Background: We proposed a population graph with Transformer-generated and clinical features for the purpose of predicting overall survival (OS) and recurrence-free survival (RFS) for patients with early stage non-small cell lung carcinomas and to compare this model with traditional models.Methods: The study included 1705 patients with lung cancer (stages I and II), and a public data set for external validation (n=127). We proposed a graph with edges representing non-imaging patient characteristics and nodes representing imaging tumour region characteristics generated by a pretrained Vision Transformer. The model was compared with a TNM model and a ResNet-Graph model. To evaluate the models' performance, the area under the receiver operator characteristic curve (ROC-AUC) was calculated for both OS and RFS prediction. The Kaplan–Meier method was used to generate prognostic and survival estimates for low- and high-risk groups, along with net reclassification improvement (NRI), integrated discrimination improvement (IDI), and decision curve analysis. An additional subanalysis was conducted to examine the relationship between clinical data and imaging features associated with risk prediction.Results: Our model achieved AUC values of 0.785 (95% confidence interval [CI]: 0.716–0.855) and 0.695 (95% CI: 0.603–0.787) on the testing and external data sets for OS prediction, and 0.726 (95% CI: 0.653–0.800) and 0.700 (95% CI: 0.615–0.785) for RFS prediction. Additional survival analyses indicated that our model outperformed the present TNM and ResNet-Graph models in terms of net benefit for survival prediction.Conclusions: Our Transformer-Graph model was effective at predicting survival in patients with early stage lung cancer, which was constructed using both imaging and non-imaging clinical features. Some high-risk patients were distinguishable by using a similarity score function defined by non-imaging characteristics such as age, gender, histology type, and tumour location, while Transformer-generated features demonstrated additional benefits for patients whose non-imaging characteristics were non-discriminatory for survival outcomes.Funding: The study was supported by the National Natural Science Foundation of China (91959126, 8210071009), and Science and Technology Commission of Shanghai Municipality (20XD1403000, 21YF1438200).

The study population contains 145 patients who were prospectively recruited for coronary CT angiography (CCTA) and fundoscopy. This study first examined the association between retinal vascular changes and the Coronary Artery Disease Reporting and Data System (CAD-RADS) as assessed on CCTA. Then, we developed a graph neural network (GNN) model for predicting the CAD-RADS as a proxy for coronary artery disease. The CCTA scans were stratified by CAD-RADS scores by expert readers, and the vascular biomarkers were extracted from their fundus images. Association analyses of CAD-RADS scores were performed with patient characteristics, retinal diseases, and quantitative vascular biomarkers. Finally, a GNN model was constructed for the task of predicting the CAD-RADS score compared to traditional machine learning (ML) models. The experimental results showed that a few retinal vascular biomarkers were significantly associated with adverse CAD-RADS scores, which were mainly pertaining to arterial width, arterial angle, venous angle, and fractal dimensions. Additionally, the GNN model achieved a sensitivity, specificity, accuracy and area under the curve of 0.711, 0.697, 0.704 and 0.739, respectively. This performance outperformed the same evaluation metrics obtained from the traditional ML models (p < 0.05). The data suggested that retinal vasculature could be a potential biomarker for atherosclerosis in the coronary artery and that the GNN model could be utilized for accurate prediction.

This study investigated the affordances, constraints, and implications of ChatGPT in education using the affordance theory and social-ecological systems theory. We employed a data mining approach that blends social media analytics including sentiment analysis and topic modelling and qualitative analysis to extract viewpoints from a collection of datasets consisting of 33,456 tweets. Key findings indicate that 42.1% of analysed tweets conveyed a positive sentiment, 39.6% were neutral, and only 18.3% conveyed a negative sentiment. We also identified five categories of ChatGPT properties (e.g., text and data analysis, AI and machine learning) and an array of affordances of ChatGPT in education (e.g., facilitating student personalised learning, classroom instruction, provision of educational resources, curriculum changes, and assessment). Meanwhile, the findings revealed key concerns, including academic dishonesty, bias, and ethics that warrant attention. This study contributes to a real-time understanding of the impact of ChatGPT on education and informs researchers, educators, and policymakers to take a holistic approach to evaluating ChatGPT in educational practices.

Neurodegenerative diseases such as Alzheimer’s disease (AD) and Parkinson’s disease (PD) are clinically heterogeneous, hampering the success of nonselective treatment strategies. Here we apply a transformer-based unsupervised clustering framework to longitudinal electronic health record data from over 100,000 patients across two UK cohorts, Clinical Practice Research Datalink Aurum and UK Biobank, to identify, validate and characterize subtypes of AD and PD. We uncover five reproducible subtypes for each condition, characterized by distinct comorbidity patterns, symptom trajectories, outcomes and genetic profiles. These include a high-mortality AD subtype with motor and cardiovascular features, and a genetically susceptible but clinically resilient PD subtype. We also identify metabolic–inflammatory and vascular–psychiatric phenotypes shared across AD and PD, suggesting cross-disease mechanisms. By integrating routinely collected electronic health record data with genetic analyses, our study provides a scalable framework for early, biologically informed subtyping, laying the groundwork for future targeted interventions in neurodegenerative diseases.

Precision medicine tailored to individual patients has gained significant attention in recent times. Machine learning techniques are now employed to process personalized data from various sources, including images, genetics, and assessments. These techniques have demonstrated good outcomes in many clinical prediction tasks. Notably, the approach of constructing graphs by linking similar patients and then applying graph neural networks (GNNs) stands out, because related information from analogous patients are aggregated and considered for prediction. However, selecting the appropriate edge feature to define patient similarity and construct the graph is challenging, given that each patient is depicted by high-dimensional features from diverse sources. Previous studies rely on human expertise to select the edge feature, which is neither scalable nor efficient in pinpointing crucial edge features for complex diseases. In this paper, we propose a novel algorithm named \ours, which can automatically select important features to construct multiple patient similarity graphs, and train GNNs based on these graphs as weak learners in adaptive boosting. \ours{} is evaluated on two real-world medical scenarios and shows superiors performance.

Background: Chronic diseases are closely linked to alterations in body composition, yet there is a need for reliable biomarkers to assess disease risk and progression. This study aimed to develop and validate a biological age indicator based on body composition derived from dual-energy X-ray absorptiometry (DXA) scans, offering a novel approach to evaluating health status and predicting disease outcomes. Methods: A deep learning model was trained on a reference population from the UK Biobank to estimate body composition biological age (BCBA). The model’s performance was assessed across various groups, including individuals with typical and atypical body composition, those with pre-existing diseases, and those who developed diseases after DXA imaging. Key metrics such as c-index were employed to examine BCBA’s diagnostic and prognostic potential for type 2 diabetes, major adverse cardiovascular events (MACE), atherosclerotic cardiovascular disease (ASCVD), and hypertension. Results: Here we show that BCBA strongly correlates with chronic disease diagnoses and risk prediction. BCBA demonstrated significant associations with type 2 diabetes (odds ratio 1.08 for females and 1.04 for males, p < 0.0005), MACE (odds ratio 1.10 for females and 1.11 for males, p < 0.0005), ASCVD (odds ratio 1.07 for females and 1.10 for males, p < 0.0005), and hypertension (odds ratio 1.06 for females and 1.04 for males, p < 0.0005). It outperformed standard cardiovascular risk profiles in predicting MACE and ASCVD. Conclusions: BCBA is a promising biomarker for assessing chronic disease risk and progression, with potential to improve clinical decision-making. Its integration into routine health assessments could aid early disease detection and personalised interventions.