MELBOURNE, Nov. 17 (Xinhua) — A research led by Australia’s University of Melbourne has developed an artificial intelligence (AI) tool capable of creating “digital twins,” or virtual patient replicas, to predict individual health trajectories.
Researchers used three datasets containing thousands of electronic patient health records to train an existing large language model, according to a university statement released Monday. The AI model called DT-GPT, hailed as “a potential gamechanger for the clinical trial sector,” analyzed medical data of patients with either Alzheimer’s disease or non-small cell lung cancer, as well as patients admitted to intensive care units, it said. The model created digital twins of patients and accurately forecast their health changes over time by leveraging pre-existing medical knowledge and analyzing patient histories, including lab results, diagnoses, and treatments, it added. The model wasn’t provided information on the health outcomes of the patients, allowing researchers to validate its predictions, the statement said, adding the DT-GPT model outperformed 14 other state-of-the-art machine learning models in predictive accuracy. “This technology paves the way for a shift from reactive to predictive and personalized medicine,” said University of Melbourne Associate Professor Michael Menden, lead researcher of the study published in NPJ Digital Medicine. “It could enable doctors to anticipate if their patient’s health will deteriorate so they can intervene earlier,” Menden said, adding it could also predict medication side effects, helping doctors customize treatments to each patient’s unique profile and improve health outcomes.The model quickly interprets complex data and features a chatbot-like interface for users to explore its predictions, he said, adding DT-GPT uses generative AI to make “zero-shot predictions,” meaning it can make educated guesses about lab values even without prior training on them.
