Pancreatic cancer treatment plans involving surgery and/or chemotherapy are highly dependent on disease stage. However, current staging systems are ineffective and poorly correlated with survival outcomes. We investigate how artificial intelligence (AI) can enhance prognostic accuracy in pancreatic cancer by integrating multiple data sources. Patients with histopathology and/or radiology/follow-up confirmed pancreatic ductal adenocarcinoma (PDAC) from a Dutch center (2004-2023) were included in the development cohort. Two additional PDAC cohorts from a Dutch and Spanish center were used for external validation. Prognostic models including clinical variables, contrast-enhanced CT images, and a combination of both were developed to predict high-risk short-term survival. All models were trained using five-fold cross-validation and assessed by the area under the time-dependent receiver operating characteristic curve (AUC). The models were developed on 401 patients (203 females, 198 males, median survival (OS) = 347 days, IQR: 171-585), with 98 (24.4%) short-term survivors (OS < 230 days) and 303 (75.6%) long-term survivors. The external validation cohorts included 361 patients (165 females, 138 males, median OS = 404 days, IQR: 173-736), with 110 (30.5%) short-term survivors and 251 (69.5%) longer survivors. The best AUC for predicting short vs. long-term survival was achieved with the multi-modal model (AUC = 0.637 (95% CI: 0.500-0.774)) in the internal validation set. External validation showed AUCs of 0.571 (95% CI: 0.453-0.689) and 0.675 (95% CI: 0.593-0.757). Multimodal AI can predict long vs. short-term survival in PDAC patients, showing potential as a prognostic tool in clinical decision-making. Question Prognostic tools for pancreatic ductal adenocarcinoma (PDAC) remain limited, with TNM staging offering suboptimal accuracy in predicting patient survival outcomes. Findings The multimodal AI model demonstrated improved prognostic performance over TNM and unimodal models for predicting short- and long-term survival in PDAC patients. Clinical relevance Multimodal AI provides enhanced prognostic accuracy compared to current staging systems, potentially improving clinical decision-making and personalized management strategies for PDAC patients.

For patients with colorectal liver metastases (CRLM), total tumor volume (TTV) is prognostic. A deep-learning segmentation model for CRLM to assess TTV called COlorectal cAncer Liver metastases Assessment (COALA) has been developed. This study evaluated COALA's performance and practical utility in the radiological picture archiving and communication system (PACS). A secondary aim was to provide lessons for future researchers on the implementation of artificial intelligence (AI) models. Patients discussed between January and December 2023 in a multidisciplinary meeting for CRLM were included. In those patients, CRLM was automatically segmented in portal-venous phase CT scans by COALA and integrated with PACS. Eight expert abdominal radiologists completed a questionnaire addressing segmentation accuracy and PACS integration. They were also asked to write down general remarks. In total, 57 patients were evaluated. Of those patients, 112 contrast-enhanced portal-venous phase CT scans were analyzed. Of eight radiologists, six (75%) evaluated the model as user-friendly in their radiological workflow. Areas of improvement of the COALA model were the segmentation of small lesions, heterogeneous lesions, and lesions at the border of the liver with involvement of the diaphragm or heart. Key lessons for implementation were a multidisciplinary approach, a robust method prior to model development and organizing evaluation sessions with end-users early in the development phase. This study demonstrates that the deep-learning segmentation model for patients with CRLM (COALA) is user-friendly in the radiologist's PACS. Future researchers striving for implementation should have a multidisciplinary approach, propose a robust methodology and involve end-users prior to model development. Many segmentation models are being developed, but none of those models are evaluated in the (radiological) workflow or clinically implemented. Our model is implemented in the radiological work system, providing valuable lessons for researchers to achieve clinical implementation. Developed segmentation models should be implemented in the radiological workflow. Our implemented segmentation model provides valuable lessons for future researchers. If implemented in clinical practice, our model could allow for objective radiological evaluation.

