With the recent explosion in the Generative AI domain, like many other industries, healthcare organizations are also exploring ways to integrate LLMs into their workflows. In this five-part article series, we will explore how Generative AI can enhance experience in healthcare, from the perspective of both the patient and the provider. This is the fourth part of this series. You can read the first,  second and third part on this blog site. 

Now we focus our attention on the personal healthcare LLM. The objective of this LLM will be to be your personal keeper of healthcare data and records. This co-pilot should also be able to advise you, superficially, based on the personal data available and public data that it has access to, and has been trained on.

It is common sense that the power of this LLM will be derived from the co-pilot app having access to various healthcare-related apps and devices (like smartwatches). Let us explore some scenarios of this type of solution. For practicality purposes, we assume that the architecture is that the LLM is hosted in the cloud, not on the edge.

You wear a smartwatch and have decided to allow the co-pilot app to have access to health data captured by the watch. The co-pilot also has access to additional data points, like the medications you currently take, and the numbers from your annual physical exams.

As you come back from work one day, and settle down on the couch to recover from the long commute, you see a notification from the co-pilot.

You have indeed been staffed on a stressful project. Since your smartwatch can measure blood pressure as well, as highlighted by your co-pilot, you ask the co-pilot to remind you to measure the blood pressure whenever the stress level rises. You then decide to hit the bath and detox from the hectic work day.

The next day is no better. However, with the help of reminders from your co-pilot, you have been able to measure your blood pressure thrice. It does look like your blood pressure spiked during those periods of high stress. You are already on medication to manage your blood pressure, but it is apparent that it is currently not sufficient. Your co-pilot agrees. It may be time to see the doctor.

You agree with the suggestion as well. But the idea of seeing Dr. NoGood makes you uncomfortable. The doctor has zero bedside manners, and just a visit to this doctor is enough to increase your blood pressure. What if I change my specialist-is a thought that runs across your mind. You start your interaction with your co-pilot to find an alternative.

You have the appointment scheduled. A couple of days before the appointment, your co-pilot reminds you of the appointment, and also shares some logistics suggestions with you.

The scenarios above are obviously just some of the examples that a LLM like this can help with. The possibilities are plenty. Now that we have covered two buckets, in the final part of this article series, we will explore examples of Generative AI applications for service providers. The final part will be published on 06/25.