Ultrasound-induced cavitation is conventionally described through nonlinear bubble dynamics, acoustic pressure modulation, microbubble oscillation or collapse, local mechanical stress, thermal or chemical activation, and subsequent biological effects. In medical contexts, such cavitation-mediated processes are increasingly relevant to sonoporation, microbubble-enhanced drug delivery, sonodynamic therapy, and histotripsy. However, a compact phenomenological framework linking measurable cavitation dynamics to delayed, channel-specific biological outputs remains useful, particularly when different endpoints such as membrane permeabilization, reactive oxygen species generation, molecular uptake, and tissue fragmentation are considered together. In this work, a phenomenological relational-informational bridge model is proposed, in which therapeutic cavitation is interpreted as biological constraint focusing. The cavitation region and its adjacent biological microenvironment are represented as a localized, acoustically driven subsystem whose effective constraint state changes during bubble or microbubble oscillation and collapse. Bubble oscillation or collapse is represented as a rapid increase in constraint loading and informational action density, whereas medically relevant effects are modeled as relaxation of a transient high-tension state into bioactive output channels, including membrane permeabilization, reactive oxygen species generation, molecular delivery, and mechanical tissue fragmentation. The model couples the bubble or microbubble radius R(t) and collapse or oscillation velocity R˙(t), obtained experimentally or from Rayleigh-Plesset-type dynamics, to a dimensionless relational constraint parameter λ(t), an informational action density Srel(t), a stored high-tension reservoir Erel(t), channel-specific motif populations Nk(t), and measurable biological outputs Bk(t). The construction is not intended to replace hydrodynamic, thermodynamic, sonochemical, or biomechanical models; rather, it provides a latent-variable layer that may organize how cavitation loading is converted into endpoint-specific biological responses. The framework yields testable expectations: biological response should correlate not only with acoustic pressure or minimum bubble radius, but also with the rate of constraint loading, reservoir buildup and depletion, relaxation-channel kinetics, and modifiers such as microbubble composition, tissue context, oxygenation, sonosensitizer availability, and molecular cargo. Ultrasound-mediated cavitation is therefore reframed as a bioengineering process in which acoustic exposure, bubble dynamics, transient energy localization, and biological endpoint formation are connected through a testable phenomenological bridge model.
Sergio Di Nitto was not included as an author in the original publication [...].
There was an error in the original publication [...].
Epicutaneous patch testing is the gold standard for diagnosing allergic contact dermatitis (ACD), yet its interpretation relies on subjective scoring and remains prone to inter-observer variability. In this study, we present a machine learning pipeline that complements subjective scoring with objective bioengineering measurements derived from the Antera 3D imaging system. A dataset of skin reactions was analyzed, with particular attention to how the data was split to avoid information leakage between patients. For this reason, methods such as GroupShuffleSplit and GroupKFold, which account for patient-level clustering, were used. Of the models tested, the Random Forest classifier showed the best overall performance, with an AUC of 0.861 (95% CI: 0.830-0.888) on patient data that had not been used during training, outperforming the Multi-Layer Perceptron model. The incorporation of additional features that capture changes between 48 and 72 h improved the results even further, raising the AUC to 0.902 and achieving a very high sensitivity of 96.8%. Overall, the results show that objective biophysical measurements derived from the Antera 3D imaging system can be combined with machine-learning techniques for objective patch-test assessment. Incorporating temporal changes between the 48- and 72-h readings further improved model performance, suggesting that temporal changes provide additional information beyond single-time-point measurements.
This study evaluates the mechanobiological responses of MC3T3-E1 cells to fluid shear stress utilizing a coupled CFD-CPM mesoscale framework. Computational fluid dynamics was utilized to calculate the distribution of fluid shear stress within the culture chamber, which was subsequently mapped onto a discrete system of lattices. The cellular Potts model was employed to simulate behaviors of the cells governed by rules for proliferation, migration, contact inhibition, and osteogenic differentiation. To accurately reflect developmental stages, the computational workflow dictated that the cells complete the phase of growth prior to the initiation of differentiation. Evaluations demonstrated that the culture region formed a relatively uniform plateau of shear stress. Within an optimal range, fluid shear stress accelerates the transition of these cells into mature osteoblasts. Furthermore, staining for alkaline phosphatase revealed responses of osteogenic differentiation strictly correlated with the local distribution of fluid shear stress. Ultimately, this study establishes a visualized framework of mesoscale modeling to analyze the collective behavior of osteoblasts under mechanical stimulation in microfluidic environments, demonstrating the feasibility of predicting subsequent extracellular matrix mineralization and providing valuable insights into the dynamic evolution of bone remodeling.
