Accurate tumour segmentation is vital for various targeted diagnostic and therapeutic procedures for cancer, e.g., planning biopsies or tumour ablations. Manual delineation is extremely labour-intensive, requiring substantial expert time. Fully-supervised machine learning models aim to automate such localisation tasks, but require a large number of costly and often subjective 3D voxel-level labels for training. The high-variance and subjectivity in such labels impacts model generalisability, even when large datasets are available. Histopathology labels may offer more objective labels but the infeasibility of acquiring pixel-level annotations to develop tumour localisation methods based on histology remains challenging in-vivo. In this work, we propose a novel weakly-supervised semantic segmentation framework called SPARS (Self-Play Adversarial Reinforcement Learning for Segmentation), which utilises an object presence classifier, trained on a small number of image-level binary cancer presence labels, to localise cancerous regions on CT scans. Such binary labels of patient-level cancer presence can be sourced more feasibly from biopsies and histopathology reports, enabling a more objective cancer localisation on medical images. Evaluating with real patient data, we observed that SPARS yielded a mean dice score of $77.3 \pm 9.4$, which outperformed other weakly-supervised methods by large margins. This performance was comparable with recent fully-supervised methods that require voxel-level annotations. Our results demonstrate the potential of using SPARS to reduce the need for extensive human-annotated labels to detect cancer in real-world healthcare settings.

High-dose-rate brachytherapy (HDR-BT) is an integral part of treatment for locally advanced cervical cancer, requiring accurate segmentation of the high-risk clinical target volume (HR-CTV) and organs at risk (OARs) on post-applicator CT (pCT) for precise and safe dose delivery. Manual contouring, however, is time-consuming and highly variable, with challenges heightened in cervical HDR-BT due to complex anatomy and low tissue contrast. An effective auto-contouring solution could significantly enhance efficiency, consistency, and accuracy in cervical HDR-BT planning. To develop a machine learning-based approach that improves the accuracy and efficiency of HR-CTV and OAR segmentation on pCT images for cervical HDR-BT. The proposed method employs two sequential deep learning models to segment target and OARs from planning CT data. The intuitive model, a U-Net, initially segments simpler structures such as the bladder and HR-CTV, utilizing shallow features and iodine contrast agents. Building on this, the sophisticated model targets complex structures like the sigmoid, rectum, and bowel, addressing challenges from low contrast, anatomical proximity, and imaging artifacts. This model incorporates spatial information from the intuitive model and uses total variation regularization to improve segmentation smoothness by applying a penalty to changes in gradient. This dual-model approach improves accuracy and consistency in segmenting high-risk clinical target volumes and organs at risk in cervical HDR-BT. To validate the proposed method, 32 cervical cancer patients treated with tandem and ovoid (T&O) HDR brachytherapy (3-5 fractions, 115 CT images) were retrospectively selected. The method's performance was assessed using four-fold cross-validation, comparing segmentation results to manual contours across five metrics: Dice similarity coefficient (DSC), 95% Hausdorff distance (HD₉₅), mean surface distance (MSD), center-of-mass distance (CMD), and volume difference (VD). Dosimetric evaluations included D90 for HR-CTV and D2cc for OARs. The proposed method demonstrates high segmentation accuracy for HR-CTV, bladder, and rectum, achieving DSC values of 0.79 ± 0.06, 0.83 ± 0.10, and 0.76 ± 0.15, MSD values of 1.92 ± 0.77 mm, 2.24 ± 1.20 mm, and 4.18 ± 3.74 mm, and absolute VD values of 5.34 ± 4.85 cc, 17.16 ± 17.38 cc, and 18.54 ± 16.83 cc, respectively. Despite challenges in bowel and sigmoid segmentation due to poor soft tissue contrast in CT and variability in manual contouring (ground truth volumes of 128.48 ± 95.9 cc and 51.87 ± 40.67 cc), the method significantly outperforms two state-of-the-art methods on DSC, MSD, and CMD metrics (p-value < 0.05). For HR-CTV, the mean absolute D90 difference was 0.42 ± 1.17 Gy (p-value > 0.05), less than 5% of the prescription dose. Over 75% of cases showed changes within ± 0.5 Gy, and fewer than 10% exceeded ± 1 Gy. The mean and variation in structure volume and D2cc parameters between manual and segmented contours for OARs showed no significant differences (p-value > 0.05), with mean absolute D2cc differences within 0.5 Gy, except for the bladder, which exhibited higher variability (0.97 Gy). Our innovative auto-contouring method showed promising results in segmenting HR-CTV and OARs from pCT, potentially enhancing the efficiency of HDR BT cervical treatment planning. Further validation and clinical implementation are required to fully realize its clinical benefits.

