to +7.26). Complications were infrequent (6/75, 8.0%) and comparable across matrices; spontaneous epithelialization occurred in 3/75 patients (4.0%). A day-14 grafting strategy, applied only after documented clinical integration, was feasible in 30/75 (40.0%) patients without excess complications. Histology/IHC at 3–4 weeks demonstrated CD105-positive, perfused capillary networks with abundant collagen; at 4–6 weeks, D2-40-positive lymphatic structures confirmed progressive neo-dermis maturation, supporting the biological plausibility of earlier grafting once integration criteria are met. In this cohort, outcomes were broadly similar across matrices after adjustment. A criteria-based early STSG approach (~day 14) appears safe and operationally advantageous when integration is confirmed, while a minority of defects may heal without grafting. Prospective multicenter studies with standardized scar/functional measures and cost analyses are needed to refine patient selection and graft timing strategies. Full article

Background: Accurate quantification of visceral (VAT) and subcutaneous adipose tissue (SAT) is critical for understanding the cardiometabolic consequences of obstructive sleep apnea (OSA) and other chronic diseases. This study validates a customization framework using pre-trained networks for the development of automated VAT/SAT segmentation models using hybrid positron emission tomography (PET)/magnetic resonance imaging (MRI) data from OSA patients. While the widespread adoption of deep learning models continues to accelerate the automation of repetitive tasks, establishing a customization framework is essential for developing models tailored to specific research questions. Methods: A UNet-ResNet50 model, pre-trained on RadImageNet, was iteratively trained on 59, 157, and 328 annotated scans within a closed-loop system on the Discovery Viewer platform. Model performance was evaluated against manual expert annotations in 10 independent test cases (with 80–100 MR slices per scan) using Dice similarity coefficients, segmentation time, intraclass correlation coefficients (ICC) for volumetric and metabolic agreement (VAT/SAT volume and standardized uptake values [SUVmean]), and Bland–Altman analysis to evaluate the bias. Results: The proposed deep learning pipeline substantially improved segmentation efficiency. Average annotation time per scan was 121.8 min (manual segmentation), 31.8 min (AI-assisted segmentation), and only 1.2 min (fully automated AI segmentation). Segmentation performance, assessed on 10 independent scans, demonstrated high Dice similarity coefficients for masks (0.98 for VAT and SAT), though lower for contours/boundary delineation (0.43 and 0.54). Agreement between AI-derived and manual volumetric and metabolic VAT/SAT measures was excellent, with all ICCs exceeding 0.98 for the best model and with minimal bias. Conclusions: This scalable and accurate pipeline enables efficient abdominal fat quantification using hybrid PET/MRI for simultaneous volumetric and metabolic fat analysis. Our framework streamlines research workflows and supports clinical studies in obesity, OSA, and cardiometabolic diseases through multi-modal imaging integration and AI-based segmentation. This facilitates the quantification of depot-specific adipose metrics that may strongly influence clinical outcomes. Full article

Background/Objectives: Constructing a comprehensive spatiotemporal map of tumor heterogeneity is essential for understanding tumor evolution, with copy number variation (CNV) being a significant feature. Existing studies often rely on tools originally developed for single-cell data, which fail to utilize spatial information, often leading to an incomplete map of clonal architecture. Our study aims to develop a model that fully leverages spatial omics data to elucidate spatio-temporal changes in tumor evolution. Methods: Here, we introduce SCOIGET (Spatial COpy number Inference by Graph on Evolution of Tumor), a novel framework using graph neural networks with graph attention layers to learn spatial neighborhood features of gene expression and infer copy number variations. This approach integrates spatial multi-omics features to create a comprehensive spatial map of tumor heterogeneity. Results: Notably, SCOIGET achieves a substantial reduction in error metrics (e.g., mean squared error, cosine similarity, and distance measures) and produces superior clustering performance, as indicated by higher Silhouette Scores compared to existing methods, validated by both simulated data with spot-level ground truth and patient cohorts. Our model significantly enhances the accuracy of tumor evolution depiction, capturing detailed spatial and temporal changes within the tumor microenvironment. It is versatile and applicable to various downstream tasks, demonstrating strong generalizability across different spatial omics platforms, including 10× Visium and Visium HD and various cancer types, including colorectal cancer and prostate cancer. This robust performance improves research efficiency and provides valuable insights into tumor progression. Conclusions: SCOIGET offers an innovative solution by integrating multiple features and advanced algorithms, providing a detailed and accurate representation of tumor heterogeneity and evolution, aiding in the development of personalized cancer treatment strategies. Full article

