Purpose: Accurate segmentation of clinical target volumes (CTV) and organs-at-risk is crucial for optimizing gynecologic brachytherapy (GYN-BT) treatment planning. However, anatomical variability, low soft-tissue contrast in CT imaging, and limited annotated datasets pose significant challenges. This study presents GynBTNet, a novel multi-stage learning framework designed to enhance segmentation performance through self-supervised pretraining and hierarchical fine-tuning strategies. Methods: GynBTNet employs a three-stage training strategy: (1) self-supervised pretraining on large-scale CT datasets using sparse submanifold convolution to capture robust anatomical representations, (2) supervised fine-tuning on a comprehensive multi-organ segmentation dataset to refine feature extraction, and (3) task-specific fine-tuning on a dedicated GYN-BT dataset to optimize segmentation performance for clinical applications. The model was evaluated against state-of-the-art methods using the Dice Similarity Coefficient (DSC), 95th percentile Hausdorff Distance (HD95), and Average Surface Distance (ASD). Results: Our GynBTNet achieved superior segmentation performance, significantly outperforming nnU-Net and Swin-UNETR. Notably, it yielded a DSC of 0.837 +/- 0.068 for CTV, 0.940 +/- 0.052 for the bladder, 0.842 +/- 0.070 for the rectum, and 0.871 +/- 0.047 for the uterus, with reduced HD95 and ASD compared to baseline models. Self-supervised pretraining led to consistent performance improvements, particularly for structures with complex boundaries. However, segmentation of the sigmoid colon remained challenging, likely due to anatomical ambiguities and inter-patient variability. Statistical significance analysis confirmed that GynBTNet's improvements were significant compared to baseline models.

The accelerating development of general medical artificial intelligence (GMAI), powered by multimodal large language models (MLLMs), offers transformative potential for addressing persistent healthcare challenges, including workforce deficits and escalating costs. The parallel development of systematic evaluation benchmarks emerges as a critical imperative to enable performance assessment and provide technological guidance. Meanwhile, as an invaluable knowledge source, the potential of medical textbooks for benchmark development remains underexploited. Here, we present MedBookVQA, a systematic and comprehensive multimodal benchmark derived from open-access medical textbooks. To curate this benchmark, we propose a standardized pipeline for automated extraction of medical figures while contextually aligning them with corresponding medical narratives. Based on this curated data, we generate 5,000 clinically relevant questions spanning modality recognition, disease classification, anatomical identification, symptom diagnosis, and surgical procedures. A multi-tier annotation system categorizes queries through hierarchical taxonomies encompassing medical imaging modalities (42 categories), body anatomies (125 structures), and clinical specialties (31 departments), enabling nuanced analysis across medical subdomains. We evaluate a wide array of MLLMs, including proprietary, open-sourced, medical, and reasoning models, revealing significant performance disparities across task types and model categories. Our findings highlight critical capability gaps in current GMAI systems while establishing textbook-derived multimodal benchmarks as essential evaluation tools. MedBookVQA establishes textbook-derived benchmarking as a critical paradigm for advancing clinical AI, exposing limitations in GMAI systems while providing anatomically structured performance metrics across specialties.

Sarcopenia, or the loss of muscle quality and quantity, has been associated with poor clinical outcomes in liver transplantation such as infection, increased length of stay, and increased patient mortality. Abdominal computed tomography (CT) scans are utilized to measure patient core musculature as a measurement of sarcopenia. Methods to extract information on core body musculature can be through either freehand region-of-interest (ROI) or machine learning algorithms to quantitate total body muscle within a given area. This study directly compares these two collection methods leveraging length of stay (LOS) outcomes previously found to be associated with freehand ROI measurements. A total of 50 individuals were included who underwent liver transplantation from our single center between January 1, 2016, and May 30, 2021, and had a non-contrast abdominal CT scan within 6-months of surgery. CT-derived skeletal muscle measures at the third lumbar vertebrae were obtained using freehand ROI and an automated deep learning system. Correlation analysis of freehand psoas muscle measures, psoas area index (PAI) and mean Hounsfield units (mHU), were significantly correlated to the automated deep learning system's total skeletal muscle measures at the level of the L3, skeletal muscle index (SMI) and skeletal muscle density (SMD), respectively (R² = 0.4221; p value < 0.0001; R² = 0.6297; p value < 0.0001). The automated deep learning model's SMI predicted ∼20% of the variability (R² = 0.2013; hospital length of stay) while the PAI variable only predicted about 10% of the variability (R² = 0.0919; total healthcare length of stay) of the length of stay variables. In contrast, both the freehand ROI mHU and the automated deep learning model's muscle density variables were associated with ∼20% of the variability in the inpatient length of stay (R² = 0.2383 and 0.1810, respectively) and total healthcare length of stay variables (R² = 0.2190 and 0.1947, respectively). Sarcopenia measurements represent an important risk stratification tool for liver transplantation outcomes. For muscle sarcopenia assessment association with LOS, freehand measures of sarcopenia perform similarly to automated deep learning system measurements.