Peritoneal metastasis is a key factor in the poor prognosis of advanced gastrointestinal cancer patients. Traditional radiological diagnostic faces challenges such as insufficient sensitivity. Through technologies like radiomics and deep learning, artificial intelligence can deeply analyze the tumor heterogeneity and microenvironment features in medical images, revealing markers of peritoneal metastasis and constructing high-precision predictive models. These technologies have demonstrated advantages in tasks such as predicting peritoneal metastasis, assessing the risk of peritoneal recurrence, and identifying small metastatic foci during surgery. This paper summarizes the representative progress and application prospects of medical imaging artificial intelligence in the diagnosis and treatment of peritoneal metastasis, and discusses potential development directions such as multimodal data fusion and large model. The integration of medical imaging artificial intelligence with clinical practice is expected to advance personalized and precision medicine in the diagnosis and treatment of peritoneal metastasis in gastrointestinal cancers.

Pulmonary embolism (PE) is a life-threatening clinical problem where early diagnosis and prompt treatment are essential to reducing morbidity and mortality. While the combination of CT images and electronic health records (EHR) can help improve computer-aided diagnosis, there are many challenges that need to be addressed. The primary objective of this study is to leverage both 3D CT images and EHR data to improve PE diagnosis. First, for 3D CT images, we propose a network combining Swin Transformers with 3D CNNs, enhanced by a Multi-Scale Feature Fusion (MSFF) module to address fusion challenges between different encoders. Secondly, we introduce a Polarized Self-Attention (PSA) module to enhance the attention mechanism within the 3D CNN. And then, for EHR data, we design the Tabular Transformer for effective feature extraction. Finally, we design and evaluate three multimodal attention fusion modules to integrate CT and EHR features, selecting the most effective one for final fusion. Experimental results on the RadFusion dataset demonstrate that our model significantly outperforms existing state-of-the-art methods, achieving an AUROC of 0.971, an F1 score of 0.926, and an accuracy of 0.920. These results underscore the effectiveness and innovation of our multimodal approach in advancing PE diagnosis.

Accurate and interpretable age estimation and gender classification are essential in forensic and clinical diagnostics, particularly when using high-dimensional medical imaging data such as Cone Beam Computed Tomography (CBCT). Traditional CBCT-based approaches often suffer from high computational costs and limited interpretability, reducing their applicability in forensic investigations. This study aims to develop a multi-task deep learning framework that enhances both accuracy and explainability in CBCT-based age estimation and gender classification using attention mechanisms. We propose a multi-task learning (MTL) model that simultaneously estimates age and classifies gender using panoramic slices extracted from CBCT scans. To improve interpretability, we integrate Convolutional Block Attention Module (CBAM) and Grad-CAM visualization, highlighting relevant craniofacial regions. The dataset includes 2,426 CBCT images from individuals aged 7 to 23 years, and performance is assessed using Mean Absolute Error (MAE) for age estimation and accuracy for gender classification. The proposed model achieves a MAE of 1.08 years for age estimation and 95.3% accuracy in gender classification, significantly outperforming conventional CBCT-based methods. CBAM enhances the model's ability to focus on clinically relevant anatomical features, while Grad-CAM provides visual explanations, improving interpretability. Additionally, using panoramic slices instead of full 3D CBCT volumes reduces computational costs without sacrificing accuracy. Our framework improves both accuracy and interpretability in forensic age estimation and gender classification from CBCT images. By incorporating explainable AI techniques, this model provides a computationally efficient and clinically interpretable tool for forensic and medical applications.