Background/Objectives: Breast cancer is one of the most common malignancies in women worldwide. Early diagnosis and accurate classification in breast cancer detection are among the most critical factors determining treatment success and patient survival. In this study, a deep learning-based model was developed that can classify benign, malignant, and normal breast tissues from ultrasound images with high accuracy and achieve better results than the methods commonly used in the literature. Methods: The proposed model was trained on a dataset of breast ultrasound images, and its classification performance was evaluated. The model is designed to effectively learn both local textural features and global contextual relationships by combining Squeeze-and-Excitation (SE) blocks, which emphasize channel-level feature importance, and Convolutional Block Attention Module (CBAM) attention mechanisms, which focus on spatial information, with the MHA structure. The model’s performance is compared with three commonly used convolutional neural networks (CNNs) and three Vision Transformer (ViT) architectures. Results: The developed model achieved an accuracy rate of 96.03% in experimental analyses, outperforming both the six compared models and similar studies in the literature. Additionally, the proposed model was tested on a second dataset consisting of histopathological images and achieved an average accuracy of 99.55%. The results demonstrate that the model can effectively learn meaningful spatial and contextual information from ultrasound data and distinguish different tissue types with high accuracy. Conclusions: This study demonstrates the potential of deep learning-based approaches in breast ultrasound-based computer-aided diagnostic systems, providing a reliable, fast, and accurate decision support tool for early diagnosis. The results obtained with the proposed model suggest that it can significantly contribute to patient management by improving diagnostic accuracy in clinical applications. Full article

Background/Objectives: Canadian Advanced Physiotherapist Practitioner (APP) roles have existed for over 25 years in pediatric rheumatology. The APP can manage many common pediatric rheumatic conditions most often in Shared-Care Models (SCMs) with pediatric rheumatologists (PRs). The quality of care children receive in an APP SCM compared to traditional care is unknown. The Pediatric Rheumatology Care and Outcomes Improvement Network (PR-COIN) tracks quality measures as Key Performance Indicators (KPIs) in juvenile idiopathic arthritis (JIA) care. This study aimed to analyze the frequency of KPIs documented in a pediatric rheumatology APP SCM from a single center and compare to PR-COIN’s performance targets to assess the quality and safety of care. Methods: A retrospective chart review of JIA cases managed in a pediatric rheumatology APP SCM over a 2-year period was conducted. KPIs for disease activity, safety monitoring and access to care were evaluated. Frequency of KPI documentation by the APP were compared to target performance goals (≥40, ≥70 or ≥80% documentation rate depending on KPI) and with PR-COIN data from the Same Center (SC) (three rheumatologists) and PR-COIN (15 centers). Results: Documented KPIs were compared between the APP SCM, SC and PR-COIN registry (138; 140; 11,431 eligible visits, respectively) between June 2022–May 2024. Demographics were similar between groups. Increased percentages of patients with polyarticular rheumatoid factor positive and psoriatic subtypes were seen by APP compared to SC and PR-COIN. Documentation frequency of all disease activity and safety monitoring KPI performance goals were either higher in the APP SCM or comparable to SC and PR-COIN. Conclusions: The pediatric rheumatology APP SCM exceeded PR-COIN performance goals for KPI documentation, establishing a high level of quality and safety of care for children with JIA when managed in this model of care. Next steps include replicating this study in other pediatric rheumatology centers with an APP SCM. Full article

Background: Precise delineation of the rectum is crucial in treatment planning for cancers in the pelvic region, such as prostate and cervical cancers. Manual segmentation is also still time-consuming and suffers from inter-observer variability. Since there are meaningful differences in rectal anatomy between males and females, incorporating sex-specific anatomical patterns can be used to enhance the performance of segmentations. Furthermore, recent deep learning advancements have provided promising solutions for automatically classifying patient sex from CT scans and leveraging this information for enhancing the accuracy of rectal segmentation. However, their clinical utility requires comprehensive validation against real-world standards. Methods: In this study, a two-stage deep learning pipeline was developed using CT scans from 186 patients with either prostate or cervical cancer. First, a CNN model automatically classified the patient’s biological sex from CT images in order to capture anatomical variations dependent on sex. Second, a sex-aware U-Net model performed automated rectal segmentation, allowing the network to adjust its feature representation based on the anatomical differences identified in stage one. The internal validation had an 80/20 train–test split, and 15% of the training portion was held out for validation to ensure balanced distribution regarding sex and diagnosis. Model performance was evaluated using spatial similarity metrics, including the Dice Similarity Coefficient (DSC), Hausdorff Distance, and Average Surface Distance. Additionally, a radiation oncologist conducted a retrospective clinical evaluation using a 3-point Likert scale. Statistical significance was examined using Wilcoxon signed-rank tests, Welch’s t-tests, and Mann–Whitney U test. Results: The sex-classification model attained an accuracy of 94.6% (AUC = 0.98, 95% CI: 0.96–0.99). Incorporation of predicted sex into the segmentation pipeline improved anatomical consistency of U-Net outputs. Mean DSC values were 0.91 (95% CI: 0.89–0.92) for prostate cases and 0.89 (95% CI: 0.87–0.91) for cervical cases, with no significant difference between groups (p = 0.12). Surface distance metrics calculated on resampled isotropic voxels showed mean HD values of 3.4 ± 0.8 mm and ASD of 1.2 ± 0.3 mm, consistent with clinically acceptable accuracy. On clinical evaluation, 89.2% of contours were rated as excellent, while 9.1% required only minor adjustments. Automated segmentation reduced the average contouring time from 12.7 ± 2.3 min manually to 4.3 ± 0.9 min. Conclusions: The proposed sex-aware deep learning framework offers accurate, robust segmentation of the rectum in pelvic CT imaging by explicitly modeling sex-specific differences in anatomical characteristics. This physiologically informed approach enhances segmentation performance and supports reliable integration of AI-based delineation into radiotherapy workflows to improve both contouring efficiency and clinical consistency. Full article

