We help address these challenges through three contributions. First, we publish a new dataset, EHRSHOT, which contains de-identified structured data from the electronic health records (EHRs) of 6,739 patients from Stanford Medicine.

The healthcare and AI communities have witnessed a growing interest in the development of AI-assisted systems for automated diagnosis of Parkinson's Disease

As rates of mental health conditions like depression spike, we desperately need new ways of identifying and treating people in distress. The last big breakthrough in treating depression was all the way back in the 1980s. That was when Prozac, the first SSRI antidepressant, was released. It and its subsequent copycats soon swept the globe, and hundreds of millions of people have now taken this kind of medication. But while three-quarters of people say the pills have helped them feel better, they don't work for everyone.

"It seems like this past week has been quite challenging for you," a disembodied voice tells me, before proceeding to ask a series of increasingly personal questions. "Have you been feeling down or depressed?" "Can you describe what this feeling has been like for you?" "Does the feeling lift at all when something good happens?" When I respond to each one, my chatbot interviewer thanks me for my honesty and empathises with any issues. By the end of the conversation, I will have also spoken about my sleep patterns, sex drive and appetite for food.

What if the idea of the autism spectrum is completely wrong? For years, we've thought of autism as lying on a spectrum, but emerging evidence suggests that it comes in several distinct types. These three words have become synonymous with autism, yet behind them lies a common misunderstanding. The idea of "the spectrum" suggests that all autistic people share similar experiences and behave in similar ways - only to a greater or lesser extent. The reality couldn't be further from the truth. Some autistic people may not speak at all; others are hyperverbal and extremely fluent.

Checklists are simple decision aids that are often used to promote safety and reliability in clinical applications. In this paper, we present a method to learn checklists for clinical decision support. We represent predictive checklists as discrete linear classifiers with binary features and unit weights.

Intelligent education stands as a prominent application of machine learning. Within this domain, cognitive diagnosis (CD) is a key research focus that aims to diagnose students' proficiency levels in specific knowledge concepts. As a crucial task within the field of education, cognitive diagnosis encompasses two fundamental requirements: accuracy and fairness. Existing studies have achieved significant success by primarily utilizing observed historical logs of student-exercise interactions. However, real-world scenarios often present a challenge, where a substantial number of students engage with a limited number of exercises.

Clinical diagnosis is a highly specialized discipline requiring both domain expertise and strict adherence to rigorous guidelines. While current AI-driven medical research predominantly focuses on knowledge graphs or natural text pretraining paradigms to incorporate medical knowledge, these approaches primarily rely on implicitly encoded knowledge within model parameters, neglecting task-specific knowledge required by diverse downstream tasks. To address this limitation, we propose Retrieval-Augmented Diagnosis (RAD), a novel framework that explicitly injects external knowledge into multimodal models directly on downstream tasks. Specifically, RAD operates through three key mechanisms: retrieval and refinement of disease-centered knowledge from multiple medical sources, a guideline-enhanced contrastive loss that constrains the latent distance between multi-modal features and guideline knowledge, and the dual transformer decoder that employs guidelines as queries to steer cross-modal fusion, aligning the models with clinical diagnostic workflows from guideline acquisition to feature extraction and decision-making. Moreover, recognizing the lack of quantitative evaluation of interpretability for multimodal diagnostic models, we introduce a set of criteria to assess the interpretability from both image and text perspectives. Extensive evaluations across four datasets with different anatomies demonstrate RAD's generalizability, achieving state-of-the-art performance. Furthermore, RAD enables the model to concentrate more precisely on abnormal regions and critical indicators, ensuring evidence-based, trustworthy diagnosis. Our code is available at https://github.com/tdlhl/RAD.