Fatima Rodriguez, MD,<\/a> Associate Professor of Cardiovascular Medicine and Section Chief of Preventive Cardiology at Stanford University School of Medicine, has a ready explanation. “The AI coach was fine-tuned specifically on the transtheoretical model of behavior change,” she says. “It generated more personalized messages tailored to each user’s stage of change. In behavior change and prevention, there is rarely a one-size-fits-all approach to engage individuals. This is the secret sauce of increasingly sophisticated generative AI models, recommendations can be more precisely tailored to specific context and data inputs.”<\/p>\nThe implications, if the finding holds up at scale, are significant. Personalized health coaching from a human professional is expensive, time-limited, and unavailable to the vast majority of people who might benefit from it. An AI system that can deliver comparable results changes that equation entirely.<\/p>\n
Proving It Works<\/p>\n
Enthusiasm is one thing; evidence is another. The new version of My Heart Counts is built around a randomized crossover trial specifically designed to separate genuine effect from wishful thinking.<\/p>\n
After a baseline monitoring period, participants are randomly assigned to receive either AI-generated, stage-matched messages or generic step reminders for one week, then they switch. Because each participant experiences both conditions, they serve as their own control group. This elegant design addresses one of the most persistent criticisms of digital health studies: that apps tend to attract people who are already motivated to change, making it impossible to know whether the technology is working or whether it’s simply tracking enthusiastic early adopters.<\/p>\n
The primary outcome measure is deliberately objective: daily step counts, analyzed using statistical models that account for the order in which participants experience each condition.\u00a0<\/p>\n
“The initial trial is intentionally short,” Johnson notes, “so it’s designed to measure near-term behavior change. Longer-term sustainment is a key next question the program is looking to study.”<\/p>\n
Also on the researchers\u2019 agenda: gathering feedback from participants on their feelings about specific nudges. At some point in the future, the app will be able to integrate that data for even further personalization and adaptability of the messages.<\/p>\n
Trust, Risk, and the Algorithm<\/p>\n
For many people, handing any aspect of their health over to an algorithm can feel uncomfortable, and the research team takes that skepticism seriously. Fatima Rodriguez draws an important distinction that often gets lost in broader debates about AI in medicine.<\/p>\n
\u201cTrust really varies a lot depending on what the AI is doing,\u201d she explains. \u201cNot all healthcare AI is the same. For low-risk behavioral coaching, the downside of a less-than-perfect output is less than for a diagnostic or therapeutic decision. AI can also be more consistent than humans and is less subject to fatigue, inconsistency, and training.\u201d<\/p>\n
My Heart Counts is not, the researchers stress, a medical device. It does not diagnose disease, recommend treatments, or replace the clinical judgment of a physician. Participation is voluntary, data sharing requires explicit permission, and sensitive integrations like electronic health record linkage are strictly opt-in.<\/p>\n
Just as importantly, the AI itself is intentionally constrained. It does not have access to full health records or unstructured personal data, and instead operates on a limited set of structured inputs\u2014such as activity patterns, survey responses, and stage-of-change signals. This bounded design helps keep outputs predictable and aligned with its narrow purpose: supporting behavior change, not making clinical decisions.<\/p>\n
The platform is also built on a foundation of unusual transparency for the health technology space: its code is fully open-source, and aggregated, anonymized data is shared with vetted researchers through governed access pathways, rather than being broadly available or controlled by a single commercial entity.<\/p>\n
Infrastructure for the Future<\/p>\n
The newest version of My Heart Counts is built on Stanford’s Spezi framework, a modular open-source architecture designed for precisely this kind of ongoing, evolving research. The rebuild was deliberate: the original platform’s aging infrastructure had begun to constrain innovation.\u00a0<\/p>\n
But now, due to the modular nature of the app, the research team will now be able to update it seamlessly, without rebuilding it from scratch, a move the team hopes will accelerate the pace of discovery across cardiovascular research broadly.<\/p>\n
And the implications of the rebuild don\u2019t stop there. Future updates to the dashboard could help users decrease their risk for diseases other than heart disease, such as breast cancer or osteoporosis.\u00a0<\/p>\n
A Study You Can Join<\/p>\n
Perhaps the most unusual thing about My Heart Counts is that it is, at its core, a study anyone can join. Adults 18 and older in the United States and United Kingdom (coming soon) can download the app, consent to participate, and begin contributing data that flows into real cardiovascular research at one of the world’s leading medical institutions.<\/p>\n
“Our goal is increased population health, not gatekept science,” the researchers say of their open-data philosophy.<\/p>\n
By blending behavioral science with generative AI, My Heart Counts offers a glimpse of a future where personalized health coaching is no longer a luxury reserved for those who can afford it. As the research suggests, a small nudge from your pocket may turn out to be an incredibly valuable (and accessible) tool for tackling the world’s leading cause of death.<\/p>\n
My Heart Counts is available for download on iOS. The study is open to adults 18 and older in the United States and the United Kingdom who speak English or Spanish.<\/p>\n","protected":false},"excerpt":{"rendered":"The Psychology of the Nudge The AI coaching system at the heart of My Heart Counts draws on…\n","protected":false},"author":2,"featured_media":29100,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[2],"tags":[24,25,19080,19081,52],"class_list":{"0":"post-29099","1":"post","2":"type-post","3":"status-publish","4":"format-standard","5":"has-post-thumbnail","7":"category-ai","8":"tag-ai","9":"tag-artificial-intelligence","10":"tag-artificial-intelligence-ai-in-medicine","11":"tag-cardiovascular-medicine","12":"tag-research"},"_links":{"self":[{"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/posts\/29099","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/comments?post=29099"}],"version-history":[{"count":0,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/posts\/29099\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/media\/29100"}],"wp:attachment":[{"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/media?parent=29099"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/categories?post=29099"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.europesays.com\/ai\/wp-json\/wp\/v2\/tags?post=29099"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}