Uncertainty estimation methods are essential for the application of artificial intelligence (AI) models in medical image segmentation, particularly in addressing reliability and feasibility challenges in clinical deployment. Despite their significance, the adoption of uncertainty estimation methods in clinical practice remains limited due to the lack of a comprehensive evaluation framework tailored to their clinical usage. To address this gap, a simulation of uncertainty-assisted clinical workflows is conducted, highlighting the roles of uncertainty in model selection, sample screening, and risk visualization. Furthermore, uncertainty evaluation is extended to pixel, sample, and model levels to enable a more thorough assessment. At the pixel level, the Uncertainty Confusion Metric (UCM) is proposed, utilizing density curves to improve robustness against variability in uncertainty distributions and to assess the ability of pixel uncertainty to identify potential errors. At the sample level, the Expected Segmentation Calibration Error (ESCE) is introduced to provide more accurate calibration aligned with Dice, enabling more effective identification of low-quality samples. At the model level, the Harmonic Dice (HDice) metric is developed to integrate uncertainty and accuracy, mitigating the influence of dataset biases and offering a more robust evaluation of model performance on unseen data. Using this systematic evaluation framework, five mainstream uncertainty estimation methods are compared on organ and tumor datasets, providing new insights into their clinical applicability. Extensive experimental analyses validated the practicality and effectiveness of the proposed metrics. This study offers clear guidance for selecting appropriate uncertainty estimation methods in clinical settings, facilitating their integration into clinical workflows and ultimately improving diagnostic efficiency and patient outcomes.

This study develops a deep-learning method for automatic segmentation of the interventricular septum (IS) in MR images to measure myocardial T2* and estimate cardiac iron deposition in patients with thalassemia. This retrospective study used multiple-gradient-echo cardiac MR scans from 419 thalassemia patients to develop and evaluate the segmentation network. The network was trained on 1.5 T images from Center 1 and evaluated on 3.0 T unseen images from Center 1, all data from Center 2, and the CHMMOTv1 dataset. Model performance was assessed using five metrics, and T2* values were obtained by fitting the network output. Bland-Altman analysis, coefficient of variation (CoV), and regression analysis were used to evaluate the consistency between automatic and manual methods. MA-BBIsegNet achieved a Dice of 0.90 on the internal test set, 0.85 on the external test set, and 0.81 on the CHMMOTv1 dataset. Bland-Altman analysis showed mean differences of 0.08 (95% LoA: -2.79 ∼ 2.63) ms (internal), 0.29 (95% LoA: -4.12 ∼ 3.54) ms (external) and 0.19 (95% LoA: -3.50 ∼ 3.88) ms (CHMMOTv1), with CoV of 8.9%, 6.8%, and 9.3%. Regression analysis yielded r values of 0.98 for the internal and CHMMOTv1 datasets, and 0.99 for the external dataset (p < 0.05). The IS segmentation network based on multiple-gradient-echo bright-blood images yielded T2* values that were in strong agreement with manual measurements, highlighting its potential for the efficient, non-invasive monitoring of myocardial iron deposition in patients with thalassemia.

The U-shaped encoder-decoder architecture with skip connections has become a prevailing paradigm in medical image segmentation due to its simplicity and effectiveness. While many recent works aim to improve this framework by designing more powerful encoders and decoders, employing advanced convolutional neural networks (CNNs) for local feature extraction, Transformers or state space models (SSMs) such as Mamba for global context modeling, or hybrid combinations of both, these methods often struggle to fully utilize pretrained vision backbones (e.g., ResNet, ViT, VMamba) due to structural mismatches. To bridge this gap, we introduce ACM-UNet, a general-purpose segmentation framework that retains a simple UNet-like design while effectively incorporating pretrained CNNs and Mamba models through a lightweight adapter mechanism. This adapter resolves architectural incompatibilities and enables the model to harness the complementary strengths of CNNs and SSMs-namely, fine-grained local detail extraction and long-range dependency modeling. Additionally, we propose a hierarchical multi-scale wavelet transform module in the decoder to enhance feature fusion and reconstruction fidelity. Extensive experiments on the Synapse and ACDC benchmarks demonstrate that ACM-UNet achieves state-of-the-art performance while remaining computationally efficient. Notably, it reaches 85.12% Dice Score and 13.89mm HD95 on the Synapse dataset with 17.93G FLOPs, showcasing its effectiveness and scalability. Code is available at: https://github.com/zyklcode/ACM-UNet.

