Brain metastases are a common complication of systemic cancer, affecting over 20% of patients with primary malignancies. Longitudinal magnetic resonance imaging (MRI) is essential for diagnosing patients, tracking disease progression, assessing therapeutic response, and guiding treatment selection. However, the manual review of longitudinal imaging is time-intensive, especially for patients with multifocal disease. Artificial intelligence (AI) offers opportunities to streamline image evaluation, but developing robust AI models requires comprehensive training data representative of real-world imaging studies. Thus, there is an urgent necessity for a large dataset with heterogeneity in imaging protocols and disease presentation. To address this, we present an open-access dataset of 11,884 longitudinal brain MRI studies from 1,430 patients with clinically confirmed brain metastases, paired with clinical and image metadata. The provided dataset will facilitate the development of AI models to assist in the long-term management of patients with brain metastasis.

Despite the importance of memories in everyday life and the progress made in understanding how they are encoded and retrieved, the neural processes by which declarative memories are maintained or forgotten remain elusive. Part of the problem is that it is empirically difficult to measure the rate at which memories fade, even between repeated presentations of the source of the memory. Without such a ground-truth measure, it is hard to identify the corresponding neural correlates. This study addresses this problem by comparing individual patterns of functional connectivity against behavioral differences in forgetting speed derived from computational phenotyping. Specifically, the individual-specific values of the speed of forgetting in long-term memory (LTM) were estimated for 33 participants using a formal model fit to accuracy and response time data from an adaptive paired-associate learning task. Individual speeds of forgetting were then used to examine participant-specific patterns of resting-state fMRI connectivity, using machine learning techniques to identify the most predictive and generalizable features. Our results show that individual speeds of forgetting are associated with resting-state connectivity within the default mode network (DMN) as well as between the DMN and cortical sensory areas. Cross-validation showed that individual speeds of forgetting were predicted with high accuracy (r = .78) from these connectivity patterns alone. These results support the view that DMN activity and the associated sensory regions are actively involved in maintaining memories and preventing their decline, a view that can be seen as evidence for the hypothesis that forgetting is a result of storage degradation, rather than of retrieval failure.

Explainability features are intended to provide insight into the internal mechanisms of an AI device, but there is a lack of evaluation techniques for assessing the quality of provided explanations. We propose a framework to assess and report explainable AI features. Our evaluation framework for AI explainability is based on four criteria: 1) Consistency quantifies the variability of explanations to similar inputs, 2) Plausibility estimates how close the explanation is to the ground truth, 3) Fidelity assesses the alignment between the explanation and the model internal mechanisms, and 4) Usefulness evaluates the impact on task performance of the explanation. Finally, we developed a scorecard for AI explainability methods that serves as a complete description and evaluation to accompany this type of algorithm. We describe these four criteria and give examples on how they can be evaluated. As a case study, we use Ablation CAM and Eigen CAM to illustrate the evaluation of explanation heatmaps on the detection of breast lesions on synthetic mammographies. The first three criteria are evaluated for clinically-relevant scenarios. Our proposed framework establishes criteria through which the quality of explanations provided by AI models can be evaluated. We intend for our framework to spark a dialogue regarding the value provided by explainability features and help improve the development and evaluation of AI-based medical devices.

Computer-aided surgical simulation is a critical component of orthognathic surgical planning, where accurately simulating face-bone shape transformations is significant. The traditional biomechanical simulation methods are limited by their computational time consumption levels, labor-intensive data processing strategies and low accuracy. Recently, deep learning-based simulation methods have been proposed to view this problem as a point-to-point transformation between skeletal and facial point clouds. However, these approaches cannot process large-scale points, have limited receptive fields that lead to noisy points, and employ complex preprocessing and postprocessing operations based on registration. These shortcomings limit the performance and widespread applicability of such methods. Therefore, we propose a Transformer-based coarse-to-fine point movement network (TCFNet) to learn unique, complicated correspondences at the patch and point levels for dense face-bone point cloud transformations. This end-to-end framework adopts a Transformer-based network and a local information aggregation network (LIA-Net) in the first and second stages, respectively, which reinforce each other to generate precise point movement paths. LIA-Net can effectively compensate for the neighborhood precision loss of the Transformer-based network by modeling local geometric structures (edges, orientations and relative position features). The previous global features are employed to guide the local displacement using a gated recurrent unit. Inspired by deformable medical image registration, we propose an auxiliary loss that can utilize expert knowledge for reconstructing critical organs. Our framework is an unsupervised algorithm, and this loss is optional. Compared with the existing state-of-the-art (SOTA) methods on gathered datasets, TCFNet achieves outstanding evaluation metrics and visualization results. The code is available at https://github.com/Runshi-Zhang/TCFNet.

