Microsoft in collaboration with SRL Diagnostics has created a new AI tool to diagnose cervical cancer faster. On the National Cancer Awareness Day, Microsoft has announced that the AI-powered cervical cancer screening API is currently in internal preview at SRL Diagnostics. According to estimates by the World Health Organization (WHO), cervical cancer is the fourth most frequent cancer among women worldwide. India accounts for 16 percent of the global burden of the disease.
The cervical cancer is also frequent in the rural regions of the country where literacy and awareness are low. In order to reduce its incidence, the two key factors are effective screening and early detection. SRL Diagnostics, the largest diagnostics laboratory company in India, has reportedly been witnessing an increase in the demand for cervical cancer screening. The two companies began collaboration to create an AI Network for Pathology in September 2018. It was designed to ease the burden of cytopathologists and histopathologists.
A digitally scanned version of a Whole Slide Imaging (WSI) slide (Photo: Microsoft)
“The proportion of cytopathologists in India is very low with respect to the number of patients,” said Dr. Arnab Roy, Technical Lead – New Initiatives & Knowledge Management, SRL Diagnostics. “We receive more than 100,000 Pap smear samples every year and there are only a few trained cytopathologists who are supposed to examine such slides,” he added. Roy added that 98 percent of these samples are normal and only remaining 2 percent require further intervention. The AI tool is designed to help cytopathologists find those remaining 2 percent of abnormal samples faster.
The AI-powered cervical cancer screening API, according to Microsoft, does the first-level check and offloads normal slides. This means that cytopathologists now have more time to spend analyzing the slides that show signs of abnormality. “Our partnership with Microsoft to co-create solutions having the potential to generate both social and business value, is another first-of-its-kind effort in the Indian diagnostics industry,” Arindam Haldar, CEO, SRL Diagnostics, said.
In order to develop Cervical Cancer Image Detection API, the cytopathologists studied digitally scanned versions of Whole Slide Imaging (WSI) slides manually. They marked their observations, which were used as training data for the AI model. SRL Diagnostics initially appointed one cytopathologist to annotate the images slides. However, each WSI comprises of as many as 1,800 tile images, making it difficult for a single individual to interpret a large quantity of image tiles from each slide in ways that can be consumed by the AI algorithm.
“We wanted to create an AI algorithm that could identify areas that everybody was looking at and create a consensus on the areas assessed,” Manish Gupta, Principal Applied Researcher, Microsoft Azure Global Engineering, said in a released statement. The project then expanded to include five cytopathologists across multiple labs and different locations. These cytopathologists annotated thousands of tile images of cervical smear with each comprising about 300-400 cells.
“The collaboration has already started showing results with the first API for screening cervical cancer ready for internal preview at SRL Diagnostics,” Microsoft said. “The AI model can now differentiate between normal and abnormal smear slides with accuracy and is currently under validation in labs. It can also classify smear slides based on the seven-subtypes of cervical cytopathological scale.”