Digital pathology relies on the morphological architecture of prostate glands to recognize cancerous tissue. Prostate cancer (PCa) originates in walnut shaped prostate gland in the male reproductive system. Deep learning (DL) pipelines can assist in identifying these regions with advanced segmentation techniques which are effective in diagnosing and treating prostate diseases. This facilitates early detection, targeted biopsy, and accurate treatment planning, ensuring consistent, reproducible results while minimizing human error. Automated segmentation techniques trained on MRI datasets can aid in monitoring disease progression which leads to clinical support by developing patient-specific models for personalized medicine. In this study, we present multiclass segmentation models designed to localize the prostate gland and its zonal regions-specifically the peripheral zone (PZ), transition zone (TZ), and the whole gland-by combining EfficientNetB4 encoders with Self-organized Operational Neural Network (Self-ONN)-based decoders. Traditional convolutional neural networks (CNNs) rely on linear neuron models, which limit their ability to capture the complex dynamics of biological neural systems. In contrast, Operational Neural Networks (ONNs), particularly Self-ONNs, address this limitation by incorporating nonlinear and adaptive operations at the neuron level. We evaluated various encoder-decoder configurations and identified that the combination of an EfficientNet-based encoder with a Self-ONN-based decoder yielded the best performance. To further enhance segmentation accuracy, we employed the STAPLE method to ensemble the top three performing models. Our approach was tested on the large-scale, recently updated PI-CAI Challenge dataset using 5-fold cross-validation, achieving Dice scores of 95.33 % for the whole gland and 92.32 % for the combined PZ and TZ regions. These advanced segmentation techniques significantly improve the quality of PCa diagnosis and treatment, contributing to better patient care and outcomes.

Aortic pulse wave velocity (PWV) is a prognostic biomarker for cardiovascular disease, which can be measured by dividing the aortic path length by the pulse transit time. However, current MRI techniques require special sequences and time-consuming manual analysis. We aimed to fully automate the process using deep learning to measure PWV from standard sequences, facilitating PWV measurement in routine clinical and research scans. A deep learning (DL) model was developed to generate high-resolution 3D aortic segmentations from routine 2D trans-axial SSFP localizer images, and the centerlines of the resulting segmentations were used to estimate the aortic path length. A further DL model was built to automatically segment the ascending and descending aorta in phase contrast images, and pulse transit time was estimated from the sampled flow curves. Quantitative comparison with trained observers was performed for path length, aortic flow segmentation and transit time, either using an external clinical dataset with both localizers and paired 3D images acquired or on a sample of UK Biobank subjects. Potential application to clinical research scans was evaluated on 1053 subjects from the UK Biobank. Aortic path length measurement was accurate with no major difference between the proposed method (125 ± 19 mm) and manual measurement by a trained observer (124 ± 19 mm) (P = 0.88). Automated phase contrast image segmentation was similar to that of a trained observer for both the ascending (Dice vs manual: 0.96) and descending (Dice 0.89) aorta with no major difference in transit time estimation (proposed method = 21 ± 9 ms, manual = 22 ± 9 ms; P = 0.15). 966 of 1053 (92 %) UK Biobank subjects were successfully analyzed, with a median PWV of 6.8 m/s, increasing 27 % per decade of age and 6.5 % higher per 10 mmHg higher systolic blood pressure. We describe a fully automated method for measuring PWV from standard cardiac MRI localizers and a single phase contrast imaging plane. The method is robust and can be applied to routine clinical scans, and could unlock the potential of measuring PWV in large-scale clinical and population studies. All models and deployment codes are available online.