The clinical utility of body composition in predicting the severity of acute pancreatitis (AP) remains unclear. We aimed to measure body composition using artificial intelligence (AI) to predict severe AP in hospitalized patients. We performed a retrospective study of patients hospitalized with AP at three tertiary care centers in 2018. Patients with computer tomography (CT) imaging of the abdomen at admission were included. A fully automated and validated abdominal segmentation algorithm was used for body composition analysis. The primary outcome was severe AP, defined as having persistent single- or multi-organ failure as per the revised Atlanta classification. 352 patients were included. Severe AP occurred in 35 patients (9.9%). In multivariable analysis, adjusting for male sex and first episode of AP, intermuscular adipose tissue (IMAT) was associated with severe AP, OR = 1.06 per 5 cm², p = 0.0207. Subcutaneous adipose tissue (SAT) area approached significance, OR = 1.05, p = 0.17. Neither visceral adipose tissue (VAT) nor skeletal muscle (SM) was associated with severe AP. In obese patients, a higher SM was associated with severe AP in unadjusted analysis (86.7 vs 75.1 and 70.3 cm² in moderate and mild, respectively p = 0.009). In this multi-site retrospective study using AI to measure body composition, we found elevated IMAT to be associated with severe AP. Although SAT was non-significant for severe AP, it approached statistical significance. Neither VAT nor SM were significant. Further research in larger prospective studies may be beneficial.

Accurately identifying individuals who are at high risk of lung cancer is critical to optimize lung cancer screening with low-dose CT (LDCT). We sought to compare the performance of traditional regression models and artificial intelligence (AI)-based models in predicting future lung cancer risk. A systematic review and meta-analysis were conducted with reporting according to Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. We searched MEDLINE, Embase, Scopus, and the Cumulative Index to Nursing and Allied Health Literature databases for studies reporting the performance of AI or traditional regression models for predicting lung cancer risk. Two researchers screened articles, and a third researcher resolved conflicts. Model characteristics and predictive performance metrics were extracted. The quality of studies was assessed using the Prediction model Risk of Bias Assessment Tool. A meta-analysis assessed the discrimination performance of models, based on area under the receiver operating characteristic curve (AUC). One hundred forty studies met inclusion criteria and included 185 traditional and 64 AI-based models. Of these, 16 AI models and 65 traditional models have been externally validated. The pooled AUC of external validations of AI models was 0.82 (95% confidence interval [CI], 0.80-0.85), and the pooled AUC for traditional regression models was 0.73 (95% CI, 0.72-0.74). In a subgroup analysis, AI models that included LDCT had a pooled AUC of 0.85 (95% CI, 0.82-0.88). Overall risk of bias was high for both AI and traditional models. AI-based models, particularly those using imaging data, show promise for improving lung cancer risk prediction over traditional regression models. Future research should focus on prospective validation of AI models and direct comparisons with traditional methods in diverse populations.

Sparse views X-ray computed tomography has emerged as a contemporary technique to mitigate radiation dose. Because of the reduced number of projection views, traditional reconstruction methods can lead to severe artifacts. Recently, research studies utilizing deep learning methods has made promising progress in removing artifacts for Sparse-View Computed Tomography (SVCT). However, given the limitations on the generalization capability of deep learning models, current methods usually train models on fixed sampling rates, affecting the usability and flexibility of model deployment in real clinical settings. To address this issue, our study proposes a adaptive reconstruction method to achieve high-performance SVCT reconstruction at various sampling rate. Specifically, we design a novel imaging degradation operator in the proposed sampling diffusion model for SVCT (CT-SDM) to simulate the projection process in the sinogram domain. Thus, the CT-SDM can gradually add projection views to highly undersampled measurements to generalize the full-view sinograms. By choosing an appropriate starting point in diffusion inference, the proposed model can recover the full-view sinograms from various sampling rate with only one trained model. Experiments on several datasets have verified the effectiveness and robustness of our approach, demonstrating its superiority in reconstructing high-quality images from sparse-view CT scans across various sampling rates.

