Deep learning models for medical image analysis often suffer from performance degradation when applied to data from different scanners or protocols, a phenomenon known as domain shift. This study investigates this challenge in the context of sex classification from 3D T1-weighted brain magnetic resonance imaging (MRI) scans using the IXI and OASIS3 datasets. While models achieved high within-domain accuracy (around 0.95) when trained and tested on a single dataset (IXI or OASIS3), we demonstrate a significant performance drop to chance level (about 0.50) when models trained on one dataset are tested on the other, highlighting the presence of a strong domain shift. To address this, we employed the ComBat harmonization technique to align the feature distributions of the two datasets. We evaluated three state-of-the-art 3D deep learning architectures (3D ResNet18, 3D DenseNet, and 3D EfficientNet) across multiple training strategies. Our results show that ComBat harmonization effectively reduces the domain shift, leading to a substantial improvement in cross-domain classification performance. For instance, the cross-domain balanced accuracy of our best model (ResNet18 3D with Attention) improved from approximately 0.50 (chance level) to 0.61 after harmonization. t-SNE visualization of extracted features provides clear qualitative evidence of the reduced domain discrepancy post-harmonization. This work underscores the critical importance of domain adaptation techniques for building robust and generalizable neuroimaging AI models.

Ear diseases affect billions of people worldwide, leading to substantial health and socioeconomic burdens. Computed tomography (CT) plays a pivotal role in accurate diagnosis, treatment planning, and outcome evaluation. The objective of this study is to present the establishment and design of UltraEar Database, a large-scale, multicentric repository of isotropic 0.1 mm ultra-high-resolution CT (U-HRCT) images and associated clinical data dedicated to ear diseases. UltraEar recruits patients from 11 tertiary hospitals between October 2020 and October 2035, integrating U-HRCT images, structured CT reports, and comprehensive clinical information, including demographics, audiometric profiles, surgical records, and pathological findings. A broad spectrum of otologic disorders is covered, such as otitis media, cholesteatoma, ossicular chain malformation, temporal bone fracture, inner ear malformation, cochlear aperture stenosis, enlarged vestibular aqueduct, and sigmoid sinus bony deficiency. Standardized preprocessing pipelines have been developed for geometric calibration, image annotation, and multi-structure segmentation. All personal identifiers in DICOM headers and metadata are removed or anonymized to ensure compliance with data privacy regulation. Data collection and curation are coordinated through monthly expert panel meetings, with secure storage on an offline cloud system. UltraEar provides an unprecedented ultra-high-resolution reference atlas with both technical fidelity and clinical relevance. This resource has significant potential to advance radiological research, enable development and validation of AI algorithms, serve as an educational tool for training in otologic imaging, and support multi-institutional collaborative studies. UltraEar will be continuously updated and expanded, ensuring long-term accessibility and usability for the global otologic research community.

Although traditional statistical techniques and machine learning methods have contributed significantly to genetics and, in particular, inherited disease diagnosis, they often struggle with complex, high-dimensional data, a challenge now addressed by state-of-the-art deep learning models. Large language models (LLMs), based on transformer architectures, have excelled in tasks requiring contextual comprehension of unstructured medical data. This systematic review examines the role of LLMs in the genetic research and diagnostics of both rare and common diseases. Automated keyword-based search in PubMed, bioRxiv, medRxiv, and arXiv was conducted, targeting studies on LLM applications in diagnostics and education within genetics and removing irrelevant or outdated models. A total of 172 studies were analyzed, highlighting applications in genomic variant identification, annotation, and interpretation, as well as medical imaging advancements through vision transformers. Key findings indicate that while transformer-based models significantly advance disease and risk stratification, variant interpretation, medical imaging analysis, and report generation, major challenges persist in integrating multimodal data (genomic sequences, imaging, and clinical records) into unified and clinically robust pipelines, facing limitations in generalizability and practical implementation in clinical settings. This review provides a comprehensive classification and assessment of the current capabilities and limitations of LLMs in transforming hereditary disease diagnostics and supporting genetic education, serving as a guide to navigate this rapidly evolving field.