To assess the prognostic utility of coronary artery calcium (CAC) scoring and coronary computed tomography angiography (CCTA)-derived quantitative plaque metrics for predicting adverse cardiovascular outcomes. The study enrolled 2404 patients with suspected coronary artery disease (CAD) but without a prior history of CAD. All participants underwent CAC scoring and CCTA, with plaque metrics quantified using an artificial intelligence (AI)-based tool (Cleerly, Inc). Percent atheroma volume (PAV) and non-calcified plaque volume percentage (NCPV%), reflecting total plaque burden and the proportion of non-calcified plaque volume normalized to vessel volume, were evaluated. The primary endpoint was a composite of all-cause mortality and non-fatal myocardial infarction (MI). Cox proportional hazard models, adjusted for clinical risk factors and early revascularization, were employed for analysis. During a median follow-up of 7.0 years, 208 patients (8.7%) experienced the primary endpoint, including 73 cases of MI (3%). The model incorporating PAV demonstrated superior discriminatory power for the composite endpoint (AUC = 0.729) compared to CAC scoring (AUC = 0.706, P = 0.016). In MI prediction, PAV (AUC = 0.791) significantly outperformed CAC (AUC = 0.699, P < 0.001), with NCPV% showing the highest prognostic accuracy (AUC = 0.814, P < 0.001). AI-driven assessment of coronary plaque burden enhances prognostic accuracy for future adverse cardiovascular events, highlighting the critical role of comprehensive plaque characterization in refining risk stratification strategies.

Accurate vertebrae segmentation is crucial for modern surgical technologies, and deep learning networks provide valuable tools for this task. This study explores the application of advanced deep learning-based methods for segmenting vertebrae in computed tomography (CT) images of adolescent idiopathic scoliosis (AIS) patients. In this study, we collected a dataset of 31 samples from AIS patients, covering a wide range of spinal regions from cervical to lumbar vertebrae. High-resolution CT images were obtained for each sample, forming the basis of our segmentation analysis. We utilized 2 popular neural networks, U-Net and Attention U-Net, to segment the vertebrae in these CT images. Segmentation performance was rigorously evaluated using 2 key metrics: the Dice Coefficient Score to measure overlap between segmented and ground truth regions, and the Hausdorff distance (HD) to assess boundary dissimilarity. Both networks performed well, with U-Net achieving an average Dice coefficient of 92.2 ± 2.4% and an HD of 9.80 ± 1.34 mm. Attention U-Net showed similar results, with a Dice coefficient of 92.3 ± 2.9% and an HD of 8.67 ± 3.38 mm. When applied to the challenging anatomy of AIS, our findings align with literature results from advanced 3D U-Nets on healthy spines. Although no significant overall difference was observed between the 2 networks (P > .05), Attention U-Net exhibited an improved Dice coefficient (91.5 ± 0.0% vs 88.8 ± 0.1%, P = .151) and a significantly better HD (9.04 ± 4.51 vs. 13.60 ± 2.26 mm, P = .027) in critical scoliosis sites (mid-thoracic region), suggesting enhanced suitability for complex anatomy. Our study indicates that U-Net neural networks are feasible and effective for automated vertebrae segmentation in AIS patients using clinical 3D CT images. Attention U-Net demonstrated improved performance in thoracic levels, which are primary sites of scoliosis and may be more suitable for challenging anatomical regions.

