With national program of healthcare for the elderly, we are going to improve dementia diagnosis by machine learning and image processing software for ECD and TRODAT nuclear medicine images. Multi-center data of age and gender differences in image patterns of normal, AD, DLB and VaD patients will also be collected. The developed software will help to improve image analysis and diagnostic accuracy.

This program is based on clinical experience contributed by dementia experts and machine learning from engineers. Through a platform, the data are uploaded from different sites to reach a goal of precision medicine.

This program can be applied to both clinical practice and researched. More importantly, the rate of more correct diagnosis of dementia types can be improved and the quality of dementia care and long term services can be better. A business model exists.