Our Goal...
According to wikipedia, Parkinson’s disease (or simply Parkinson’s) is a long-term degenerative disorder of the central nervous system that mainly affects the motor system. Given that Parkinson’s is long-term, the symptoms take a very long time to fully develop. For that reason, it is necessary to diagnose early so that specialists have a better chance of treating Parkinson’s at its earliest stages. However, unfortunately, we do not currently have a specific test to diagnose Parkinson’s. Specialists have to examine a patient's medical history and then the specialist would suggest a neurological examination to determine the correct diagnosis. None of these would ultimately yield a “yes” or “no” answer to whether or not a patient has Parkinson’s but instead, the providers would recommend regular follow-ups to evaluate symptoms as time goes on.
Currently, parktest is a framework that helps in detecting Parkinson’s symptoms by performing some neurological tests. The goal of this platform is to ensure that anyone from anywhere in the world gets an early diagnosis in case they show symptoms of Parkinson’s. We plan to add on to the park test framework by improving the data collection framework and adding new features that would ultimately help in detecting Parkinson’s symptoms.
The goal of this project is to use Machine Learning to analyze videos recorded by the park test framework in terms of their quality (making sure the videos are not blurry) and their background. This is supposed to boost data collection since we will have well-recorded data which would be more useful for improving the general state-of-art of the framework.
Since providing real-time feedback may be difficult, users will start the test and then the entire video will be recorded and then analyzed with machine learning techniques. After this, feedback will be sent to the users about the quality of the video and results of the diagnosis (symptoms) based on this video. Users who had poor quality videos will be encouraged to complete the video again to make sure that the results are more accurate.