To investigate whether the deep learning reconstruction (DLR) combined with contrast-enhancement-boost (CE-boost) technique can improve the diagnostic quality of CT pulmonary angiography (CTPA) at low radiation and contrast doses, compared with routine CTPA using hybrid iterative reconstruction (HIR). This prospective two-center study included 130 patients who underwent CTPA for suspected pulmonary embolism. Patients were randomly divided into two groups: the routine CTPA group, reconstructed using HIR; and the dual-low dose CTPA group, reconstructed using HIR and DLR, additionally combined with the CE-boost to generate HIR-boost and DLR-boost images. Signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) of pulmonary arteries were quantitatively assessed. Two experienced radiologists independently ordered CT images (5, best; 1, worst) based on overall image noise and vascular contrast. Diagnostic performance for PE detection was calculated for each dataset. Patient demographics were similar between groups. Compared to HIR images of the routine group, DLR-boost images of the dual-low dose group were significantly better at qualitative scores (p < 0.001). The CT values of pulmonary arteries between the DLR-boost and the HIR images were comparable (p > 0.05), whereas the SNRs and CNRs of pulmonary arteries in the DLR-boost images were the highest among all five datasets (p < 0.001). The AUCs of DLR, HIR-boost, and DLR-boost were 0.933, 0.924, and 0.986, respectively (all p > 0.05). DLR combined with CE-boost technique can significantly improve the image quality of CTPA with reduced radiation and contrast doses, facilitating a more accurate diagnosis of pulmonary embolism. Question The dual-low dose protocol is essential for detecting pulmonary emboli (PE) in follow-up CT pulmonary angiography (PA), yet effective solutions are still lacking. Findings Deep learning reconstruction (DLR)-boost with reduced radiation and contrast doses demonstrated higher quantitative and qualitative image quality than hybrid-iterative reconstruction in the routine CTPA. Clinical relevance DLR-boost based low-radiation and low-contrast-dose CTPA protocol offers a novel strategy to further enhance the image quality and diagnosis accuracy for pulmonary embolism patients.

Atrial fibrillation (AF) is one of the primary etiologies for ischemic stroke, and it is of paramount importance to delineate the risk phenotypes among elderly AF patients and to investigate more efficacious models for predicting stroke risk. This single-center prospective cohort study collected clinical data and cardiac computed tomography angiography (CTA) images from elderly AF patients. The clinical phenotypes and left atrial appendage (LAA) radiomic phenotypes of elderly AF patients were identified through K-means clustering. The independent correlations between these phenotypes and stroke risk were subsequently analyzed. Machine learning algorithms-Logistic Regression, Naive Bayes, Support Vector Machine (SVM), Random Forest, and Extreme Gradient Boosting-were selected to develop a predictive model for stroke risk in this patient cohort. The model was assessed using the Area Under the Receiver Operating Characteristic Curve, Hosmer-Lemeshow tests, and Decision Curve Analysis. A total of 419 elderly AF patients (≥ 65 years old) were included. K-means clustering identified three clinical phenotypes: Group A (cardiac enlargement/dysfunction), Group B (normal phenotype), and Group C (metabolic/coagulation abnormalities). Stroke incidence was highest in Group A (19.3%) and Group C (14.5%) versus Group B (3.3%). Similarly, LAA radiomic phenotypes revealed elevated stroke risk in patients with enlarged LAA structure (Group B: 20.0%) and complex LAA morphology (Group C: 14.0%) compared to normal LAA (Group A: 2.9%). Among the five machine learning models, the SVM model achieved superior prediction performance (AUROC: 0.858 [95% CI: 0.830-0.887]). The stroke-risk prediction model for elderly AF patients constructed based on the SVM algorithm has strong predictive efficacy.

Identification of bony landmarks in medical images is of high importance for 3D planning in orthopaedic surgery. Automated landmark identification has the potential to optimize clinical routines and allows for the scientific analysis of large databases. To the authors' knowledge, no direct comparison of different methods for automated landmark detection on the same dataset has been published to date. We compared 3 methods for automated femoral landmark identification: an artificial neural network, a statistical shape model and a geometric approach. All methods were compared against manual measurements of two raters on the task of identifying 6 femoral landmarks on CT data or derived surface models of 202 femora. The accuracy of the methods was in the range of the manual measurements and comparable to those reported in previous studies. The geometric approach showed a significantly higher average deviation compared to the manually selected reference landmarks, while there was no statistically significant difference for the neural network and the SSM. All fully automated methods show potential for use, depending on the use case. Characteristics of the different methods, such as the input data required (raw CT/segmented bone surface models, amount of training data required) and/or the methods robustness, can be used for method selection in the individual application.