Enlarged perivascular spaces (PVS) are increasingly recognized as biomarkers of cerebral small vessel disease, Alzheimer's disease, stroke, and aging-related neurodegeneration. However, manual segmentation of PVS is time-consuming and subject to moderate inter-rater reliability, while existing automated deep learning models have moderate performance and typically fail to generalize across diverse clinical and research MRI datasets. We adapted MedNeXt-L-k5, a Transformer-inspired 3D encoder-decoder convolutional network, for automated PVS segmentation. Two models were trained: one using a homogeneous dataset of 200 T2-weighted (T2w) MRI scans from the Human Connectome Project-Aging (HCP-Aging) dataset and another using 40 heterogeneous T1-weighted (T1w) MRI volumes from seven studies across six scanners. Model performance was evaluated using internal 5-fold cross validation (5FCV) and leave-one-site-out cross validation (LOSOCV). MedNeXt-L-k5 models trained on the T2w images of the HCP-Aging dataset achieved voxel-level Dice scores of 0.88+/-0.06 (white matter, WM), comparable to the reported inter-rater reliability of that dataset, and the highest yet reported in the literature. The same models trained on the T1w images of the HCP-Aging dataset achieved a substantially lower Dice score of 0.58+/-0.09 (WM). Under LOSOCV, the model had voxel-level Dice scores of 0.38+/-0.16 (WM) and 0.35+/-0.12 (BG), and cluster-level Dice scores of 0.61+/-0.19 (WM) and 0.62+/-0.21 (BG). MedNeXt-L-k5 provides an efficient solution for automated PVS segmentation across diverse T1w and T2w MRI datasets. MedNeXt-L-k5 did not outperform the nnU-Net, indicating that the attention-based mechanisms present in transformer-inspired models to provide global context are not required for high accuracy in PVS segmentation.

Medical image segmentation has advanced rapidly over the past two decades, largely driven by deep learning, which has enabled accurate and efficient delineation of cells, tissues, organs, and pathologies across diverse imaging modalities. This progress raises a fundamental question: to what extent have current models overcome persistent challenges, and what gaps remain? In this work, we provide an in-depth review of medical image segmentation, tracing its progress and key developments over the past decade. We examine core principles, including multiscale analysis, attention mechanisms, and the integration of prior knowledge, across the encoder, bottleneck, skip connections, and decoder components of segmentation networks. Our discussion is organized around seven key dimensions: (1) the shift from supervised to semi-/unsupervised learning, (2) the transition from organ segmentation to lesion-focused tasks, (3) advances in multi-modality integration and domain adaptation, (4) the role of foundation models and transfer learning, (5) the move from deterministic to probabilistic segmentation, (6) the progression from 2D to 3D and 4D segmentation, and (7) the trend from model invocation to segmentation agents. Together, these perspectives provide a holistic overview of the trajectory of deep learning-based medical image segmentation and aim to inspire future innovation. To support ongoing research, we maintain a continually updated repository of relevant literature and open-source resources at https://github.com/apple1986/medicalSegReview

Explainable Artificial Intelligence (XAI) can decode the 'black box' models, enhancing trust in clinical decision-making. XAI makes the predictions of deep learning models interpretable, transparent, and trustworthy. This study employed XAI techniques to explain the predictions made by a deep learning-based model for diagnosing autism and identifying the memory regions responsible for children's academic performance. This study utilized publicly available sMRI data from the ABIDE-II repository. First, a deep learning model, FaithfulNet, was developed to aid in the diagnosis of autism. Next, gradient-based class activation maps and the SHAP gradient explainer were employed to generate explanations for the model's predictions. These explanations were integrated to develop a novel and faithful visual explanation, Faith_CAM. Finally, this faithful explanation was quantified using the pointing game score and analyzed with cortical and subcortical structure masks to identify the impaired brain regions in the autistic brain. This study achieved a classification accuracy of 99.74% with an AUC value of 1. In addition to facilitating autism diagnosis, this study assesses the degree of impairment in memory regions responsible for the children's academic performance, thus contributing to the development of personalized treatment plans.

Osteoporosis is a bone disease characterized by reduced bone mineral density and quality, which increases the risk of fragility fractures. The current diagnostic gold standard, dual-energy X-ray absorptiometry (DXA), faces limitations such as low equipment penetration, high testing costs, and radiation exposure, restricting its feasibility as a screening tool. To address these limitations, We retrospectively collected data from 1995 patients who visited Daping Hospital in Chongqing from January 2019 to August 2024. We developed an opportunistic screening method using chest X-rays. Furthermore, we designed three innovative deep neural network models using transfer learning: Inception v3, VGG16, and ResNet50. These models were evaluated based on their classification performance for osteoporosis using chest X-ray images, with external validation via multi-center data. The ResNet50 model demonstrated superior performance, achieving average accuracies of 87.85 % and 90.38 % in the internal test dataset across two experiments, with AUC values of 0.945 and 0.957, respectively. These results outperformed traditional convolutional neural networks. In the external validation, the ResNet50 model achieved an AUC of 0.904, accuracy of 89 %, sensitivity of 90 %, and specificity of 88.57 %, demonstrating strong generalization ability. And the model shows robust performance despite concurrent pulmonary pathologies. This study provides an automatic screening method for osteoporosis using chest X-rays, without additional radiation exposure or cost. The ResNet50 model's high performance supports clinicians in the early identification and treatment of osteoporosis patients.

