Let’s see five areas where AI can be used for personal health management.
Andrea Koncz
Key Takeaways
Despite the continuous buzz surrounding AI over the past year, many individuals remain unaware of how to utilize AI for their personal health management.
Various use cases are emerging, ranging from skin-checking apps and specialized health chatbots to employing generative AI for personalized health purposes.
This change is monumental, but it doesn’t happen overnight, which is just as well, as this is also a cultural shift, and as such, it needs time.
AI and Generative AI are the biggest hype of the past year. Everyone talks about what artificial intelligence can/will be able to do. It has become the most prominent topic on our site as well, overshadowing “traditional” digital health. The FDA has cleared hundreds of AI and/or machine learning-enabled devices.
I reckon most healthcare professionals would struggle to provide a solid, practical answer to this.
Rolling up our sleeves we try to offer assistance and list actual ways to use artificial intelligence right now. Some of these won’t be available for everyone at once, most likely not all will be equally useful for all of us, but here they are as a start.
1) Using AI with EMR and personal health records
Most news reports highlight the integration of generative AI and electronic medical records (EMR), emphasizing tools that assist doctors in either collecting information or generating medical records. Nonetheless, we are beginning to see solutions that are equally beneficial for patients as well: the Abridge software automatically sends medical appointment summaries to patients, translating medical terminology into plain English along the way.
Similarly, solutions like Virtuoso aim to provide a single point of contact with the health system, and the consumer-facing extension allows users to view their health information and access an array of available self-service features, from 24/7 online chat and nurse-led call center to appointment booking and scheduling and insurance coverage checker.
Of course, these solutions are yet scattered scarcely, but will eventually become more and more widespread. So it is worth asking your doctor from time to time if there is a patient-accessible AI tool for your use as well.
2) Skin-checking apps
Skin-checking apps are a signature digital health solution for a number of reasons explained here, but they are also a signature AI use case for everyday people. There are many AI-powered apps to check our suspicious lesions or a wider range of skin conditions, like Miiskin, Cube, SkinVision, aysa or Skinive, just to name a few.
Emerging Trend Alert – Skin Checking Algorithms
As skin cancer is one of the most common cancer types worldwide, early detection and treatment are invaluable: almost all skin cancers (both melanoma and nonmelanoma) can be cured if found and treated early. Prevention and detection are the keys. And of course, everyone should have a skin check done from time to time – but not everyone has the option to have it.
Using our smartphones, a tool most of the earth’s population already has is a great access point to providing affordable care. That is why these apps were one of our bets for the first real-life applications of AI that will be widely used by patients.
3) Using ChatGPT for health purposes
We can all benefit from the generative AI revolution as this experiment from patient scholar Dave deBronkart (aka: e-Patient Dave) demonstrates: he used ChatGPT to summarise his last medical records and organised them in an easy-to-digest form, providing patient-ready takeaways. It is worth reading through his post, as he not only explains how he did it but also details potential pitfalls and concerns.
While it’s crucial not to mistake ChatGPT for our trusted family doctor, the AI agent can quite efficiently answer questions like what is GGT in your lab results or whether a blood pressure reading of 150/85 warrants further investigation. This infographic shows a range of use cases both for physicians and patients, you can dive deeper into the topic in this study The Medical Futurist published with Eric J. Topol.
AI in healthcare now
4) Patient care apps
The CommonSpirit health system has deployed a text-based, AI-driven patient outreach and care coordination tool. The tool uses patient data and artificial intelligence (AI) to predict patient care needs, refer patients to services within care pathways, and then coordinate that care. Patient navigators also help guide patients to different community resources and prepare patients for different elements of their care.
This tool aids not only in informing patients but also showed significant improvements in care outcomes for both maternity and orthopedic surgery patients during the pilot phase of the program. New Neonates in the Medicaid population likewise saw their length of stay go down, from 6.13 days to 4.3. The program also yielded a 37 percent decrease in pre-term births among the Medicaid population. Patients getting orthopedic surgery likewise saw improvements in outcomes. The average length of stay went down from 3.5 days to 1.93, a 45 percent decrease, while the 30-day readmission rate went down 71 percent, from a 3.5 percent readmission rate to 1 percent.
5) Non-generative AI chatbots
Maybe it’s a little surprising these days, but not all health chatbots come from the generative AI realm, there are differently built existing tools addressing specific problems. For example, Northwell Health launched one to help reduce “no-shows” for colonoscopies, a procedure elemental in colorectal cancer diagnosis. This issue is particularly concerning as 40 percent of less privileged patients don’t follow through with the procedure. This study from March 2023 reports how an app developed to help patients’ bowel preparation can increase compliance and thus, the number of successful colonoscopies.
medical chatbot AI algorithm person man phone TMF
In some cases, health chatbots are also able to connect patients with clinicians for diagnosis or treatment, but that is one step further down the line. The general idea is that in the future, these talking or texting smart algorithms might become the first contact point for primary care. Patients will not get in touch with physicians, nurses, or any medical professional with every one of their health questions but will turn to chatbots first. If the little medical helper cannot comfortably respond to the raised issues, it will transfer the case to a real-life doctor.
As the number of health chatbots multiplies at an incredible speed, we decided to list the most promising ones in this article to have a clue about where the health chatbot industry is heading.
Are we in Promise Land yet?
Obviously not. These tools can help a group of patients and not help others, we’ll surely see huge sociological and geographical gaps in how fast AI-supported healthcare arrives, and on top of that, we already discussed that increasing health equity through AI and digital health is not even a tech question.
However, we are witnessing changes of monumental scale and significance. First of all, AI became widely accessible in the past year, a huge step forward from the times when only a select few – a limited group of high-end developers, researchers, and especially lucky physicians – had access to it. Now (almost) everyone has a chance to try and learn to work with these tools and gain useful new skills.
AI solutions will arrive gradually in our lives, today you may notice nothing, but next year one of your regular health checkups may already be supported by an algorithm. We will eventually get used to the new AI team member in our care. It is not happening overnight, but that is just right this way. This is not only technological advancement but also a cultural change, and we all need time to adjust mentally.
Leave a Reply