An effective diagnosis system and suitable treatment planning require the precise segmentation of thyroid nodules in ultrasound imaging. The advancement of imaging technologies has not resolved traditional imaging challenges, which include noise issues, limited contrast, and dependency on operator choices, thus highlighting the need for automated, reliable solutions. The researchers developed TATHA, an innovative deep learning architecture dedicated to improving thyroid ultrasound image segmentation accuracy. The model is evaluated using the digital database of thyroid ultrasound images, which includes 99 cases across three subsets containing 134 labelled images for training, validation, and testing. It incorporates data pre-treatment procedures that reduce speckle noise and enhance contrast, while edge detection provides high-quality input for segmentation. TATHA outperforms U-Net, PSPNet, and Vision Transformers across various datasets and cross-validation folds, achieving superior Dice scores, accuracy, and AUC results. The distributed thyroid segmentation framework generates reliable predictions by combining results from multiple feature extraction units. The findings confirm that these advancements make TATHA an essential tool for clinicians and researchers in thyroid imaging and clinical applications.

Ulcerative colitis (UC) is a chronic inflammatory disorder necessitating precise severity stratification to facilitate optimal therapeutic interventions. This study harnesses a triple-pronged deep learning methodology-including multimodal inference pipelines that eliminate domain-specific training, few-shot meta-learning, and Vision Transformer (ViT)-based ensembling-to classify UC severity within the HyperKvasir dataset. We systematically evaluate multiple vision transformer architectures, discovering that a Swin-Base model achieves an accuracy of 90%, while a soft-voting ensemble of diverse ViT backbones boosts performance to 93%. In parallel, we leverage multimodal pre-trained frameworks (e.g., CLIP, BLIP, FLAVA) integrated with conventional machine learning algorithms, yielding an accuracy of 83%. To address limited annotated data, we deploy few-shot meta-learning approaches (e.g., Matching Networks), attaining 83% accuracy in a 5-shot context. Furthermore, interpretability is enhanced via SHapley Additive exPlanations (SHAP), which interpret both local and global model behaviors, thereby fostering clinical trust in the model's inferences. These findings underscore the potential of contemporary representation learning and ensemble strategies for robust UC severity classification, highlighting the pivotal role of model transparency in facilitating medical image analysis.

Early detection of steatotic liver disease (SLD) is critically important. In clinical practice, hepatic steatosis is frequently diagnosed using computed tomography (CT) performed for unrelated clinical indications. An equation for estimating magnetic resonance proton density fat fraction (MR-PDFF) using liver attenuation on non-contrast CT exists, but no equivalent equation exists for post-contrast CT. We sought to (1) determine whether an automated workflow can accurately measure liver attenuation, (2) validate previously identified optimal thresholds for liver or liver-spleen attenuation in post-contrast studies, and (3) develop a method for estimating MR-PDFF (FF) on post-contrast CT. The fully automated TotalSegmentator 'total' machine learning model was used to segment 3D liver and spleen from non-contrast and post-contrast CT scans. Mean attenuation was extracted from liver (L) and spleen (S) volumes and from manually placed regions of interest (ROIs) in multi-phase CT scans of two cohorts: derivation (n = 1740) and external validation (n = 1044). Non-linear regression was used to determine the optimal coefficients for three phase-specific (arterial, venous, delayed) increasing exponential decay equations relating post-contrast L to non-contrast L. MR-PDFF was estimated from non-contrast CT and used as the reference standard. The mean attenuation for manual ROIs versus automated volumes were nearly perfectly correlated for both liver and spleen (r > .96, p < .001). For moderate-to-severe steatosis (L < 40 HU), the density of the liver (L) alone was a better classifier than either liver-spleen difference (L-S) or ratio (L/S) on post-contrast CTs. Fat fraction calculated using a corrected post-contrast liver attenuation measure agreed with non-contrast FF > 15% in both the derivation and external validation cohort, with AUROC between 0.92 and 0.97 on arterial, venous, and delayed phases. Automated volumetric mean attenuation of liver and spleen can be used instead of manually placed ROIs for liver fat assessments. Liver attenuation alone in post-contrast phases can be used to assess the presence of moderate-to-severe hepatic steatosis. Correction equations for liver attenuation on post-contrast phase CT scans enable reasonable quantification of liver steatosis, providing potential opportunities for utilizing clinical scans to develop large scale screening or studies in SLD.