Accurate differentiation between benign and malignant adnexal masses is crucial for patients to avoid unnecessary surgical interventions. Ultrasound (US) is the most widely utilized diagnostic and screening tool for gynecological diseases, with contrast-enhanced US (CEUS) offering enhanced diagnostic precision by clearly delineating blood flow within lesions. According to the Ovarian and Adnexal Reporting and Data System (O-RADS), masses classified as categories 4 and 5 carry the highest risk of malignancy. However, the diagnostic accuracy of US remains heavily reliant on the expertise and subjective interpretation of radiologists. Radiomics has demonstrated significant value in tumor differential diagnosis by extracting microscopic information imperceptible to the human eye. Despite this, no studies to date have explored the application of CEUS-based radiomics for differentiating adnexal masses. This study aims to develop and validate a multimodal US-based nomogram that integrates clinical variables, radiomics, and deep learning (DL) features to effectively distinguish adnexal masses classified as O-RADS 4-5. From November 2020 to March 2024, we enrolled 340 patients who underwent two-dimensional US (2DUS) and CEUS and had masses categorized as O-RADS 4-5. These patients were randomly divided into a training cohort and a test cohort in a 7:3 ratio. Adnexal masses were manually segmented from 2DUS and CEUS images. Using machine learning (ML) and DL techniques, five models were developed and validated to differentiate adnexal masses. The diagnostic performance of these models was evaluated using the area under the receiver operating characteristic (ROC) curve (AUC), accuracy, sensitivity, specificity, precision, and F1-score. Additionally, a nomogram was constructed to visualize outcome measures. The CEUS-based radiomics model outperformed the 2DUS model (AUC: 0.826 vs. 0.737). Similarly, the CEUS-based DL model surpassed the 2DUS model (AUC: 0.823 vs. 0.793). The ensemble model combining clinical variables, radiomics, and DL features achieved the highest AUC (0.929). Our study confirms the effectiveness of CEUS-based radiomics for distinguishing adnexal masses with high accuracy and specificity using a multimodal US-based radiomics DL nomogram. This approach holds significant promise for improving the diagnostic precision of adnexal masses classified as O-RADS 4-5.

To create and validate a machine learning (ML)-based model for predicting the adverse perinatal outcome (APO) in foetal growth restriction (FGR) at diagnosis. A retrospective study. Multi-centre in China. Pregnancies affected by FGR. We enrolled singleton foetuses with a perinatal diagnosis of FGR who were admitted between January 2021 and November 2023. A total of 361 pregnancies from Beijing Obstetrics and Gynecology Hospital were used as the training set and the internal test set. In comparison, data from 50 pregnancies from Haidian Maternal and Child Health Hospital were used as the external test set. Feature screening was performed using the random forest (RF), the Least Absolute Shrinkage and Selection Operator (LASSO) and logistic regression (LR). Subsequently, six ML methods, including Stacking, were used to construct models to predict the APO of FGR. Model's performance was evaluated through indicators such as the area under the receiver operating characteristic curve (AUROC). The Shapley Additive Explanation analysis was used to rank each model feature and explain the final model. Mean ± SD gestational age at diagnosis was 32.3 ± 4.8 weeks in the absent APO group and 27.3 ± 3.7 in the present APO group. Women enrolled in the present APO group had a higher rate of hypertension related to pregnancy (74.8% vs. 18.8%, p < 0.001). Among 17 candidate predictors (including maternal characteristics, maternal comorbidities, obstetric characteristics and ultrasound parameters), the integration of RF, LASSO and LR methodologies identified maternal body mass index, hypertension, gestational age at diagnosis of FGR, estimated foetal weight (EFW) z score, EFW growth velocity and abnormal umbilical artery Doppler (defined as a pulsatility index above the 95th percentile or instances of absent/reversed diastolic flow) as significant predictors. The Stacking model demonstrated a good performance in both the internal test set [AUROC: 0.861, 95% confidence interval (CI), 0.838-0.896] and the external test set [AUROC: 0.906, 95% CI, 0.875-0.947]. The calibration curve showed high agreement between the predicted and observed risks. The Hosmer-Lemeshow test for the internal and external test sets was p = 0.387 and p = 0.825, respectively. The ML algorithm for APO, which integrates maternal clinical factors and ultrasound parameters, demonstrates good predictive value for APO in FGR at diagnosis. This suggested that ML techniques may be a valid approach for the early detection of high-risk APO in FGR pregnancies.

