This paper introduces MedExpMem, an experience memory framework designed to enhance medical vision-language models (VLMs) with differential diagnosis expertise. It allows diagnostic agents to learn from their own failures by memorizing discriminative experiences as pairwise differential notes.
This research introduces Transcriptome-based Drug Design (TBDD) as a generative inverse problem to design drug molecules conditioned on desired transcriptomic state transitions. It proposes "ACURE" (A Cellular Response Engine), a multi-resolution transcriptome-guided diffusion framework, to address the challenges of this complex task.
The University of Geneva's MangroveGS AI system predicts cancer metastasis with 80% accuracy by analyzing gene expression patterns, marking a breakthrough in personalized cancer care. Its successful adoption demands structured data, rigorous validation, interdisciplinary collaboration, and careful consideration of ethics and data privacy.
AI is being utilized to scan 1,500 approved drugs for hidden brain-disease signals, aiming to cut research time for conditions like MND, Parkinson's, and dementia from decades to years. Researchers are exploring if existing medications could be repurposed as treatments for neurological conditions.
Boston Children's Hospital utilizes OpenAI technology to enhance patient care and reduce operational burdens. This implementation has also facilitated the diagnosis of over 40 rare disease cases.
This article provides a friendly introduction to Foundation Models applied in Computational Pathology. It explores how artificial intelligence can analyze microscope slides to extract valuable information.
This article unpacks Google DeepMind CEO Demis Hassabis's statement at Google I/O, where he declared the company's ambition to "solve all disease" through reimagining drug discovery. The Verge's newsletter discusses the implications and context of such a bold claim.