Three-dimensional (3D) ultrasound (US) aims to provide sonographers with the spatial relationships of anatomical structures, playing a crucial role in clinical diagnosis. Recently, deep-learning-based freehand 3D US has made significant advancements. It reconstructs volumes by estimating transformations between images without external tracking. However, image-only reconstruction poses difficulties in reducing cumulative drift and further improving reconstruction accuracy, particularly in scenarios involving complex motion trajectories. In this context, we propose an enhanced motion network (MoNetV2) to enhance the accuracy and generalizability of reconstruction under diverse scanning velocities and tactics. First, we propose a sensor-based temporal and multi-branch structure that fuses image and motion information from a velocity perspective to improve image-only reconstruction accuracy. Second, we devise an online multi-level consistency constraint that exploits the inherent consistency of scans to handle various scanning velocities and tactics. This constraint exploits both scan-level velocity consistency, path-level appearance consistency, and patch-level motion consistency to supervise inter-frame transformation estimation. Third, we distill an online multi-modal self-supervised strategy that leverages the correlation between network estimation and motion information to further reduce cumulative errors. Extensive experiments clearly demonstrate that MoNetV2 surpasses existing methods in both reconstruction quality and generalizability performance across three large datasets.

The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) causes Coronavirus Disease 2019 (COVID-19) and induces activation of inflammatory pathways, including the inflammasome. The aim was to construct Machine Learning (ML) models to predict critical disease and death in patients with COVID-19. A total of 528 individuals with SARS-CoV-2 infection were included, comprising 308 with critical and 220 with non-critical COVID-19. The ML models included imaging, demographic, inflammatory biomarkers, NLRP3 (rs10754558 and rs10157379) and IL18 (rs360717 and rs187238) inflammasome variants. Individuals with critical COVID-19 were older, higher male/female ratio, body mass index (BMI), rate of type 2 diabetes mellitus (T2DM), hypertension, inflammatory biomarkers, need of orotracheal intubation, intensive care unit admission, incidence of death, and sickness symptom complex (SSC) scores and lower peripheral oxygen saturation (SpO₂) compared to those with non-critical disease. We found that 49.5 % of the variance in the severity of critical COVID-19 was explained by SpO₂ and SSC (negatively associated), chest computed tomography alterations (CCTA), inflammatory biomarkers, severe acute respiratory syndrome (SARS), BMI, T2DM, and age (positively associated). In this model, the NLRP3/IL18 variants showed indirect effects on critical COVID-19 that were mediated by inflammatory biomarkers, SARS, and SSC. Neural network models yielded a prediction of critical disease and death due to COVID-19 with an area under the receiving operating characteristic curve of 0.930 and 0.927, respectively. These ML methods increase the accuracy of predicting severity, critical illness, and mortality caused by COVID-19 and show that the genetic variants contribute to the predictive power of the ML models.

The new artificial intelligence-based software, Roadmap (HeartFlow), may assist in evaluating coronary artery stenosis during cardiac computed tomography (CT) for transcatheter aortic valve replacement (TAVR). Consecutive TAVR candidates who underwent both cardiac CT angiography (CTA) and invasive coronary angiography were enrolled. We evaluated the ability of three methods to predict obstructive coronary artery disease (CAD), defined as ≥50 ​% stenosis on quantitative coronary angiography (QCA), and the need for percutaneous coronary intervention (PCI) within one year: Roadmap, clinician CT specialists with Roadmap, and CT specialists alone. The area under the curve (AUC) for predicting QCA ≥50 ​% stenosis was similar for CT specialists with or without Roadmap (0.93 [0.85-0.97] vs. 0.94 [0.88-0.98], p ​= ​0.82), both significantly higher than Roadmap alone (all p ​< ​0.05). For PCI prediction, no significant differences were found between QCA and CT specialists, with or without Roadmap, while Roadmap's AUC was lower (all p ​< ​0.05). The negative predictive value (NPV) of CT specialists with Roadmap for ≥50 ​% stenosis was 97 ​%, and for PCI prediction, the NPV was comparable to QCA (p ​= ​1.00). In contrast, the positive predictive value (PPV) of Roadmap alone for ≥50 ​% stenosis was 49 ​%, the lowest among all approaches, with a similar trend observed for PCI prediction. While Roadmap alone is insufficient for clinical decision-making due to low PPV, Roadmap may serve as a "second observer", providing a supportive tool for CT specialists by flagging lesions for careful review, thereby enhancing workflow efficiency and maintaining high diagnostic accuracy with excellent NPV.

