A research result published in the journal Nature on the 17th claims that a new artificial intelligence (AI) model has emerged, which can predict the trend of health changes throughout a person's life. This job can help doctors and health planners better understand and respond to personalized health needs. Most people will get sick more than once in their lifetime, but predicting how different diseases (such as cardiovascular disease and cancer) will affect each other is a challenge. Medical decision-making increasingly relies on predicting individual health trends based on medical history. AI provides powerful tools for identifying disease progression patterns by analyzing large datasets of patient records. But the full potential of these models has not yet been fully explored, especially in terms of population size. This time, a research team from the German cancer center DKFZ has created an AI model - Delphi-2M - to identify when a specific disease occurs relative to other events in patient records, such as lifestyle factors and other health conditions. The model was trained using health data from 400000 people in the UK and tested using data from nearly 2 million people in Denmark. They found that Delphi-2M predicted the incidence of over 1000 diseases based on personal medical history, with accuracy close to or better than existing tools, which can only predict a very small number of diseases individually. This model can also simulate possible pathways of health changes for up to 20 years in the future and generate synthetic data that can be used to train other AI models while protecting privacy. This method is expected to help identify people at high disease risk, guide screening plans, and support long-term planning of medical services. Future versions may include more types of health information and help improve personalized diagnosis and treatment. But scientists point out that the model will reflect biases in the training data, and its predictions should not be used for direct medical decision-making until further testing. (New Society)