Subspecialty knowledge barriers have limited the adoption of large language models (LLMs) in oncology. We introduce Woollie, an open-source, oncology-specific LLM trained on real-world data from Memorial Sloan Kettering Cancer Center (MSK) across lung, breast, prostate, pancreatic, and colorectal cancers, with external validation using University of California, San Francisco (UCSF) data. Woollie surpasses ChatGPT in medical benchmarks and excels in eight non-medical benchmarks. Analyzing 39,319 radiology impression notes from 4002 patients, it achieved an overall area under the receiver operating characteristic curve (AUROC) of 0.97 for cancer progression prediction on MSK data, including a notable 0.98 AUROC for pancreatic cancer. On UCSF data, it achieved an overall AUROC of 0.88, excelling in lung cancer detection with an AUROC of 0.95. As the first oncology specific LLM validated across institutions, Woollie demonstrates high accuracy and consistency across cancer types, underscoring its potential to enhance cancer progression analysis.

Sudden cardiac death from ventricular arrhythmias is a leading cause of mortality worldwide. Arrhythmic death prognostication is challenging in patients with hypertrophic cardiomyopathy (HCM), a setting where current clinical guidelines show low performance and inconsistent accuracy. Here, we present a deep learning approach, MAARS (Multimodal Artificial intelligence for ventricular Arrhythmia Risk Stratification), to forecast lethal arrhythmia events in patients with HCM by analyzing multimodal medical data. MAARS' transformer-based neural networks learn from electronic health records, echocardiogram and radiology reports, and contrast-enhanced cardiac magnetic resonance images, the latter being a unique feature of this model. MAARS achieves an area under the curve of 0.89 (95% confidence interval (CI) 0.79-0.94) and 0.81 (95% CI 0.69-0.93) in internal and external cohorts and outperforms current clinical guidelines by 0.27-0.35 (internal) and 0.22-0.30 (external). In contrast to clinical guidelines, it demonstrates fairness across demographic subgroups. We interpret MAARS' predictions on multiple levels to promote artificial intelligence transparency and derive risk factors warranting further investigation.

To evaluate the effectiveness of deep learning radiomics nomogram in distinguishing early intracranial hypertension (IH) following primary decompressive craniectomy (DC) in patients with severe traumatic brain injury (TBI) and to demonstrate its potential clinical value as a noninvasive tool for guiding timely intervention and improving patient outcomes. This study included 238 patients with severe TBI (training cohort: n = 166; testing cohort: n = 72). Postoperative ultrasound images of the optic nerve sheath (ONS) and Spectral doppler imaging of middle cerebral artery (MCASDI) were obtained at 6 and 18 h after DC. Patients were grouped according to threshold values of 15 mmHg and 20 mmHg based on invasive intracranial pressure (ICPi) measurements. Clinical-semantic features were collected, and radiomics features were extracted from ONS images, and Additionally, deep transfer learning (DTL) features were generated using RseNet101. Predictive models were developed using the Light Gradient Boosting Machine (light GBM) machine learning algorithm. Clinical-ultrasound variables were incorporated into the model through univariate and multivariate logistic regression. A combined nomogram was developed by integrating DLR (deep learning radiomics) features with clinical-ultrasound variables, and its diagnostic performance over different thresholds was evaluated using Receiver Operating Characteristic (ROC) curve analysis and decision curve analysis (DCA). The nomogram model demonstrated superior performance over the clinical model at both 15 mmHg and 20 mmHg thresholds. For 15 mmHg, the AUC was 0.974 (95% confidence interval [CI]: 0.953-0.995) in the training cohort and 0.919 (95% CI: 0.845-0.993) in the testing cohort. For 20 mmHg, the AUC was 0.968 (95% CI: 0.944-0.993) in the training cohort and 0.889 (95% CI: 0.806-0.972) in the testing cohort. DCA curves showed net clinical benefit across all models. Among DLR models based on ONS, MCASDI, or their pre-fusion, the ONS-based model performed best in the testing cohorts. The nomogram model, incorporating clinical-semantic features, radiomics, and DTL features, exhibited promising performance in predicting early IH in post-DC patients. It shows promise for enhancing non-invasive ICP monitoring and supporting individualized therapeutic strategies.