Ventral incisional hernia is a frequent postoperative complication following abdominal surgery, with recurrence largely driven by biomechanical factors such as elevated intra-abdominal pressure and stress concentration at the suture-mesh-tissue interface. This study presents a patient-specific finite element analysis (FEA) framework to investigate stress transfer mechanisms within ventral incisional hernia repair systems under physiologically relevant loading conditions. A patient-specific abdominal wall geometry was reconstructed from computed tomography images and modeled as a nonlinear hyper-elastic material using a second-order Ogden formulation. Polypropylene surgical meshes and sutures were represented using finite elements, and an intra-lay mesh placement with a midline incision was simulated. Loading conditions corresponding to regular breathing, forceful breathing, and heavy-load lifting were applied. The results reveal that stress concentrations consistently localize at the sutures, with stress magnitudes increasing markedly under higher physiological loads. Under heavy lifting, suture stresses approached the material yield limit, whereas mesh stresses remained comparatively low. These findings identify suture failure as a critical mechanical factor contributing to hernia recurrence and highlight the importance of postoperative load management and improved support strategies. This work provides a basis for more comprehensive future patient-specific analyses.
Osteochondral defects, involving both articular cartilage and subchondral bone, can lead to joint degeneration and osteoarthritis. Recent advances in 3D-printed biphasic scaffolds offer promising opportunities to recreate physiological microenvironments for tissue regeneration. In tissue engineering, these scaffolds can be mechanically stimulated to promote targeted cartilage and bone formation. While computational models have been widely used to study mechanically induced cellular responses in monophasic scaffolds, time-dependent modelling of biphasic osteochondral systems remains relatively scarce. In this study, a fluid-structure interaction (FSI) framework coupled with a mechanoregulatory algorithm was developed to predict mechanically induced early-stage mesenchymal stem cell (MSC) differentiation in biphasic open-porous osteochondral scaffolds comprising chondral and bone layers designed for direct ink writing (DIW). In a second model, an interfacial barrier layer representing the native osteochondral interface was integrated. Dynamic compressive loading (1 Hz, 2.5% strain) was applied. The simulations predicted region-specific differentiation patterns in both the chondral and subchondral bone regions. In the scaffold without a barrier layer, approximately 68.9% of MSCs in the chondral layer and 93.4% of MSCs in the bone layer underwent chondrogenic and osteogenic differentiation, respectively. Incorporation of the barrier layer caused only minor changes, reducing predicted cartilage and bone differentiation by approximately 1.5% and 3.9%, respectively. Overall, this study highlights the capability of computational modelling to predict mechanobiological responses in complex osteochondral systems and support scaffold design and effective mechanical stimulation protocols.
Background/Objectives: Although insulin therapy has been fundamental in the management of diabetes mellitus since its discovery, limitations associated with conventional administration routes continue to drive the development of alternative delivery strategies. This study reports the design, synthesis, and characterization of a chitosan-alginate nanoparticulate system for the encapsulation and pH-responsive release behavior of recombinant human insulin, developed via ionic crosslinking with sodium tripolyphosphate (TPP). Methods: Nanoparticles were prepared by ionic crosslinking. Physicochemical characterization was carried out by UV-vis spectroscopy, Fourier-transform infrared spectroscopy with attenuated total reflectance (FTIR-ATR), dynamic light scattering (DLS), zeta potential analysis, fluorescence spectroscopy, and scanning electron microscopy (SEM). In vitro release studies were conducted at pH 4.5 and 7.4 to simulate physiological environments. Results: The nanoparticles achieved an encapsulation efficiency (EE%) of 30 ± 2.6% and a zeta potential of 41 ± 1.7 mV (pH 4.0). SEM analysis revealed mostly elongated nanoparticles, while DLS measurements confirmed nanometric sizes. The release profile demonstrated rapid insulin release at acidic pH (4.5) and sustained release at slightly basic pH (7.4). After three months of storage at room temperature, the lyophilized nanoparticles allowed continued release at pH 7.4. Conclusions: The synthesized chitosan-alginate nanoparticles provide a biocompatible platform for the encapsulation and release of recombinant human insulin. These findings can contribute to the development of nanosystems as an alternative for insulin delivery, thereby improving therapeutic adherence and accessibility.
