Wow, I didn’t expect that the Stanford heart study I participated in 11 years ago, *My Heart Counts*, has now evolved into one of the largest publicly available wearable device datasets. I only recently discovered this: in June this year, the Stanford team released a preprint for OpenMHC, systematically organizing the wearable device data accumulated over the past decade from *My Heart Counts* and preparing to open it up to the research community. This dataset covers over 10,000 participants, 19 data channels, and approximately 60 million hours of data. The data comes from iPhones, Apple Watches, and devices integrated with HealthKit. Even after compression, the entire dataset is still close to 40 GB. Applications for access will open once the paper is officially accepted. *My Heart Counts* was first announced at Apple’s March 2015 keynote, as a collaboration between Stanford University and Apple. At that time, the Apple Watch hadn’t even officially launched yet, so the early data collection relied mainly on iPhones, including activity data like step counts, walking/running distances, floors climbed, as well as some personal health information and a 6-minute walking test. Later, the study recruited over 100,000 participants. Among them, only about 10,000 met the data requirements, completed enough of the research process, and agreed to share their data with researchers. Since then, *My Heart Counts* has continued to evolve, gradually incorporating Apple Watch and more HealthKit data sources, with increasingly diverse data types. Numerous studies have been published based on this dataset, including in journals like *JAMA Cardiology* and *The Lancet*, yielding some key findings: 1. Smartphones can independently support large-scale real-world health studies, but participant bias and long-term dropout are very real challenges. 2. Smartphone reminders can indeed boost short-term physical activity, but the average effect is often just a few hundred extra steps per day. 3. Personalized exercise recommendations may be more effective than simply telling everyone to "walk 10,000 steps a day." 4. People who recover more slowly after COVID tend to also show slower recovery in their daily walking and running distances, though current evidence in this area is based on relatively small samples. 5. Long-term accumulated smartphone activity and heart rate data may even leave identifiable digital signals before certain diseases are officially diagnosed. The truly groundbreaking aspect of this isn’t just the creation of a 60-million-hour wearable dataset. More importantly, it proves something that seemed ahead of its time back in 2015: consumer electronics can accumulate real-world health data over years without significantly altering people’s lifestyles, eventually forming a reusable research infrastructure for the entire scientific community. Traditionally, medical research involved "recruiting a specific group of people and collecting data once to answer a single question." This model is now shifting to "long-term recording of real-world data from people, allowing future researchers to continuously ask new questions." In a way, the seemingly ordinary step counts, distances, and heart rate data we contributed 11 years ago are now becoming a new kind of public research infrastructure. Sources: https://arxiv.org/abs/2607.16235 https://(jamanetwork.com)/journals/jamacardiology/fullarticle/2592965 https://pubmed.ncbi.nlm.nih.gov/37794870/ https://www.(nature.com)/articles/s41746-023-00974-w Here are some screenshots and related records from when I was using it back then.