Smart questionnaire systems in digital health: Combining UX design and machine learning to improve data accuracy

Because the demand for precise, convenient and scalable data collection in digital health is growing, it has been realized that traditional health questionnaires are not effective in capturing trustworthy data people share about themselves. The main innovation in this work is to design a health questionnaire by using both UX and ML techniques to create an adaptive system. It changes the order of questions to help the user, using their actions as well as suggestions from estimated data reliability. When dealing with chronic patients, case studies showed that filling out forms online is easier, more data is captured, and users report better satisfaction after using the new approach. Because the system is built modularly, it includes adaptive questionnaires, monitors participants’ behavior and applies machine learning to deal with issues such as tiredness during a survey and biased answers. The study supports the idea that combining UX design and AI could make data collection in digital health more reliable, helpful to all and trustworthy. Researchers should explore the impact of multimodal data, federated learning and explainable AI to help AI be adopted more widely in the clinical field.