Convergence insufficiency (CI) is a common binocular vision disorder that causes eye strain, headaches, and blurriness. Although pencil push-ups (PPU) are widely used to treat CI, their real-world effectiveness may be limited by inconsistent performance and poor adherence. MobileS (the App) is an AI-based smartphone application that tracks eye movements through the front camera, enabling guided "digital push-ups" with instant visual feedback and automatic performance monitoring. This follow-up study evaluated the feasibility, recorded adherence, and preliminary clinical signals of MobileS as a home-based therapy for CI compared with conventional PPU. Twenty-five participants with CI were randomly assigned to either the App group (n = 12) or the PPU group (n = 13) for 8 weeks of home-based training. Both groups performed convergence push-ups-using the app or a pencil. NPC was measured using a RAF ruler at baseline, mid-point, and endpoint. Symptom severity was assessed using the Convergence Insufficiency Symptom Survey (CISS), and adherence was recorded automatically by the app or manually by participants. All 25 randomized participants completed all three NPC assessment points. The App group showed a statistically significant improvement in NPC from baseline to study completion (mean improvement = 2.81 cm, p = 0.007), whereas the PPU group showed no significant change. Recorded adherence was higher in the App group compared with the PPU group (90.1% vs. 75.4%, p = 0.0069). The App group showed a significant reduction in symptom severity (p = 0.049). Between-group differences in symptom change were not statistically significant. MobileS appears feasible as a home-based CI management tool, and was associated with higher recorded adherence and preliminary NPC improvement signals. Larger, adequately powered studies are needed to confirm efficacy, particularly for symptoms, and guide teleophthalmology implementation.
This 24-month retrospective, exploratory observational study evaluated longitudinal spinal morphometric changes in 128 postmenopausal women with osteoporosis receiving anabolic therapy (n = 30) or non-anabolic anti-osteoporotic therapy (n = 98). Lumbar BMD, lower lumbar vertebral heights, disc height indices, and sagittal spinopelvic parameters were assessed at baseline and 24 months. Unweighted and baseline-severity adjusted linear regression models were used; IPTW and overlap-weighted analyses were performed as sensitivity analyses. Both groups showed significant lumbar BMD improvement. Patients receiving anabolic therapy were older and had lower baseline lumbar BMD and greater thoracolumbar kyphosis, indicating confounding by indication. In unweighted models, anabolic therapy was associated with greater L5 anterior height loss and L5/S1 disc height index preservation; however, these associations were attenuated after adjustment for baseline lumbar BMD and thoracolumbar kyphosis and were not consistently supported by propensity-score-weighted sensitivity analyses. Pelvic incidence showed more consistent level-specific associations, being negatively associated with L5 anterior height change and positively associated with L5/S1 disc height index preservation. Given substantial confounding by indication and limited treatment-group overlap, treatment-related associations should not be interpreted causally. These findings suggest that spinopelvic morphology may influence lower lumbar structural adaptation, whereas treatment-related morphometric findings remain exploratory and require confirmation.
Background: HER2 status guides targeted therapy in breast cancer but is currently determined by invasive biopsy. Imaging-based HER2 prediction from dynamic contrast-enhanced MRI (DCE-MRI) could provide a non-invasive adjunct decision-support signal, but published models are typically single-center with heterogeneous preprocessing that limits reproducibility. Methods: We trained a Triple-Head Dual-Attention ResNet (THDA-ResNet) that processes three DCE phases (pre-contrast, early post-contrast, and late post-contrast) on the multicenter BreastDCEDL dataset (n = 1149, I-SPY trials), and we compared it with Vision Transformer (ViT) and Convolutional Vision Transformer (CvT) baselines, all ImageNet-pretrained. We benchmarked 14 preprocessing strategies, with and without N4 bias-field correction. External validation used the independent BreastDCEDL_AMBL cohort (43 lesions). AUC confidence intervals used stratified bootstrap; model comparisons used DeLong's test. Results: THDA-ResNet achieved the highest AUC, 0.74 (95% CI 0.65-0.83), versus 0.66 for ViT and 0.63 for CvT, with the advantage reaching borderline significance over CvT (p=0.054) and not significant over ViT (p=0.14). At a threshold of 0.7, it retained discrimination (sensitivity 0.41, specificity 0.86), while transformers collapsed to near-trivial classifiers. External AUC was 0.66 (0.49-0.81). N4 correction did not improve performance. Conclusions: Attention-driven CNNs are a strong default for HER2 prediction from DCE-MRI on medium-sized cohorts, and N4 correction can be omitted, simplifying the pipeline.
