This work combines the existing "Big Health Data for the Elderly" to fill the shortcomings of the traditional "self-diagnosis system. Based on the three-dimensional health portrait model, the health data of different elderly groups are analyzed and presented with data such as common comorbidity, health index change, comorbidity index and high incidence of diseases. At the same time, a three-dimensional human model system is constructed based on web3D technology, so as to realize more accurate and multi-faceted interaction forms and research measures. The work follows the trend of the times and introduces Web3D technology in medical diagnosis and treatment. Web3D technology can embed three-dimensional models of organs and bones in web pages and give them hot spots. Patients can interact with human organs and bones by operating interactive hot spots, so as to understand the corresponding diseases of painful parts from all angles. For some elderly patients who lack basic medical knowledge, this can make it easier for them to seek medical treatment. At the same time, the presentation of diseases in this work can make medical diagnosis develop in the direction of intelligence, speed and three-dimensional. It has a high development prospect in medical treatment.