Microsoft Partners With Novartis To Develop A Kinect-Based MS Assessment


Microsoft announces a new strategic partnership with Swiss pharmaceutical company Novartis to co-develop a Kinect-based assessment tool that could be used to track patient symptoms and quantify whether certain multiple sclerosis treatments were having a positive effect. Novartis has been working on developing more consistent methods of measuring disease progress in MS patients for years, with little to show for its efforts thus far. The disease is well known for the unpredictable way that it progresses in patients. Two similar patients can often have entirely different experiences with the condition, with one losing mobility in a matter of years, while the other goes on to live a normal life with few debilitating symptoms.

Because MS often progresses over decades, doctors have developed a series of tests to objectively measure each patient’s symptoms and disease progress. Called the Multiple Sclerosis Functional Composite, the test is administered by a trained evaluator and attempts to objectively score a patient’s symptoms as they walk and perform various tasks using both their dominant and non-dominant hand. The test has been used with a fair degree of reliability by research organizations conducting clinical trials on MS thus far.

Novartis hopes to improve upon the test by mimicking the assessment over the Kinect platform. The primary goal is to improve inter-rater agreement for the test. Because each assessment is scored by an evaluator, there is a degree of subjectivity to the results that Novartis is hoping can be eliminated if software scores the movements rather than a clinician. A press release announcing the partnership explains, “Even when the same doctor is shown the same movement on two different days, that doctor may give that patient a different rating.” Novartis was careful to note that the goal of the project is not to replace the clinician, but to augment their assessment of the patient with objective, computer-generated scores. “What we’re doing is giving them a set of data that they can then weave into their judgment,” explains Abagail Sellen, a Microsoft researcher working on the project.

Researchers will turn to machine-learning algorithms to analyze real-time video of the patient as they perform specific tasks with the Kinect. These algorithms have become more sophisticated in recent years and perform particularly well at tasks requiring detailed image analysis. Researchers are using the same types of algorithms to support radiologists and pathologists as they screen medical images for potential tumors, and researchers at Google have been using the same types of algorithms to help its driverless cars spot hazards in the roadway.

Enjoy HIStalk Connect? Sign up for update alerts, or follow us at @HIStalkConnect.

↑ Back to top

Founding Sponsors

Platinum Sponsors