The United States Medical Licensing Examination (USMLE) is a rigorous, multi-step assessment that all medical students must pass to be licensed to practice medicine in the United States. This comprehensive exam assesses a…
Our team investigated the feasibility of using GPT-4, a cutting-edge natural language processing model, to automate the conversion of free-text radiology reports into structured templates. We evaluated GPT-4’s performance on 170 detailed CT…
Our team developed a deep learning pipeline for the automated detection of active inflammatory and structural changes indicative of axial spondyloarthritis (axSpA) on biparametric MRI of the sacroiliac joints. The pipeline was trained…
Prostate cancer and kidney cancer are among the most prevalent cancers and have a profound negative impact on the quality of life of those affected and on our healthcare systems.
The COMFORT project (Computational Models FOR patienT stratification in urologic cancers – Creating robust and trustworthy multimodal AI for health care) aims to assist medical professionals in delivering improved care for people affected by prostate cancer or kidney cancer. The multidisciplinary research teams are developing a cutting-edge decision support system that uses artificial intelligence (AI) and data-driven insights.
The project strives to develop transparent and accurate computational models by integrating complex health data from multiple sources. These models will use advanced AI-powered risk stratification methods to help healthcare professionals select the right treatments, prevent disease progression, and improve the patient journey. Ultimately, the project will produce the first multi-national evaluation of AI models in a clinical setting and offer new insights to maximise the usefulness and acceptance of the technology.
The goal of inteRAGt is the development of a secure Retrieval-Augmented-Generation (RAG) system for the interactive analysis of patient records. Using large language models, documents and information can be quickly and intelligently queried in natural language. This system aims to augment traditional database queries and filtering methods, which are time-consuming for doctors and hospital staff and can lead to serious consequences if errors occur in the search results. Another primary concern of the project is not only to achieve a qualitative improvement in search and analysis methods but also to guarantee the protection of personal data, primarily when processing patient information, by implementing advanced AI security technologies. The system will be tested in a clinical setting under real-world conditions.
AIAH Lab 