Delineation of multiple organs in murine µCT images is crucial for preclinical studies but requires manual volumetric segmentation, a tedious and time-consuming process prone to inter-observer variability. Automatic deep learning-based segmentation can improve speed and reproducibility. While 2D and 3D deep learning models have been developed for anatomical segmentation, their generalization to external datasets has not been extensively investigated. Furthermore, ensemble learning, combining predictions of multiple 2D models, and partially-supervised learning (PSL), enabling training on partially-labeled datasets, have not been explored for preclinical purposes. This study demonstrates the first use of PSL frameworks and the superiority of 3D models in accuracy and generalizability to external datasets. Ensemble methods performed on par or better than the best individual 2D network, but only 3D models consistently generalized to external datasets (Dice Similarity Coefficient (DSC) > 0.8). PSL frameworks showed promising results across various datasets and organs, but its generalization to external data can be improved for some organs. This work highlights the superiority of 3D models over 2D and ensemble counterparts in accuracy and generalizability for murine µCT image segmentation. Additionally, a promising PSL framework is presented for leveraging multiple datasets without complete annotations. Our model can increase time-efficiency and improve reproducibility in preclinical radiotherapy workflows by circumventing manual contouring bottlenecks. Moreover, high segmentation accuracy of 3D models allows monitoring multiple organs over time using repeated µCT imaging, potentially reducing the number of mice sacrificed in studies, adhering to the 3R principle, specifically Reduction and Refinement.

The detection of brain tumors is crucial in medical imaging, because accurate and early diagnosis can have a positive effect on patients. Because traditional deep learning models store all their data together, they raise questions about privacy, complying with regulations and the different types of data used by various institutions. We introduce the anisotropic-residual capsule hybrid Gorilla Badger optimized network (Aniso-ResCapHGBO-Net) framework for detecting brain tumors in a privacy-preserving, decentralized system used by many healthcare institutions. ResNet-50 and capsule networks are incorporated to achieve better feature extraction and maintain the structure of images' spatial data. To get the best results, the hybrid Gorilla Badger optimization algorithm (HGBOA) is applied for selecting the key features. Preprocessing techniques include anisotropic diffusion filtering, morphological operations, and mutual information-based image registration. Updates to the model are made secure and tamper-evident on the Ethereum network with its private blockchain and SHA-256 hashing scheme. The project is built using Python, TensorFlow and PyTorch. The model displays 99.07% accuracy, 98.54% precision and 99.82% sensitivity on assessments from benchmark CT imaging of brain tumors. This approach also helps to reduce the number of cases where no disease is found when there is one and vice versa. The framework ensures that patients' data is protected and does not decrease the accuracy of brain tumor detection.

This study aimed at investigating a novel technique for designing and manufacturing a sealing socket abutment (SSA) using artificial intelligence (AI)-driven tooth segmentation and 3D printing technologies. A validated AI-powered module was used to segment the tooth to be replaced on the presurgical Cone Beam Computed Tomography (CBCT) scan. Following virtual surgical planning, the CBCT and intraoral scan (IOS) were imported into Mimics software. The AI-segmented tooth was aligned with the IOS, sliced horizontally at the temporary abutment's neck, and further trimmed 2 mm above the gingival margin to capture the emergence profile. A conical cut, 2 mm wider than the temporary abutment with a 5° taper, was applied for a passive fit. This process produced a custom sealing socket abutment, which was then 3D-printed. After atraumatic tooth extraction and immediate implant placement, the temporary abutment was positioned, followed by the SealPrint atop. A flowable composite was used to fill the gap between the temporary abutment and the SealPrint; the whole structure sealing the extraction socket, providing by design support for the interdental papilla and protecting the implant and (bio)materials used. True to planning, the SealPrint passively fits on the temporary abutment. It provides an optimal seal over the entire surface of the extraction socket, preserving the emergence profile of the extracted tooth, protecting the dental implant and stabilizing the graft material and blood clot. The SealPrint technique provides a reliable and fast solution for protection and preservation of the soft-, hard-tissues and emergence profile following immediate implant placement.

Accurate segmentation of white matter hyperintensities (WMH) is crucial for clinical decision-making, particularly in the context of multiple sclerosis. However, domain shifts, such as variations in MRI machine types or acquisition parameters, pose significant challenges to model calibration and uncertainty estimation. This comparative study investigates the impact of domain shift on WMH segmentation, proposing maximum-entropy regularization techniques to enhance model calibration and uncertainty estimation. The purpose is to identify errors appearing after model deployment in clinical scenarios using predictive uncertainty as a proxy measure, since it does not require ground-truth labels to be computed. We conducted experiments using a classic U-Net architecture and evaluated maximum entropy regularization schemes to improve model calibration under domain shift on two publicly available datasets: the WMH Segmentation Challenge and the 3D-MR-MS dataset. Performance is assessed with Dice coefficient, Hausdorff distance, expected calibration error, and entropy-based uncertainty estimates. Entropy-based uncertainty estimates can anticipate segmentation errors, both in-distribution and out-of-distribution, with maximum-entropy regularization further strengthening the correlation between uncertainty and segmentation performance, while also improving model calibration under domain shift. Maximum-entropy regularization improves uncertainty estimation for WMH segmentation under domain shift. By strengthening the relationship between predictive uncertainty and segmentation errors, these methods allow models to better flag unreliable predictions without requiring ground-truth annotations. Additionally, maximum-entropy regularization contributes to better model calibration, supporting more reliable and safer deployment of deep learning models in multi-center and heterogeneous clinical environments.