Electrocardiogram (ECG) signals contain important clinical information associated with serum potassium abnormalities. However, in Taiwan, raw patient data and original medical signals generally cannot be taken outside the hospital environment, thereby limiting their subsequent reuse and cross-institutional applications. The objective of this study is to transform classification-related information contained in raw ECG signals into high-level SHAP features and to evaluate their feasibility as auxiliary or alternative features to the original wave-segment model outputs. To this end, this study proposes a time-series classification framework that integrates ECG wave-segment submodels, SHAP-based feature augmentation, and stacking ensemble learning for serum potassium abnormality prediction. Submodels are first trained separately using different ECG wave segments and their combinations. SHAP is then applied to transform the contributions of the wave-segment submodel outputs to the prediction results into high-level features. In addition, PCA features are included as a comparison baseline to analyze the effects of different feature transformation methods on classification performance. The experimental results show that incorporating SHAP-based augmented features improves ECG-based serum potassium abnormality prediction performance under most settings. Even when only SHAP-based augmented features are used for training, some models still maintain performance comparable to or better than the Baseline. Although PCA provides more stable classification balance for some patients, SHAP-based augmented features can still represent classification-related model contribution information under most settings while achieving better or comparable classification performance. Overall, even without directly using the original ECG signals in the final classification stage, these features retain a certain degree of discriminative information and demonstrate the potential to serve as alternative features to the original wave-segment model outputs. Therefore, the findings of this study provide a preliminary reference for future medical data reuse, cross-institutional collaboration, and privacy risk assessment.
Within the Diagnostic Criteria for Temporomandibular Disorders, masticatory myalgia is framed around muscle hyperactivity, yet a recognizable subgroup shows the opposite profile-low muscle tone, fatigue, masticatory pain, and a persistent sense that the bite no longer fits, without structural dental change-and responds poorly to conventional conservative care. We introduce G-SCORE (Gum-chewing-mediated Sensorimotor Calibration of Occlusion through Rhythmic Exercise), a proprioceptive recalibration strategy delivered as a bilateral gum-chewing (BGC) protocol: two equal halves of one sugar-free gum chewed simultaneously and symmetrically on both posterior segments in short, patient-titrated sessions, delivering balanced, rhythmic, low-load input to the trigeminal sensorimotor system. In a retrospective series of 41 patients, we characterized within-subject changes in pain, mandibular mobility, and occlusal parameters. Masticatory pain fell from 5.0 ± 2.9 to 0.7 ± 0.9 (0-10; n = 27; p < 0.001; r = 0.87), with 85% achieving a ≥2-point reduction. Occluding tooth positions and mouth opening also recovered; in an anterior open-bite subgroup, overbite improved (n = 6; p = 0.031). The observed changes were dominated by pain relief and restored bilateral occlusal contact rather than by raw force-a pattern more compatible with sensorimotor recalibration than with muscle hypertrophy. This low-cost, non-invasive technique may help selected patients avoid irreversible occlusal treatment.
Background: Osteoarthritis (OA) is a complex whole-joint disease imparting a substantial global socioeconomic burden. Photobiomodulation (PBM) has attracted increasing research interest as a non-pharmacological intervention for OA, but the intellectual structure and developmental trajectory of this research field remain incompletely understood. Objective: This study systematically mapped the intellectual landscape, research hotspots, thematic structure, and translational characteristics of PBM research in OA using bibliometric methods. Methods: Bibliographic records were retrieved from the Web of Science Core Collection database. Following data cleaning, 422 publications (1988-2026) were analyzed using the bibliometrix package in R and VOSviewer software. Results: Publication output increased substantially after 2015, with a marked rise after 2020. Keyword co-occurrence analysis classified 71 core keywords into seven clusters, revealing a dual-axis knowledge structure comprising a mechanistic biology axis (inflammation, chondrocytes, oxidative stress, and cartilage) and a clinical rehabilitation axis (pain, WOMAC, exercise, and physical therapy). Overlay visualization and thematic map analyses indicated a gradual shift in research focus from symptom-oriented rehabilitation research toward mechanistic investigations and regenerative medicine-related approaches involving platelet-rich plasma and mesenchymal stem cells. Reference co-citation analysis identified two major citation clusters connected through studies related to inflammation, pain management, and rehabilitation. Conclusions: The PBM-OA literature is characterized by a translational knowledge structure integrating mechanistic biology and rehabilitation-oriented research. Notably, recent publication trends indicate increasing scholarly attention to regenerative medicine-related approaches while continuing to position PBM within exercise-centered conservative management. To strengthen the evidence base and guide future investigations, future research should prioritize protocol standardization, dose-response validation, and long-term structural outcomes.