The Varian Ethos system allows not only on-treatment-table plan adaptation but also automated contouring with the aid of artificial intelligence. This study evaluates the initial clinical implementation of an adaptive pelvic radiotherapy system, focusing on the treatment times and the associated learning curve. We analyzed the data from 903 consecutive treatments for most urogenital cancers at our center. The treatment time was calculated from the time of the first cone-beam computed tomography scan used for replanning until the end of treatment. To calculate whether treatments were generally shorter over time, we divided the date of the first treatment into 3-months quartiles. Differences between the groups were calculated using t-tests. The mean time from the first cone-beam computed tomography scan to the end of treatment was 25.9min (standard deviation: 6.9min). Treatment time depended on the number of planning target volumes and treatment of the pelvic lymph nodes. The mean time from cone-beam computed tomography to the end of treatment was 37 % longer if the pelvic lymph nodes were treated and 26 % longer if there were more than two planning target volumes. There was a learning curve: in linear regression analysis, both quartiles of months of treatment (odds ratio [OR]: 1.3, 95 % confidence interval [CI]: 1.8-0.70, P<0.001) and the number of planning target volumes (OR: 3.0, 95 % CI: 2.6-3.4, P<0.001) were predictive of treatment time. Approximately two-thirds of the treatments were delivered within 33min. Treatment time was strongly dependent on the number of separate planning target volumes. There was a continuous learning curve.

Inter reader variability and cross site domain shift challenge the automatic segmentation of prostate anatomy using T2 weighted MRI images. This study investigates whether transformer models can retain precision amid such heterogeneity. We compare the performance of UNETR and SwinUNETR in prostate gland segmentation against our previous 3D UNet model [1], based on 546 MRI (T2weighted) volumes annotated by two independent experts. Three training strategies were analyzed: single cohort dataset, 5 fold cross validated mixed cohort, and gland size based dataset. Hyperparameters were tuned by Optuna. The test set, from an independent population of readers, served as the evaluation endpoint (Dice Similarity Coefficient). In single reader training, SwinUNETR achieved an average dice score of 0.816 for Reader#1 and 0.860 for Reader#2, while UNETR scored 0.8 and 0.833 for Readers #1 and #2, respectively, compared to the baseline UNets 0.825 for Reader #1 and 0.851 for Reader #2. SwinUNETR had an average dice score of 0.8583 for Reader#1 and 0.867 for Reader#2 in cross-validated mixed training. For the gland size-based dataset, SwinUNETR achieved an average dice score of 0.902 for Reader#1 subset and 0.894 for Reader#2, using the five-fold mixed training strategy (Reader#1, n=53; Reader#2, n=87) at larger gland size-based subsets, where UNETR performed poorly. Our findings demonstrate that global and shifted-window self-attention effectively reduces label noise and class imbalance sensitivity, resulting in improvements in the Dice score over CNNs by up to five points while maintaining computational efficiency. This contributes to the high robustness of SwinUNETR for clinical deployment.

Synthesizing high quality CT projection data remains a significant challenge due to the limited availability of annotated data and the complex nature of CT imaging. In this work, we present PRO, a projection domain synthesis foundation model for CT imaging. To the best of our knowledge, this is the first study that performs CT synthesis in the projection domain. Unlike previous approaches that operate in the image domain, PRO learns rich structural representations from raw projection data and leverages anatomical text prompts for controllable synthesis. This projection domain strategy enables more faithful modeling of underlying imaging physics and anatomical structures. Moreover, PRO functions as a foundation model, capable of generalizing across diverse downstream tasks by adjusting its generative behavior via prompt inputs. Experimental results demonstrated that incorporating our synthesized data significantly improves performance across multiple downstream tasks, including low-dose and sparse-view reconstruction. These findings underscore the versatility and scalability of PRO in data generation for various CT applications. These results highlight the potential of projection domain synthesis as a powerful tool for data augmentation and robust CT imaging. Our source code is publicly available at: https://github.com/yqx7150/PRO.