We present the first comparative study of two fundamentally distinct feature extraction techniques: Histogram of Oriented Gradients (HOG) and Topological Data Analysis (TDA), for medical image classification using retinal fundus images. HOG captures local texture and edge patterns through gradient orientation histograms, while TDA, using cubical persistent homology, extracts high-level topological signatures that reflect the global structure of pixel intensities. We evaluate both methods on the large APTOS dataset for two classification tasks: binary detection (normal versus diabetic retinopathy) and five-class diabetic retinopathy severity grading. From each image, we extract 26244 HOG features and 800 TDA features, using them independently to train seven classical machine learning models with 10-fold cross-validation. XGBoost achieved the best performance in both cases: 94.29 percent accuracy (HOG) and 94.18 percent (TDA) on the binary task; 74.41 percent (HOG) and 74.69 percent (TDA) on the multi-class task. Our results show that both methods offer competitive performance but encode different structural aspects of the images. This is the first work to benchmark gradient-based and topological features on retinal imagery. The techniques are interpretable, applicable to other medical imaging domains, and suitable for integration into deep learning pipelines.

IntroductionThe urethra is a recommended avoidance structure for prostate cancer treatment. However, even subspecialist physicians often struggle to accurately identify the urethra on available imaging. Automated segmentation tools show promise, but a lack of reliable ground truth or appropriate evaluation standards has hindered validation and clinical adoption. This study aims to establish a reference-standard dataset with expert consensus contours, define clinically meaningful evaluation metrics, and assess the performance and generalizability of a deep-learning-based segmentation model. Materials and MethodsA multidisciplinary panel of four experienced subspecialists in prostate MRI generated consensus contours of the male urethra for 71 patients across six imaging centers. Four of those cases were previously used in an international study (PURE-MRI), wherein 62 physicians attempted to contour the prostate and urethra on the patient images. Separately, we developed a deep-learning AI model for urethra segmentation using another 151 cases from one center and evaluated it against the consensus reference standard and compared to human performance using Dice Score, percent urethra Coverage, and Maximum 2D (axial, in-plane) Hausdorff Distance (HD) from the reference standard. ResultsIn the PURE-MRI dataset, the AI model outperformed most physicians, achieving a median Dice of 0.41 (vs. 0.33 for physicians), Coverage of 81% (vs. 36%), and Max 2D HD of 1.8 mm (vs. 1.6 mm). In the larger dataset, performance remained consistent, with a Dice of 0.40, Coverage of 89%, and Max 2D HD of 2.0 mm, indicating strong generalizability across a broader patient population and more varied imaging conditions. ConclusionWe established a multidisciplinary consensus benchmark for segmentation of the urethra. The deep-learning model performs comparably to specialist physicians and demonstrates consistent results across multiple institutions. It shows promise as a clinical decision-support tool for accurate and reliable urethra segmentation in prostate cancer radiotherapy planning and studies of dose-toxicity associations.

To quantitatively characterize the degree of perirenal fat adhesion using artificial intelligence in renal cell carcinoma. This retrospective study analyzed a total of 596 patients from three cohorts, utilizing corticomedullary phase computed tomography urography (CTU) images. The nnUNet v2 network combined with numerical computation was employed to segment the perirenal fat region. Pyradiomics algorithms and a computed tomography foundation model were used to extract features from CTU images separately, creating single-modality predictive models for identifying perirenal fat adhesion. By concatenating the Pyradiomics and foundation model features, an early fusion multimodal predictive signature was developed. The prognostic performance of the single-modality and multimodality models was further validated in two independent cohorts. The nnUNet v2 segmentation model accurately segmented both kidneys. The neural network and thresholding approach effectively delineated the perirenal fat region. Single-modality models based on radiomic and computed tomography foundation features demonstrated a certain degree of accuracy in diagnosing and identifying perirenal fat adhesion, while the early feature fusion diagnostic model outperformed the single-modality models. Also, the perirenal fat adhesion score showed a positive correlation with surgical time and intraoperative blood loss. AI-based radiomics and foundation models can accurately identify the degree of perirenal fat adhesion and have the potential to be used for surgical risk assessment.