Prostate cancer (PCa) is the most commonly diagnosed malignancy in men worldwide. Accurate triage of patients based on tumor aggressiveness and staging is critical for selecting appropriate management pathways. While magnetic resonance imaging (MRI) has become a mainstay in PCa diagnosis, most predictive models rely on multiparametric imaging or invasive inputs, limiting generalizability in real-world clinical settings. This study aimed to develop and validate machine learning (ML) models using radiomic features extracted from T2-weighted MRI--alone and in combination with clinical variables--to predict ISUP grade (tumor aggressiveness), lymph node involvement (cN) and distant metastasis (cM). A retrospective multicenter cohort from three European sites in the Chaimeleon project was analyzed. Radiomic features were extracted from prostate zone segmentations and lesion masks, following standardized preprocessing and ComBat harmonization. Feature selection and model optimization were performed using nested cross-validation and Bayesian tuning. Hybrid models were trained using XGBoost and interpreted with SHAP values. The ISUP model achieved an AUC of 0.66, while the cN and cM models reached AUCs of 0.77 and 0.80, respectively. The best-performing models consistently combined prostate zone radiomics with clinical features such as PSA, PIRADSv2 and ISUP grade. SHAP analysis confirmed the importance of both clinical and texture-based radiomic features, with entropy and non-uniformity measures playing central roles in all tasks. Our results demonstrate the feasibility of using T2-weighted MRI and zonal radiomics for robust prediction of aggressiveness, nodal involvement and distant metastasis in PCa. This fully automated pipeline offers an interpretable, accessible and clinically translatable tool for first-line PCa triage, with potential integration into real-world diagnostic workflows.

Synthetic CT (sCT) generation from cone-beam CT (CBCT) must maintain stable performance and allow for accurate dose calculation across all treatment fractions to effectively support adaptive proton therapy. This study evaluated a 3D deep-learning (DL) network for sCT generation for prostate cancer patients over the full treatment course. Patient data from 25/6 prostate cancer patients were used to train/test the DL network. Patients in the test set had a planning CT, 39 CBCT images, and at least one repeat CT (reCT) used for replanning. The generated sCT images were compared to fan-beam planning and reCT images in terms of i) CT number accuracy and stability within spherical regions-of-interest (ROIs) in the bladder, prostate, and femoral heads, ii) proton range calculation accuracy through single-spot plans, and iii) dose trends in target coverage over the treatment course (one patient). The sCT images demonstrated image quality comparable to CT, while preserving the CBCT anatomy. The mean CT numbers on the sCT and CT images were comparable, e.g. for the prostate ROI they ranged from 29 HU to 59 HU for sCT, and from 36 HU to 50 HU for CT. The largest median proton range difference was 1.9 mm. Proton dose calculations showed excellent target coverage (V95%≥99.6 %) for the high-dose target. The DL network effectively generated high-quality sCT images with CT numbers, proton range, and dose characteristics comparable to fan-beam CT. Its robustness against intra-patient variations makes it a feasible tool for adaptive proton therapy.

Our objective is to develop and validate a deep learning radiomics nomogram (DLRN) based on preoperative ultrasound images and clinical features, for predicting the malignancy of thyroid nodules with indeterminate cytology (Bethesda III). Between June 2017 and June 2022, we conducted a retrospective study on 194 patients with surgically confirmed indeterminate cytology (Bethesda III) in our hospital. The training and internal validation cohorts were comprised of 155 and 39 patients, in a 7:3 ratio. To facilitate external validation, we selected an additional 80 patients from each of the remaining two medical centers. Utilizing preoperative ultrasound data, we obtained imaging markers that encompass both deep learning and manually radiomic features. After feature selection, we developed a comprehensive diagnostic model to evaluate the predictive value for Bethesda III benign and malignant cases. The model's diagnostic accuracy, calibration, and clinical applicability were systematically assessed. The results showed that the prediction model, which integrated 512 DTL features extracted from the pre-trained Resnet34 network, ultrasound radiomics, and clinical features, exhibited superior stability in distinguishing between benign and malignant indeterminate thyroid nodules (Bethesda Class III). In the validation set, the AUC was 0.92 (95% CI: 0.831-1.000), and the accuracy, sensitivity, specificity, precision, and recall were 0.897, 0.882, 0.909, 0.882, and 0.882, respectively. The comprehensive multidimensional data model based on deep transfer learning, ultrasound radiomics features, and clinical characteristics can effectively distinguish the benign and malignant indeterminate thyroid nodules (Bethesda Class III), providing valuable guidance for treatment selection in patients with indeterminate thyroid nodules (Bethesda Class III).