Background: Manual extraction of pancreatic cystic lesion (PCL) features from radiology reports is labor-intensive, limiting large-scale studies needed to advance PCL research. Purpose: To develop and evaluate large language models (LLMs) that automatically extract PCL features from MRI/CT reports and assign risk categories based on guidelines. Materials and Methods: We curated a training dataset of 6,000 abdominal MRI/CT reports (2005-2024) from 5,134 patients that described PCLs. Labels were generated by GPT-4o using chain-of-thought (CoT) prompting to extract PCL and main pancreatic duct features. Two open-source LLMs were fine-tuned using QLoRA on GPT-4o-generated CoT data. Features were mapped to risk categories per institutional guideline based on the 2017 ACR White Paper. Evaluation was performed on 285 held-out human-annotated reports. Model outputs for 100 cases were independently reviewed by three radiologists. Feature extraction was evaluated using exact match accuracy, risk categorization with macro-averaged F1 score, and radiologist-model agreement with Fleiss' Kappa. Results: CoT fine-tuning improved feature extraction accuracy for LLaMA (80% to 97%) and DeepSeek (79% to 98%), matching GPT-4o (97%). Risk categorization F1 scores also improved (LLaMA: 0.95; DeepSeek: 0.94), closely matching GPT-4o (0.97), with no statistically significant differences. Radiologist inter-reader agreement was high (Fleiss' Kappa = 0.888) and showed no statistically significant difference with the addition of DeepSeek-FT-CoT (Fleiss' Kappa = 0.893) or GPT-CoT (Fleiss' Kappa = 0.897), indicating that both models achieved agreement levels on par with radiologists. Conclusion: Fine-tuned open-source LLMs with CoT supervision enable accurate, interpretable, and efficient phenotyping for large-scale PCL research, achieving performance comparable to GPT-4o.

Radiography imaging protocols target on specific anatomical regions, resulting in highly consistent images with recurrent structural patterns across patients. Recent advances in medical anomaly detection have demonstrated the effectiveness of CNN- and transformer-based approaches. However, CNNs exhibit limitations in capturing long-range dependencies, while transformers suffer from quadratic computational complexity. In contrast, Mamba-based models, leveraging superior long-range modeling, structural feature extraction, and linear computational efficiency, have emerged as a promising alternative. To capitalize on the inherent structural regularity of medical images, this study introduces SP-Mamba, a spatial-perception Mamba framework for unsupervised medical anomaly detection. The window-sliding prototype learning and Circular-Hilbert scanning-based Mamba are introduced to better exploit consistent anatomical patterns and leverage spatial information for medical anomaly detection. Furthermore, we excavate the concentration and contrast characteristics of anomaly maps for improving anomaly detection. Extensive experiments on three diverse medical anomaly detection benchmarks confirm the proposed method's state-of-the-art performance, validating its efficacy and robustness. The code is available at https://github.com/Ray-RuiPan/SP-Mamba.

BackgroundPericardial disease exhibits a wide clinical spectrum, ranging from mild effusions to life-threatening tamponade or constriction pericarditis. While transthoracic echocardiography (TTE) is the primary diagnostic modality, its effectiveness is limited by operator dependence and incomplete evaluation of functional impact. Existing artificial intelligence models focus primarily on effusion detection, lacking comprehensive disease assessment. MethodsWe developed a deep learning (DL)-based framework that sequentially assesses pericardial disease: (1) morphological changes, including pericardial effusion amount (normal/small/moderate/large) and pericardial thickening or adhesion (yes/no), using five B-mode views, and (2) hemodynamic significance (yes/no), incorporating additional inputs from Doppler and inferior vena cava measurements. The developmental dataset comprises 2,253 TTEs from multiple Korean institutions (225 for internal testing), and the independent external test set consists of 274 TTEs. ResultsIn the internal test set, the model achieved diagnostic accuracy of 81.8-97.3% for pericardial effusion classification, 91.6% for pericardial thickening/adhesion, and 86.2% for hemodynamic significance. Corresponding accuracy in the external test set was 80.3-94.2%, 94.5%, and 85.5%, respectively. Area under the receiver operating curves (AUROCs) for the three tasks in the internal test set was 0.92-0.99, 0.90, and 0.79; and in the external test set, 0.95-0.98, 0.85, and 0.76. Sensitivity for detecting pericardial thickening/adhesion and hemodynamic significance was modest (66.7% and 68.8% in the internal test set), but improved substantially when cases with poor image quality were excluded (77.3%, and 80.8%). Similar performance gains were observed in subgroups with complete target views and a higher number of available video clips. ConclusionsThis study presents the first DL-based TTE model capable of comprehensive evaluation of pericardial disease, integrating both morphological and functional assessments. The proposed framework demonstrated strong generalizability and aligned with the real-world diagnostic workflow. However, caution is warranted when interpreting results under suboptimal imaging conditions.

Medical imaging has significantly advanced computer-aided diagnosis, yet its re-identification (ReID) risks raise critical privacy concerns, calling for de-identification (DeID) techniques. Unfortunately, existing DeID methods neither particularly preserve medical semantics, nor are flexibly adjustable towards different privacy levels. To address these issues, we propose a divide-and-conquer framework comprising two steps: (1) Identity-Blocking, which blocks varying proportions of identity-related regions, to achieve different privacy levels; and (2) Medical-Semantics-Compensation, which leverages pre-trained Medical Foundation Models (MFMs) to extract medical semantic features to compensate the blocked regions. Moreover, recognizing that features from MFMs may still contain residual identity information, we introduce a Minimum Description Length principle-based feature decoupling strategy, to effectively decouple and discard such identity components. Extensive evaluations against existing approaches across seven datasets and three downstream tasks, demonstrates our state-of-the-art performance.