This article presents the IST-Yeasts Culture Collection (IST-Yeasts CC), which is a newly established repository dedicated to non-conventional blue yeasts isolated from (micro)algae-associated environments in Portugal. This collection currently comprises 115 yeast strains, the majority of which belong to the phylum Basidiomycota (92%), the genus Rhodotorula (69%), including R. mucilaginosa, R. diobovata, R. sphaerocarpa, and R. taiwanensis species. Other Basidiomycota species in the collection are: Cystobasidium minutum, C. slooffiae, Vishniacozyma carnescens, Moesziomyces aphidis, Sporobolomyces roseus, S. salmonicolor, and Naganishia diffluens. The collection also includes species from the Ascomycota phylum, such as Meyerozyma guilliermondii, Yamadazyma atlantica, and Cyberlindnera vartiovaarae. An initial functional screening performed, with one representative isolate per species, revealed the ability of these strains to produce carotenoids, lipids, riboflavin, biosurfactants, bioemulsifiers and auxins, as well as their capacity to grow on a broad range of carbon sources. These traits underscore their biotechnological potential within the circular bioeconomy and sustainable bioprocesses, positioning them as promising sources of bioactive natural products. By providing access to a diverse panel of marine-associated yeasts, the IST-Yeasts CC supports the development of innovative solutions in marine biotechnology and marine drug discovery, including the design of more resilient and high-performance algal cultivation systems through targeted co-cultivation strategies. Further information on the collection can be found at IST-Yeasts CC website.
Brain aging exhibits substantial interindividual heterogeneity, yet separating aging-related neuroanatomical variation from pathological influences remains methodologically challenging. To address this issue, we constructed a Clinically Defined Aging Reference (CDAR) cohort from the UK Biobank by excluding individuals with overt clinical pathology and applied the Surreal-GAN framework to characterize latent patterns of age-associated structural variations. A total of 26,251 participants were included. The model identified two co-occurring dimensions of brain aging, referred to as R1 and R2, that were stable across subsamples (R1: r = 0.873, R2: r = 0.953) and remained consistent when refitted separately in males and females (female: r = 0.792, male: r = 0.818). R1 was characterized by widespread gray matter reduction involving cortical, subcortical, and cerebellar regions and was associated with broadly poorer cognitive performance, adverse lifestyle profiles, metabolic and inflammatory alterations, and age-related diseases. R2 exhibited relative preservation of subcortical structures together with widespread preservation of cortical surface area and more selective differences in cortical thickness. Compared with R1, R2 showed weaker associations with cognition and peripheral physiological measures but retained associations with cardiovascular-related outcomes. These findings suggest that brain aging within a clinically defined aging reference cohort may involve multiple partially dissociable neuroanatomical dimensions rather than a single pattern, providing an operational reference for studying aging-related structural heterogeneity under reduced clinical confounding.
Photobiomodulation (PBM) has gained increasing attention in tissue engineering and cell biology, yet reproducible in vitro studies remain limited by the availability of standardized irradiation platforms. This study presents the design, construction, and technical validation of Ce3Light, a modular LED-based irradiation system developed for controlled photobiomodulation experiments under standard cell culture conditions. The platform was fabricated using additive manufacturing (PA12) and incorporates interchangeable LED modules, integrated monitoring of illuminance and temperature, and compatibility with conventional CO2 incubators. To demonstrate its applicability for in vitro research, the system was evaluated using MRC-5 fibroblasts exposed to four wavelengths (460, 530, 660, and 800 nm). Cellular responses were assessed by measuring intracellular glutathione levels and dehydrogenase activity. The experiments confirmed that the platform enabled stable and reproducible irradiation while detecting wavelength-dependent cellular responses consistent with previous reports in the literature. These findings primarily validate the functionality and suitability of Ce3Light as an experimental platform rather than establish new biological mechanisms. The developed system provides a versatile, reproducible, and adaptable tool for future photobiomodulation studies requiring precise control of irradiation conditions.