Background/Objectives: We measured hepatic T2*, T1, and T2 values in N = 81 transfusion-dependent thalassemia (TDT) patients to assess and compare their reproducibility, evaluate their correlations with demographics and clinical parameters, and explore their association with disease-related complications. Methods: All TDT patients (52 females, 38.13 ± 10.79 years), were enrolled in the Extension-Myocardial Iron Overload in Thalassaemia Network. The magnetic resonance imaging protocol (1.5 T) included: multi-echo gradient echo sequences for T2* relaxometry, modified look-locker inversion recovery (MOLLI) sequences for T1 mapping, and multi-echo fast-spin-echo (MEFSE) sequences for T2 mapping. Results: All three relaxation times demonstrated good intra- and inter-observer reproducibility and were significantly correlated with each other. Of the 59 patients with reduced T2*, 45 (76.3%) also had reduced T1, and 42 (71.2%) had reduced T2 values. Among 22 patients with normal T2*, 3 (13.6%) exhibited reduced T1. No patients showed increased T1, and only one had elevated T2. Liver relaxation times were not associated with gender or splenectomy status. All relaxation times inversely correlated with serum ferritin levels, while T2 and T2* inversely correlated with mean alanine aminotransferase levels. Cirrhosis and glucose metabolism alterations were associated with lower relaxation times. All three relaxation times effectively discriminated between the absence and presence of cirrhosis [areas under the curve (AUCs) with 95% confidence intervals (CIs): 0.85 (0.75–0.92) for T2*, 0.78 (0.68–0.87) for T1, and 0.92 (0.84–0.97) for T2]. T2* showed comparable accuracy to T1 and T2, while a significant difference was observed between T1 and T2 values. All liver relaxation times demonstrated similar diagnostic performance in identifying glucose metabolism alterations [AUCs with 95% CIs: 0.67 (0.55–0.77) for T2*, 0.69 (0.57–0.79) for T1, and 0.67 (0.56–0.77) for T2]. Conclusions: In TDT, a comprehensive assessment of hepatic relaxation times may enhance clinical monitoring and management of iron overload and its related complications. Full article

Objectives: This review aims to review the application of deep learning (DL) techniques in the imaging analysis of abdominal aortic aneurysm (AAA), with a specific focus on the segmentation of intraluminal thrombus (ILT). Methods: A comprehensive literature review was conducted through searches of PUBMED and Web of Science up to September 2025. Only English-language studies applying DL-based networks for ILT segmentation in patients with AAA on computed tomography angiography were included. After screening 664 articles, 22 met the eligibility criteria and were included. The reported methodological frameworks and segmentation performance metrics were extracted for comparison and analysis. Results: Among the studies included, the reported Dice similarity coefficients ranged from 0.81 to 0.93 for 2D networks and from 0.804 to 0.9868 for 3D networks. Notably, 2D Multiview fusion models outperform other 2D approaches, while 3D U-Net remains a strong baseline. Methods using preoperative images demonstrated great applicability for surgical planning, while postoperative segmentation faced challenges related to imaging artifacts caused by stent. Conclusions: This review provides a comprehensive overview of recent DL-based ILT segmentation methods for AAA patients on CTA, offering perspectives for applications in advanced preoperative planning and postoperative surveillance. Despite the promising results, the lack of standardized datasets limits model development and external validation. Future research should address these limitations by focusing on multicenter standardized datasets and seamless integration into clinical workflows. Full article