Unsupervised anomaly detection (UAD) aims to detect anomalies without labeled data, a necessity in many machine learning applications where anomalous samples are rare or not available. Most state-of-the-art methods fall into two categories: reconstruction-based approaches, which often reconstruct anomalies too well, and decoupled representation learning with density estimators, which can suffer from suboptimal feature spaces. While some recent methods attempt to couple feature learning and anomaly detection, they often rely on surrogate objectives, restrict kernel choices, or introduce approximations that limit their expressiveness and robustness. To address this challenge, we propose a novel method that tightly couples representation learning with an analytically solvable one-class SVM (OCSVM), through a custom loss formulation that directly aligns latent features with the OCSVM decision boundary. The model is evaluated on two tasks: a new benchmark based on MNIST-C, and a challenging brain MRI subtle lesion detection task. Unlike most methods that focus on large, hyperintense lesions at the image level, our approach succeeds to target small, non-hyperintense lesions, while we evaluate voxel-wise metrics, addressing a more clinically relevant scenario. Both experiments evaluate a form of robustness to domain shifts, including corruption types in MNIST-C and scanner/age variations in MRI. Results demonstrate performance and robustness of our proposed mode,highlighting its potential for general UAD and real-world medical imaging applications. The source code is available at https://github.com/Nicolas-Pinon/uad_ocsvm_guided_repr_learning

Purpose: To develop a reconstruction framework for 3D real-time cine cardiovascular magnetic resonance (CMR) from highly undersampled data without requiring fully sampled training data. Methods: We developed a multi-dynamic low-rank deep image prior (ML-DIP) framework that models spatial image content and temporal deformation fields using separate neural networks. These networks are optimized per scan to reconstruct the dynamic image series directly from undersampled k-space data. ML-DIP was evaluated on (i) a 3D cine digital phantom with simulated premature ventricular contractions (PVCs), (ii) ten healthy subjects (including two scanned during both rest and exercise), and (iii) five patients with PVCs. Phantom results were assessed using peak signal-to-noise ratio (PSNR) and structural similarity index measure (SSIM). In vivo performance was evaluated by comparing left-ventricular function quantification (against 2D real-time cine) and image quality (against 2D real-time cine and binning-based 5D-Cine). Results: In the phantom study, ML-DIP achieved PSNR > 29 dB and SSIM > 0.90 for scan times as short as two minutes, while recovering cardiac motion, respiratory motion, and PVC events. In healthy subjects, ML-DIP yielded functional measurements comparable to 2D cine and higher image quality than 5D-Cine, including during exercise with high heart rates and bulk motion. In PVC patients, ML-DIP preserved beat-to-beat variability and reconstructed irregular beats, whereas 5D-Cine showed motion artifacts and information loss due to binning. Conclusion: ML-DIP enables high-quality 3D real-time CMR with acceleration factors exceeding 1,000 by learning low-rank spatial and temporal representations from undersampled data, without relying on external fully sampled training datasets.

Precise and automated segmentation of the liver and its tumor within CT scans plays a pivotal role in swift diagnosis and the development of optimal treatment plans for individuals with liver diseases and malignancies. However, automated liver and tumor segmentation faces significant hurdles arising from the inherent heterogeneity of tumors and the diverse visual characteristics of livers across a broad spectrum of patients. Aiming to address these challenges, we present a novel Transformer-aware Multiscale Progressive Encoder-Decoder Network (T-MPEDNet) for automated segmentation of tumor and liver. T-MPEDNet leverages a deep adaptive features backbone through a progressive encoder-decoder structure, enhanced by skip connections for recalibrating channel-wise features while preserving spatial integrity. A Transformer-inspired dynamic attention mechanism captures long-range contextual relationships within the spatial domain, further enhanced by multi-scale feature utilization for refined local details, leading to accurate prediction. Morphological boundary refinement is then employed to address indistinct boundaries with neighboring organs, capturing finer details and yielding precise boundary labels. The efficacy of T-MPEDNet is comprehensively assessed on two widely utilized public benchmark datasets, LiTS and 3DIRCADb. Extensive quantitative and qualitative analyses demonstrate the superiority of T-MPEDNet compared to twelve state-of-the-art methods. On LiTS, T-MPEDNet achieves outstanding Dice Similarity Coefficients (DSC) of 97.6% and 89.1% for liver and tumor segmentation, respectively. Similar performance is observed on 3DIRCADb, with DSCs of 98.3% and 83.3% for liver and tumor segmentation, respectively. Our findings prove that T-MPEDNet is an efficacious and reliable framework for automated segmentation of the liver and its tumor in CT scans.