This study investigates deep learning methods for automated classification of dental conditions in panoramic X-ray images. A dataset of 1,512 radiographs with 11,137 expert-verified annotations across four conditions fillings, cavities, implants, and impacted teeth was used. After preprocessing and class balancing, three approaches were evaluated: a custom convolutional neural network (CNN), hybrid models combining CNN feature extraction with traditional classifiers, and fine-tuned pre-trained architectures. Experiments employed 5 fold cross validation with accuracy, precision, recall, and F1 score as evaluation metrics. The hybrid CNN Random Forest model achieved the highest performance with 85.4% accuracy, surpassing the custom CNN baseline of 74.3%. Among pre-trained models, VGG16 performed best at 82.3% accuracy, followed by Xception and ResNet50. Results show that hybrid models improve discrimination of morphologically similar conditions and provide efficient, reliable performance. These findings suggest that combining CNN-based feature extraction with ensemble classifiers offers a practical path toward automated dental diagnostic support, while also highlighting the need for larger datasets and further clinical validation.

Cerebral microbleeds (CMB) are small hypointense lesions visible on gradient echo (GRE) or susceptibility-weighted (SWI) MRI, serving as critical biomarkers for various cerebrovascular and neurological conditions. Accurate quantification of CMB is essential, as their number correlates with the severity of conditions such as small vessel disease, stroke risk and cognitive decline. Current detection methods depend on manual inspection, which is time-consuming and prone to variability. Automated detection using deep learning presents a transformative solution but faces challenges due to the heterogeneous appearance of CMB, high false-positive rates, and similarity to other artefacts. This study investigates the application of deep learning techniques to public (ADNI and AIBL) and private datasets (OATS and MAS), leveraging GRE and SWI MRI modalities to enhance CMB detection accuracy, reduce false positives, and ensure robustness in both clinical and normal cases (i.e., scans without cerebral microbleeds). A 3D convolutional neural network (CNN) was developed for automated detection, complemented by a You Only Look Once (YOLO)-based approach to address false positive cases in more complex scenarios. The pipeline incorporates extensive preprocessing and validation, demonstrating robust performance across a diverse range of datasets. The proposed method achieves remarkable performance across four datasets, ADNI: Balanced accuracy: 0.953, AUC: 0.955, Precision: 0.954, Sensitivity: 0.920, F1-score: 0.930, AIBL: Balanced accuracy: 0.968, AUC: 0.956, Precision: 0.956, Sensitivity: 0.938, F1-score: 0.946, MAS: Balanced accuracy: 0.889, AUC: 0.889, Precision: 0.948, Sensitivity: 0.779, F1-score: 0.851, and OATS dataset: Balanced accuracy: 0.93, AUC: 0.930, Precision: 0.949, Sensitivity: 0.862, F1-score: 0.900. These results highlight the potential of deep learning models to improve early diagnosis and support treatment planning for conditions associated with CMB.

Breast cancer is a leading malignancy threatening womens health globally, making early and accurate diagnosis crucial. Ultrasound is a key screening and diagnostic tool due to its non- invasive, real-time, and cost-effective nature. However, its diagnostic accuracy is highly dependent on operator experience, and conventional single-image analysis often fails to capture the comprehensive features of a lesion. This study introduces a computer-aided diagnosis (CAD) system that emulates a clinicians multi-view diagnostic process. We developed a novel hybrid deep learning model that integrates a Convolutional Neural Network (CNN) with a Transformer architecture. The model uses a pretrained EfficientNetV2 to extract spatial features from multiple, unordered ultrasound images of a single lesion. These features are then processed by a Transformer encoder, whose self-attention mechanism globally models and fuses their intrinsic correlations. A strict lesion-level data partitioning strategy ensured a rigorous evaluation. On an internal test set, our CNN-Transformer model achieved an accuracy of 0.93, a sensitivity of 0.92, a specificity of 0.94, and an Area Under the Curve (AUC) of 0.98. On an external test set, it demonstrated an accuracy of 0.93, a sensitivity of 0.94, a specificity of 0.91, and an AUC of 0.97. These results significantly outperform those of a baseline single-image model, which achieved accuracies of 0.88 and 0.89 and AUCs of 0.95 and 0.94 on the internal and external test sets, respectively. This study confirms that combining CNNs with Transformers yields a highly accurate and robust diagnostic system for breast ultrasound. By effectively fusing multi-view information, our model aligns with clinical logic and shows immense potential for improving diagnostic reliability.