Biofabrication has significant potential to advance medicine and research by creating complex and anatomically accurate engineered tissues for implantation or in vitro modeling. However, a persisting challenge of the current biofabrication methods, such as hydrogel-based bioprinting, is to find an efficient, affordable, and reproducible method for the generation of hollow and branched geometries. This is critical for the recapitulation of anatomically realistic structures representative of cardiovascular, respiratory, and other organ systems. Existing bioprinting approaches require complex, multi-step processes and expensive specialized equipment, limiting accessibility and extending fabrication time. Here, we present an alternative 'sacrificial' method for the rapid and accessible creation of hollow and/or branched hydrogel constructs, which we term 'soft templating' (Sof-T). Sof-T utilizes ionic diffusion from a 3D-printed water-soluble polymer to crosslink surface-adsorbed hydrogels followed by the dissolution of the polymer, thus leaving behind the anatomically patterned hydrogels. Using this technique, we readily generated: (1) vascular-like bifurcated aortic conduits, with or without aneurysmal deformities; (2) upper and lower (branched) trachea models; and scale-reduced (3) human hearts and (4) bladders. Overall, Sof-T offers a simple, rapid, and cost-effective strategy for fabricating relatively complex, hollow hydrogel architectures, broadening the access to anatomically relevant constructs for biomedical research and translational and/or educational applications.
Background: Artificial intelligence (AI)-driven gait analysis using plantar pressure and ground reaction force (GRF) signals may provide objective digital biomarkers for rehabilitation, but clinical translation remains uncertain. This scoping review and evidence map aimed to summarize clinical applications, compare evidence maturity, and identify methodological and translational gaps. Methods: PubMed, Web of Science, Embase, and Scopus were searched from the earliest available indexed records in each database to May 2026. Original clinical studies using plantar pressure- or GRF-derived signals with AI methods for disease recognition, severity assessment, risk prediction, rehabilitation monitoring, or decision support were included. Results: Fifteen studies met the eligibility criteria. Evidence was concentrated in Parkinson's disease (PD), particularly PD recognition and freezing of gait prediction, where relatively more mature evidence was supported by multiple studies and participant-level or cross-dataset validation. Evidence for PD severity assessment, knee osteoarthritis monitoring, chronic ankle instability rehabilitation, fall-risk stratification, sarcopenia screening, peripheral artery disease recognition, and functional gait disorder classification remained less mature. Translation was limited by small or single-center samples, unclear participant-level data splitting, limited external validation, absent calibration, sparse explainable AI reporting, and insufficient real-world workflow testing. Conclusions: Future studies should prioritize prospective, externally validated, interpretable, calibrated, and clinically embedded models before routine rehabilitation implementation.
Total anomalous pulmonary venous connection (TAPVC) is a severe congenital heart disease, yet its prenatal detection rate remains suboptimal. To support prenatal ultrasound screening of TAPVC, the post-left atrium space (PLAS) index and the left-atrial posterior-space-to-diagonal (LAPSD) ratio measured in the four-chamber view (4CV) have been proposed as useful biometric parameters. In this study, we developed a novel approach that integrates automated 4CV extraction (AE) from fetal cardiac ultrasound videos with automated measurement of these indices. The heart, crux, and descending aorta were segmented using DeepLabv3+, UNet3+, and SegFormer. The screening performance of the AE-based methods was comparable to that of manual 4CV extraction, as demonstrated by similar mean areas under the receiver operating characteristic curve (AUCs). In a clinical comparison study, the mean AUC values for residents, fellows, experts, AE-DeepLabv3+, AE-UNet3+, and AE-SegFormer were 0.784, 0.801, 0.996, 0.903, 0.928, and 0.940, respectively, for the PLAS index and 0.797, 0.801, 0.996, 0.919, 0.916, and 0.940, respectively, for the LAPSD ratio. Although experts demonstrated the best overall performance, the fully automated methods consistently outperformed both the residents and fellows. This approach may support less experienced examiners, improve screening accuracy, streamline clinical workflows, and ultimately enhance the prenatal detection of TAPVC.