Agenda Resource Management logo

Blog

Artificial Intelligence used for Parkinson’s Diagnosis

Posted: 19th October 2018 in Industry News

Click image to view full size.

This year, Agenda’s chosen charity is Parkinson’s UK and we have been fundraising for them throughout 2018. As many as 10 million people around the world are thought to have Parkinson’s disease, and this number is expected to rise with an aging population, in which the illness becomes more prominent. Current medications help to manage symptoms, but we don’t yet have a treatment that can cure, slow, stop or reverse its progression.

Parkinson’s was in the news recently as a new technological advancement is being tested to hopefully speed up the diagnosis of the disease. If this system is successful, it is also hopeful that it could be a useful first step in identifying patients suitable to participate in research to understand the disease better, and work on a cure.

Chinese tech company Tencent and London medical firm Medopad have teamed up to use artificial intelligence in the diagnosis of Parkinson’s disease. They have developed a camera that captures the way patients move their hands to determine the severity of their symptoms, without patients wearing any sensors or devices. The camera was programmed with existing videos of patients that had been assessed by doctors at the Parkinson’s Centre of Excellence at King’s College Hospital in London.

The aim of this new process is to speed up the motor function assessment process which usually takes more than half an hour. Using this new technology, patients could be assessed in three minutes and may not even have to attend a hospital.Medopad already develops mobile apps to monitor various health conditions including diabetes, cancer, and heart disease.

The new system works by filming fine hand movements such as opening and closing fists, which are then digitally analysed. This will allow doctors to remotely examine strength and movement tests to make a diagnosis. In the new project, the aim is also to train the AI software to automatically score the tests, making it a quicker process for doctors to determine the onset of Parkinson’s disease.

This would be especially important for early diagnosis, which in turn could help other scientists working on research for Parkinson’s find a cure for the disease.

At the moment, many Parkinson’s patients put off seeing a doctor until their motor symptoms require treatment or medication. Additionally, there are often unwanted delays in times between setting an appointment and being seen – an issue which could be helped with this new technology project.

As many patients wait too long to seek medical attention it leaves researchers with a small group suitable for medical trials, as it is important for patients to participate in trials before they start taking symptomatic medications. Another problem is that many patients are placed on symptomatic medication even if treatment is not yet required, further narrowing the group of suitable Parkinson’s patients who are able to participate in clinical trials.

Interference from medications makes it difficult for researchers to tell if the potential treatment being tested is slowing disease progression, or if they are just seeing the effects from other therapies already in use. Hopefully, the new AI system will be able to diagnose symptoms earlier and easier, giving researchers a bigger pool of patients who will be able to help find better treatment, or even a cure for Parkinson’s disease.