Skip to main content
Industry

The transformation of healthcare through machine learning and cloud-based predictive analytics

Microsoft_in_Health_blog_opt2_1200x628

In my last blog post, I focused on how organizations are harnessing the potential of the Internet of Things. For this post, I will address how the application of Machine Learning, with powerful cloud-based predictive analytics, has the potential to transform health in many ways, from early detection of dyslexia in Sweden to fraud detection and management in Croatia.

Helping schools identify students at risk for dyslexia. Optolexia, founded by researchers at Karolinska Institutet in Stockholm, Sweden, built a dyslexia screening tool for young children, using a repository of eye-tracking data and an analytical engine built with cloud-based Microsoft Azure Machine Learning, Optolexia aims to help schools identify students at risk for dyslexia significantly earlier than current screening tests. With early diagnosis, students can receive appropriate treatments to boost learning skills and improve academic performance. The impact of such preventive machine learning capabilities is substantial, if we consider that as many as 10 to 15 percent of school-age children are dyslexic, and the International Dyslexia Association estimates that there are 1 billion people with dyslexia worldwide.

Identifying insurance fraud. A recent report from PKF Littlejohn LLP highlights fraud as a serious and growing problem for the health sector and shows losses to average 5.6% across all sectors, but to be even higher in healthcare at 6.1% in the UK NHS across all areas of expenditures – payroll, procurement, GP, dental, optical, and pharmaceutical services, as well as losses to income from patient charges.

When it comes to fraud detection, Machine Learning finds valuable areas of application, for example by analyzing past claims data in order to identify algorithm for detection of patterns used by hospitals to up-code claims. In Croatia, Microsoft is working with a customer to analyze over 30 million historical claims data. The power of advanced analytics, Power BI visualizations and Azure Machine Learning revealed massive practices of certain hospitals inflating certain claims. Just two of the identified cases resulted in $100k of overpaid claims by CHFI.

Empowering clinicians with KPIs at their fingertips. At Sana Kliniken AG, the third largest private hospital chain in Germany, physicians use mobile solutions, Microsoft Power BI, and enhanced natural language processing capabilities to provide better care. For example, chief clinicians can access clinical KPIs anywhere, anytime, to help them make informed decisions at the point of care that align with evidence-based best practices.

Analyzing data from patient medical devices. Millions of asthma sufferers worldwide depend on Aerocrine monitoring devices to diagnose and treat their disease effectively. The devices are sensitive to small changes in ambient environment, so Aerocrine is using a Microsoft Azure cloud analytics solution to boost reliability. As a result, the company can see data from the devices in real time. This helps Aerocrine relay valuable information from their customer service team to the end users and predict when consumable sensors will need to be replenished.

A truly ubiquitous transformation of health can only happen if even more health organizations take advantage of the powerful opportunities offered by advanced analytics and machine learning powered by the cloud.

And for more health organizations to take advantage of advanced analytics, they need tools that are as easy to use as they are powerful. Microsoft delivers just that with its cloud-first, mobile-first, advanced analytics solutions, such as Cortana Analytics Suite. Our solutions are designed to empower everyone across a health organization to gain the insight they need when and where they need it.

What’s your advanced analytics and cloud story? Please share it with us via email, Facebook, or Twitter. And please let us know if you have any questions or feedback.