<i>"Just Accepted" papers have undergone full peer review and have been accepted for publication in <i>Radiology: Artificial Intelligence</i>. This article will undergo copyediting, layout, and proof review before it is published in its final version. Please note that during production of the final copyedited article, errors may be discovered which could affect the content.</i> Purpose To develop a deep learning segmentation model that can segment abdominal organs on CT and MR images with high accuracy and generalization ability. Materials and Methods In this study, an extended nnU-Net model was trained for abdominal organ segmentation. A domain randomization method in both the image and feature space was developed to improve the generalization ability under cross-site and cross-modality settings on public prostate MRI and abdominal CT and MRI datasets. The prostate MRI dataset contains data from multiple health care institutions with domain shifts. The abdominal CT and MRI dataset is structured for cross-modality evaluation, training on one modality (eg, MRI) and testing on the other (eg, CT). This domain randomization method was then used to train a segmentation model with enhanced generalization ability on the abdominal multiorgan segmentation challenge (AMOS) dataset to improve abdominal CT and MR multiorgan segmentation, and the model was compared with two commonly used segmentation algorithms (TotalSegmentator and MRSegmentator). Model performance was evaluated using the Dice similarity coefficient (DSC). Results The proposed domain randomization method showed improved generalization ability on the cross-site and cross-modality datasets compared with the state-of-the-art methods. The segmentation model using this method outperformed two other publicly available segmentation models on data from unseen test domains (Average DSC: 0.88 versus 0.79; <i>P</i> < .001 and 0.88 versus 0.76; <i>P</i> < .001). Conclusion The combination of image and feature domain randomizations improved the accuracy and generalization ability of deep learning-based abdominal segmentation on CT and MR images. © RSNA, 2025.

Liver cancer, especially hepatocellular carcinoma (HCC), remains one of the most fatal cancers globally, emphasizing the critical need for accurate tumor segmentation to enable timely diagnosis and effective treatment planning. Traditional imaging techniques, such as CT and MRI, rely on manual interpretation, which can be both time-intensive and subject to variability. This study introduces FasNet, an innovative hybrid deep learning model that combines ResNet-50 and VGG-16 architectures, incorporating Channel and Spatial Attention mechanisms alongside Monte Carlo Dropout to improve segmentation precision and reliability. FasNet leverages ResNet-50's robust feature extraction and VGG-16's detailed spatial feature capture to deliver superior liver tumor segmentation accuracy. Channel and spatial attention mechanisms could selectively focus on the most relevant features and spatial regions for suitable segmentation with good accuracy and reliability. Monte Carlo Dropout estimates uncertainty and adds robustness, which is critical for high-stakes medical applications. Tested on the LiTS17 dataset, FasNet achieved a Dice Coefficient of 0.8766 and a Jaccard Index of 0.8487, surpassing several state-of-the-art methods. The Channel and Spatial Attention mechanisms in FasNet enhance feature selection, focusing on the most relevant spatial and channel information, while Monte Carlo Dropout improves model robustness and uncertainty estimation. These results position FasNet as a powerful diagnostic tool, offering precise and automated liver tumor segmentation that aids in early detection and precise treatment, ultimately enhancing patient outcomes.

Deep learning methods provide enormous promise in automating manually intense tasks such as medical image segmentation and provide workflow assistance to clinical experts. Deep neural networks (DNN) require a significant amount of training examples and a variety of expert opinions to capture the nuances and the context, a challenging proposition in oncological studies (H. Wang et al., Nature, vol. 620, no. 7972, pp. 47-60, Aug 2023). Inter-reader variability among clinical experts is a real-world problem that severely impacts the generalization of DNN reproducibility. This study proposes quantifying the variability in DNN performance using expert opinions and exploring strategies to train the network and adapt between expert opinions. We address the inter-reader variability problem in the context of prostate gland segmentation using a well-studied DNN, the 3D U-Net model. Reference data includes magnetic resonance imaging (MRI, T2-weighted) with prostate glandular anatomy annotations from two expert readers (R#1, n = 342 and R#2, n = 204). 3D U-Net was trained and tested with individual expert examples (R#1 and R#2) and had an average Dice coefficient of 0.825 (CI, [0.81 0.84]) and 0.85 (CI, [0.82 0.88]), respectively. Combined training with a representative cohort proportion (R#1, n = 100 and R#2, n = 150) yielded enhanced model reproducibility across readers, achieving an average test Dice coefficient of 0.863 (CI, [0.85 0.87]) for R#1 and 0.869 (CI, [0.87 0.88]) for R#2. We re-evaluated the model performance across the gland volumes (large, small) and found improved performance for large gland size with an average Dice coefficient to be at 0.846 [CI, 0.82 0.87] and 0.872 [CI, 0.86 0.89] for R#1 and R#2, respectively, estimated using fivefold cross-validation. Performance for small gland sizes diminished with average Dice of 0.8 [0.79, 0.82] and 0.8 [0.79, 0.83] for R#1 and R#2, respectively.