Basal ganglia intracranial hemorrhage (bgICH) morphology is associated with postoperative functional outcomes. We hypothesized that bgICH spatial representation modeling could be automated for functional outcome prediction after minimally invasive surgical (MIS) evacuation. A training set of 678 computed tomography head and computed tomography angiography images from 63 patients were used to train key-point detection and instance segmentation convolutional neural network-based models for anatomic landmark identification and bgICH segmentation. Anatomic landmarks included the bilateral orbital rims at the globe's maximum diameter and the posterior-most aspect of the tentorial incisura, which were used to define a universal stereotactic reference frame across patients. Convolutional neural network models were tested using volumetric computed tomography head/computed tomography angiography scans from 45 patients who underwent MIS bgICH evacuation with recorded modified Rankin Scales within one year after surgery. bgICH volumes were highly correlated (R2 = 0.95, P < .001) between manual (median 39-mL) and automatic (median 38-mL) segmentation methods. The absolute median difference between groups was 2-mL (IQR: 1-6 mL). Median localization accuracy (distance between automated and manually designated coordinate frames) was 4 mm (IQR: 3-6). Landmark coordinates were highly correlated in the x- (medial-lateral), y- (anterior-posterior), and z-axes (rostral-caudal) for all 3 landmarks (R2 range = 0.95-0.99, P < .001 for all). Functional outcome (modified Rankin Scale 4-6) was predicted with similar model performance using automated (area under the receiver operating characteristic curve = 0.81, 95% CI: 0.67-0.94) and manually (area under the receiver operating characteristic curve = 0.84, 95% CI: 0.72-0.96) constructed spatial representation models (P = .173). Computer vision models can accurately replicate bgICH manual segmentation, stereotactic localization, and prognosticate functional outcomes after MIS bgICH evacuation.

Cone-beam computed tomography (CBCT) imaging plays a crucial role in dentistry, with automatic prediction of anatomical structures on CBCT images potentially enhancing diagnostic and planning procedures. This study aims to predict anatomical structures automatically on CBCT images using a deep learning algorithm. CBCT images from 70 patients were analyzed. Anatomical structures were annotated using a regional segmentation tool within an annotation software by two dentomaxillofacial radiologists. Each volumetric dataset comprised 405 slices, with relevant anatomical structures marked in each slice. Seventy DICOM images were converted to Nifti format, with seven reserved for testing and the remaining sixty-three used for training. The training utilized nnUNetv2 with an initial learning rate of 0.01, decreasing by 0.00001 at each epoch, and was conducted for 1000 epochs. Statistical analysis included accuracy, Dice score, precision, and recall results. The segmentation model achieved an accuracy of 0.99 for nasal fossa, maxillary sinus, nasopalatine canal, mandibular canal, foramen mentale, and foramen mandible, with corresponding Dice scores of 0.85, 0.98, 0.79, 0.73, 0.78, and 0.74, respectively. Precision values ranged from 0.73 to 0.98. Maxillary sinus segmentation exhibited the highest performance, while mandibular canal segmentation showed the lowest performance. The results demonstrate high accuracy and precision across most structures, with varying Dice scores indicating the consistency of segmentation. Overall, our segmentation model exhibits robust performance in delineating anatomical features in CBCT images, promising potential applications in dental diagnostics and treatment planning.

This study aimed to construct a novel model, Multi-Spatial Attention U-Net (MSAU-Net) by incorporating our proposed Multi-Spatial Attention (MSA) block into the U-Net for the automated segmentation of the gallbladder on CT images. The gallbladder dataset consists of CT images of retrospectively-collected 152 liver cancer patients and corresponding ground truth delineated by experienced physicians. Our proposed MSAU-Net model was transformed into two versions V1(with one Multi-Scale Feature Extraction and Fusion (MSFEF) module in each MSA block) and V2 (with two parallel MSEFE modules in each MSA blcok). The performances of V1 and V2 were evaluated and compared with four other derivatives of U-Net or state-of-the-art models quantitatively using seven commonly-used metrics, and qualitatively by comparison against experienced physicians' assessment. MSAU-Net V1 and V2 models both outperformed the comparative models across most quantitative metrics with better segmentation accuracy and boundary delineation. The optimal number of MSA was three for V1 and two for V2. Qualitative evaluations confirmed that they produced results closer to physicians' annotations. External validation revealed that MSAU-Net V2 exhibited better generalization capability. The MSAU-Net V1 and V2 both exhibited outstanding performance in gallbladder segmentation, demonstrating strong potential for clinical application. The MSA block enhances spatial information capture, improving the model's ability to segment small and complex structures with greater precision. These advantages position the MSAU-Net V1 and V2 as valuable tools for broader clinical adoption.