These stories, research projects and technology capabilities from Microsoft and its partners around the world, are a great start to unravel the magic behind AI in health and understand how systems of intelligence are changing the lives of patients and the work of medical practitioners.
ScanDiags shortens time and cost of MRI-based diagnosis through image analysis that leverages Artificial Intelligence. The solution detects visible structures and pathologies and generates a written report that can then be completed with a radiologist's or doctor's conclusion. A catalog of 400 different structures is currently being developed. An initial set will be available during Spring 2017.
DevScope – HVITAL processes massive amounts of hospital data, 24/7, giving clinicians advance notice of oncoming patient risk, and thus the opportunity to avoid them. It can predict up to 30% of ICU admissions, 7 days in advance, based on assessments of clinical deterioration, infection management, and antibiotic misuse.
The project’s main focus is in the treatment of tumors and monitoring the progression of cancer in temporal studies.
InnerEye builds upon many years of research in computer vision and machine learning. It employs decision forests (as used already in Kinect and Hololens) to help radiation oncologists and radiologists deliver better care, more efficiently and consistently to their cancer patients.
Optolexia puts dyslexia risk assessment within reach during a child’s early primary schooling, making early diagnosis and treatment possible and learning easier. Optolexia plots and compares a child’s eye movements while reading to the relevant population at large, to determine if significant anomalies suggest further investigation and treatment.
Diagnosing and treating epilepsy is complicated. myCareCentric Epilepsy is the first solution to combine all available data sources to create a network of expert care that improves treatment for patients and reduces stresses on the healthcare system.
One of Epimed’s customers, Rede D’Or, reduced the incidence of hospital-induced infections by 20 percent and achieved and reached comparable international benchmark figures in its mortality rate by using the Epimed Monitor System—while minimizing IT costs.
With Azure Machine Learning, predictive analytics are significantly reducing their Hospital Acquired Infection and mortality rates. Epimed is also utilizing health information systems to predict and manage ICU Lengths of Stay (LOS). ICU’s become more efficient, producing results comparable to a 30% increase in hospital beds. Data-driven healthcare is the future.
Personnel from Akademiska Sjukhuset (Uppsala University Hospital), AbbVie, Bristol-Myers Squibb and Microsoft will work together for 52 hours, to develop and try new ideas and models for early diagnosis and better treatment of cancer, with a clear focus on patient benefit.
Project Hanover is making progress in three directions:
Information is only as good as the ability to use it, and Jorge Cortell ‘s vision is to make it available and usable to every medical professional involved in a case, to facilitate the most customized, effective treatment. Using machine learning and Microsoft Azure, images are standardized for all platforms, genomic data is shared, and relevant biological information is amassed to enable the best treatment.