IT Leadership

Apollo Hospitals Goes for Mix of AI, Big Data and Cognitive to Predict Disease Risks

The Indian healthcare industry is undergoing sweeping digital transformation in a bid to make healthcare delivery more efficient and patient centric. Apollo Hospitals, one of India’s leading private hospitals, has been at the forefront being one of the early adopters of digital technologies in healthcare. The hospital is aggressively working on a combination of Artificial Intelligence (AI) and big data analytics in the area of disease prediction on the clinical front and operational excellence on the non-clinical front.

After recently undertaking an AI initiative for preventive cardiac assessment through the use of advanced AI/ML algorithms that help predict cardiac diseases, the hospital is now working on a mix of digital technologies to build models that can help predict the disease risk in several other disease areas as well. These include infection, oncology and neuroscience departments to begin with and will gradually be extended to cover every possible clinical front and disease area.

While infection control has already been implemented, work is underway in the oncology and neurosciences areas. While the infection control initiative was mainly undertaken by the in-house team along with the clinicians, the hospital is working with different partners for the various other disease risks prediction projects.

According to Arvind Sivaramakrishnan, CIO, Apollo Hospitals Group, “We are trying to use the best of computing technologies to look at all the different disease areas to help predict the disease risk in each area to get the desired clinical outcomes. These technologies are not limited to AI/ML alone and will include a mix of AI/ML, cognitive computing, big data and predictive analytics.”

Apollo has implemented big data analytics for predicting infection risks for better infection control. It’s entire data in terms of the antibiotics, microbial systems, microbiology tests that go along with it and the diagnosis pattern of diseases has been analyzed for predicting and prescribing patterns to look at how to prevent and control hospital acquired infections. “Hospital acquired infections are always a risk from a tertiary care patient perspective and keeping that close to zero is an absolute delight for clinical outcomes. The tertiary care patients are already difficult clinical patients. So, ensuring that we have the least amount of complexity from other sources ensures that the treatment gets good clinical outcomes. Being a tertiary hospital this is very important for us,” explains Sivaramakrishnan.

Going forward, Sivaramakrishnan sees a lot of promise around ML based imaging analysis and running that into models of cognitive computing to provide predictive patterns.

The current and future models and algorithms for helping predict the diseases are trained on close to 4 million records of the patients at Apollo. This data from Electronic Medical Records (EMR) is completely scrambled and anonymized when fed into the models.

Importantly, predicting disease risk is just one part of ensuring effective healthcare delivery for hospitals, the other part being operational efficiency and service excellence. Sivaramakrsihnan recognizes this and is ensuring equal emphasis on bringing about efficiencies on the non-clinical front as well. “We are targeting the entire health system workforce, which is not limited to doctors and nurses alone. It includes the entire gamut of paramedical teams, ancillary support teams and administrative teams who have to function at high levels of effectiveness so that the outcomes are always delightful, ensuring that patient safety and patient care is always at its highest possible level,” explains Sivaramakrishnan.

On the operational efficiency and service excellence front, the hospital is looking at using a mix of AI/ML, IoT and mobility for predicting the length of stay, predicting patterns and understanding how the various diagnostic tests help in decision making for speeding up diagnosis time, tracing and understanding patterns of service levels from the hospital equipments, looking for patterns from the maintenance and engineering services that runs the hospital services. Essentially, the hospital is looking to understand patterns in all operational and service areas to help improve efficiencies.

Many of these use cases on the operational front are currently in the works and in different stages of maturity.

Leave a Comment

Your email address will not be published.

You may also like