Deep learning (DL)-based methods show promise in detecting pulmonary embolism (PE) on CT pulmonary angiography (CTPA), potentially improving diagnostic accuracy and workflow efficiency. This meta-analysis aimed to (1) determine pooled performance estimates of DL algorithms for PE detection; and (2) compare the diagnostic efficacy of convolutional neural network (CNN)- versus U-Net-based architectures. Following PRISMA guidelines, we searched PubMed and EMBASE through April 15, 2025 for English-language studies (2010-2025) reporting DL models for PE detection with extractable 2 × 2 data or performance metrics. True/false positives and negatives were reconstructed when necessary under an assumed 50 % PE prevalence (with 0.5 continuity correction). We approximated AUROC as the mean of sensitivity and specificity if not directly reported. Sensitivity, specificity, accuracy, PPV and NPV were pooled using a DerSimonian-Laird random-effects model with Freeman-Tukey transformation; AUROC values were combined via a fixed-effect inverse-variance approach. Heterogeneity was assessed by Cochran's Q and I². Subgroup analyses contrasted CNN versus U-Net models. Twenty-four studies (n = 22,984 patients) met inclusion criteria. Pooled estimates were: AUROC 0.895 (95 % CI: 0.874-0.917), sensitivity 0.894 (0.856-0.923), specificity 0.871 (0.831-0.903), accuracy 0.857 (0.833-0.882), PPV 0.832 (0.794-0.869) and NPV 0.902 (0.874-0.929). Between-study heterogeneity was high (I² ≈ 97 % for sensitivity/specificity). U-Net models exhibited higher sensitivity (0.899 vs 0.893) and CNN models higher specificity (0.926 vs 0.900); subgroup Q-tests confirmed significant differences for both sensitivity (p = 0.0002) and specificity (p < 0.001). DL algorithms demonstrate high diagnostic accuracy for PE detection on CTPA, with complementary strengths: U-Net architectures excel in true-positive identification, whereas CNNs yield fewer false positives. However, marked heterogeneity underscores the need for standardized, prospective validation before routine clinical implementation.

Purpose: To evaluate the impact of harmonization and multi-region CT image feature integration on survival prediction in non-small cell lung cancer (NSCLC) patients, using handcrafted radiomics, pretrained foundation model (FM) features, and clinical data from a multicenter dataset. Methods: We analyzed CT scans and clinical data from 876 NSCLC patients (604 training, 272 test) across five centers. Features were extracted from the whole lung, tumor, mediastinal nodes, coronary arteries, and coronary artery calcium (CAC). Handcrafted radiomics and FM deep features were harmonized using ComBat, reconstruction kernel normalization (RKN), and RKN+ComBat. Regularized Cox models predicted overall survival; performance was assessed using the concordance index (C-index), 5-year time-dependent area under the curve (t-AUC), and hazard ratio (HR). SHapley Additive exPlanations (SHAP) values explained feature contributions. A consensus model used agreement across top region of interest (ROI) models to stratify patient risk. Results: TNM staging showed prognostic utility (C-index = 0.67; HR = 2.70; t-AUC = 0.85). The clinical + tumor radiomics model with ComBat achieved a C-index of 0.7552 and t-AUC of 0.8820. FM features (50-voxel cubes) combined with clinical data yielded the highest performance (C-index = 0.7616; t-AUC = 0.8866). An ensemble of all ROIs and FM features reached a C-index of 0.7142 and t-AUC of 0.7885. The consensus model, covering 78% of valid test cases, achieved a t-AUC of 0.92, sensitivity of 97.6%, and specificity of 66.7%. Conclusion: Harmonization and multi-region feature integration improve survival prediction in multicenter NSCLC data. Combining interpretable radiomics, FM features, and consensus modeling enables robust risk stratification across imaging centers.