Cloud Hosting

By Monojit Mazumdar, Partner and Krishatanu Ghosh, Manager, Deloitte IndiaIn Deloitte Centre for Health Solutions 2020 survey conducted in January 20201, 83% of respondents have mentioned Artificial Intelligence and Machine Learning (AI/ML) as one of their top two priorities.

Conventional wisdom has it that physicians cannot work from home. In the field of healthcare, traditional leverage of AI has been on disease detection and prediction. AI engines have generally been efficient in predicting anomalies in CT scans to detect onset of a disease.

Does it need to remain restricted to detection only? At specific scenario. Many of the Type1 diabetes patients now use a Continuous Glucose Monitor (CGM) to get a near real time reading of their blood sugar levels to determine insulin dosage. These commercially available devices pull the data and load into a cloud based data set-up at a regular interval.

Physicians look at the data during review and suggest adjustment to foods and dosage. A simple AI algorithm can take this further by recommending precise set of treatment recommendations for physicians to validate.

Since routine visits are getting deferred, this simple intervention has the potential to increase both precision and accuracy of the treatment process for all conditions that require timely and routine physician visits.

This opens up the possibility of AI being used as a recommendation tool as opposed to a detection only model. This single change has the ability to transform the entire business model of physical healthcare. From a facility to physically host healthcare professionals along with patients, hospitals and clinics may start operating as a digitally driven operations nerve center.

AI based scheduling service may listen to the patients conditions through a chat bot or voice application. It can ask a series of questions, look at the clinical records of the patients in the system and get a basic hypothesis ready for diagnosis based on data.

It can then schedule an appointment with the most competent physician available depending on the urgency. Before the appointment, the AI engine may prepare a complete briefing with potential diagnosis and recommended treatments. It can answer a set of follow on questions and allow the recommendations to be overridden.

In case of a required diagnostic intervention, AI driven scheduler should be able to arrange for an agent to collect the samples and add them in the patient dossier. Post tele or video consultation, a personal yet non-intervening Voice AI service may do regular follow-throughs, a reminder on medication and other recommended treatment follow through along with any future treatment recommendation. AI engine can sharpen this recommendation by constantly looking through data stream coming from devices that monitor the patient, by consulting physicians.

While this sounds futuristic, we have the technology components commercially available. With a strong and progressively cheaper data network, communication has just got easier. Cloud based storage and delivery of information has cut down the cost of computing infrastructure to a fraction. AI can process faster with advanced hardware gaining speed. Finally, a compulsive situation out of a pandemic has changed our mindset to believe things can be equally good if not better in a remote mode.

Through an efficient sharing of this data with suppliers, typical gaps of demand and supply can be bridged as well. Most important component of making the system work, the need for healthcare professionals may be calibrated as well and with increasing load on healthcare system, a changing model of treatment aided by AI seems to be a good option for future.

DISCLAIMER: The views expressed are solely of the author and ETHealthworld.com does not necessarily subscribe to it. ETHealthworld.com shall not be responsible for any damage caused to any person/organisation directly or indirectly.

Continue reading here:
Artificial Intelligence in Healthcare: Beyond disease prediction - ETHealthworld.com