By Elton Gomes
Microsoft has joined forces with SRL Diagnostics, one of India’s leading pathology labs, to train its artificial intelligence (AI) system in detecting cancer. The American technology company plans to teach its AI to find traits of cancerous cells by looking at a slice of the patient’s tissue on a glass slide.
Microsoft’s AI will use SRL’s bank of more than one million slides from older diagnosis as training material. Microsoft’s collaboration with SRL will help in improving the quality of digital pathology for screening by bringing together Microsoft’s Azure and AI innovations and SRL’s expertise in the study of cells and tissues (histology).
“Microsoft is committed to empower both the healthcare industry and patients by using AI to democratise healthcare for all. Our partnership with SRL Diagnostics will assist pathologists by equipping them with technology that will augment their capability,” Anil Bhansali, managing director at Microsoft India, said in a statement, IANS reported.
What will Microsoft’s AI do?
Microsoft said that its cancer-detecting AI will not replace the need of an oncologist. Through the AI, Microsoft aims to quicken the diagnosis and make more accurate and valid pointers to the doctor. The company said that this could likely lower down the cost of treatment, and trained specialists could have more time on their hands to reach out to more patients.
Bhansali said that the company is looking to partner with other qualified health care providers to acquire the large amounts of data that could be useful in training the AI. After the initial training is complete, multiple cases will be presented to the AI to assess its performance. As the AI is constantly learning, every new or different case it comes across can prove to be helpful in improving its diagnosis.
How is Microsoft changing healthcare
Microsoft has been working to improve healthcare for a long time. In February 2017, it launched Healthcare NExT, a new initiative that looked to transform healthcare while integrating greenfield research and health technology product development. Healthcare NExT also aimed to develop a new model at Microsoft for strategic health industry partnerships.
In June 2018, Microsoft expanded its healthcare initiative by hiring two industry veterans – Jim Weinstein and Joshua Mandel. Weinstein was the former CEO of the Dartmouth-Hitchcock Health system and joined Microsoft as the vice president of Microsoft Healthcare, and will work with healthcare organizations to transfer systems to the cloud.
How can AI help in detecting diseases?
Opportunities with AI seem to be endless, and the futuristic technology is being applied in variegated fields. DeepMind, an artificial intelligence company owned by Google, has developed an AI system that can accurately identify 50 different types of eye conditions as accurately as any doctor can. The system was reportedly capable of analysing 3D retinal scans for early signs of conditions like glaucoma, diabetic eye disease, and macular degeneration. It was developed as part of a joint research partnership with Moorfields Eye Hospital in London. The AI learned to detect eye conditions by studying roughly 15,000 anonymous eye scans.
Google and its health-tech subsidiary Verily were also involved in discovering a new way to use machine learning to assess a person’s risk of heart disease. Google’s software analyzes scans of the back of the patient’s eye and is accurately able to deduce data, including finding out the individual’s age, blood pressure, and whether or not they smoke.
This data is then used to predict their risk of suffering a major cardiac event, with almost the same accuracy as current leading methods. More importantly, the algorithm does not require a blood test to do this, and it makes it quicker and easier for doctors to monitor a patient’s cardiovascular risk.
Elton Gomes is a staff writer at Qrius.