Brain lesion segmentation is crucial for neurological disease research and diagnosis. As different types of lesions exhibit distinct characteristics on different imaging modalities, segmentation methods are typically developed in a task-specific manner, where each segmentation model is tailored to a specific lesion type and modality. However, the use of task-specific models requires predetermination of the lesion type and imaging modality, which complicates their deployment in real-world scenarios. In this work, we propose a universal foundation model for brain lesion segmentation on magnetic resonance imaging (MRI), which can automatically segment different types of brain lesions given input of various MRI modalities. We develop a novel Mixture of Modality Experts (MoME) framework with multiple expert networks attending to different imaging modalities. A hierarchical gating network is proposed to combine the expert predictions and foster expertise collaboration. Moreover, to avoid the degeneration of each expert network, we introduce a curriculum learning strategy during training to preserve the specialisation of each expert. In addition to MoME, to handle the combination of multiple input modalities, we propose MoME+, which uses a soft dispatch network for input modality routing. We evaluated the proposed method on nine brain lesion datasets, encompassing five imaging modalities and eight lesion types. The results show that our model outperforms state-of-the-art universal models for brain lesion segmentation and achieves promising generalisation performance onto unseen datasets.

Long scan time significantly hinders the widespread applications of three-dimensional multi-contrast cardiac magnetic resonance (3D-MC-CMR) imaging. This study aims to accelerate 3D-MC-CMR acquisition by a novel method based on score-based diffusion models with self-supervised learning. Specifically, we first establish a mapping between the undersampled k-space measurements and the MR images, utilizing a self-supervised Bayesian reconstruction network. Secondly, we develop a joint score-based diffusion model on 3D-MC-CMR images to capture their inherent distribution. The 3D-MC-CMR images are finally reconstructed using the conditioned Langenvin Markov chain Monte Carlo sampling. This approach enables accurate reconstruction without fully sampled training data. Its performance was tested on the dataset acquired by a 3D joint myocardial $ \text {T}_{{1}}$ and $ \text {T}_{{1}\rho }$ mapping sequence. The $ \text {T}_{{1}}$ and $ \text {T}_{{1}\rho }$ maps were estimated via a dictionary matching method from the reconstructed images. Experimental results show that the proposed method outperforms traditional compressed sensing and existing self-supervised deep learning MRI reconstruction methods. It also achieves high quality $ \text {T}_{{1}}$ and $ \text {T}_{{1}\rho }$ parametric maps close to the reference maps, even at a high acceleration rate of 14.

Radiation therapy is regarded as the mainstay treatment for cancer in clinic. Kilovoltage cone-beam CT (CBCT) images have been acquired for most treatment sites as the clinical routine for image-guided radiation therapy (IGRT). However, repeated CBCT scanning brings extra irradiation dose to the patients and decreases clinical efficiency. Sparse CBCT scanning is a possible solution to the problems mentioned above but at the cost of inferior image quality. To decrease the extra dose while maintaining the CBCT quality, deep learning (DL) methods are widely adopted. In this study, planning CT was used as prior information, and the corresponding strictly structure-preserved CBCT was simulated based on the attenuation information from the planning CT. We developed a hyper-resolution ultra-sparse-view CBCT reconstruction model, known as the planning CT-based strictly-structure-preserved neural network (PSSP-NET), using a generative adversarial network (GAN). This model utilized clinical CBCT projections with extremely low sampling rates for the rapid reconstruction of high-quality CBCT images, and its clinical performance was evaluated in head-and-neck cancer patients. Our experiments demonstrated enhanced performance and improved reconstruction speed.

Against the metal artifact reduction (MAR) task in computed tomography (CT) imaging, most of the existing deep-learning-based approaches generally select a single Hounsfield unit (HU) window followed by a normalization operation to preprocess CT images. However, in practical clinical scenarios, different body tissues and organs are often inspected under varying window settings for good contrast. The methods trained on a fixed single window would lead to insufficient removal of metal artifacts when being transferred to deal with other windows. To alleviate this problem, few works have proposed to reconstruct the CT images under multiple-window configurations. Albeit achieving good reconstruction performance for different windows, they adopt to directly supervise each window learning in an equal weighting way based on the training set. To improve the learning flexibility and model generalizability, in this paper, we propose an adaptive weighting algorithm, called AdaW, for the multiple-window metal artifact reduction, which can be applied to different deep MAR network backbones. Specifically, we first formulate the multiple window learning task as a bi-level optimization problem. Then we derive an adaptive weighting optimization algorithm where the learning process for MAR under each window is automatically weighted via a learning-to-learn paradigm based on the training set and validation set. This rationality is finely substantiated through theoretical analysis. Based on different network backbones, experimental comparisons executed on five datasets with different body sites comprehensively validate the effectiveness of AdaW in helping improve the generalization performance as well as its good applicability. We will release the code at https://github.com/hongwang01/AdaW.