Second Opinion, a Collaborative Online Game for Medical Diagnosis

Information technology has made tremendous impacts on the health care system. This work aims to develop an online intelligent diagnosis system, which will cooperate with human collaborators to improve its machine intelligence to provide fast and accurate diagnosis. We have developed an online interface so that a user can interact with the system as either a diagnoser or a diagnosee. As a diagnosee, the user can interact with the system to provide their symptoms. The system will use its intelligence to plan a strategy of questioning and finally provide a possible diagnosis. As a diagnoser, the user can review stored cases, plan his/her own strategy to interact with the system to learn symptoms of the selected case, and finally give his/her diagnosis. The system encourages the user to compete for points and for a good diagnoser ranking as a component of the game. Using such interaction with human collaborators, we have developed a machine learning algorithm so that the system can automatically improve its questioning strategy and provide faster and more reliable diagnosis. The system now focuses only on sexual transmitted diseases and will be extended to other diseases in future work.