Recent contributions


Contribution list

Recent comments

(1) View all

Erin commented on FacePrint

Hi Elizabeth! Great question. I built FacePrint using real Parkinson’s patients and control subject data that I collected during a research study. The algorithms that I built using these responses were able to accurately predict Parkinson’s disease using the facial features at an accuracy of 88%. I am currently expanding data collection and using new machine learning techniques in order to further improve the accuracy. It’s also important to note that this tool is designed to be used as a decision support tool, not as the sole means of diagnosis. This machine learning tool will enable clinicians to have a more comprehensive, objective understanding of disease onset, further improving accuracy. Hopefully this helps clarify. Let me know if you still have any more questions :)