Gene Doe

Gene Doe is a direct-to-consumer iPhone app that serves as a platform for health citizen science. App users will be given opportunities to join research studies that are conducted using their iOS and consumer health devices. The most relevant studies for the user are featured over time. Maintaining user privacy is paramount, and is achieved in large part by ensuring that users are anonymous to the platform and maintain full control over their collected data.  

Gene Doe is one example of what can be created using status/post, an iOS application that integrates a number of APIs to allow rapid and inexpensive production and deployment of mobile apps for medical research. The integrated APIs include REDCap, a popular web-based application that acts as the content management system and HIPAA-compliant data repository, and ResearchKit, an open-source software library for robust in-app data management and a polished user interface. Other APIs are integrated and available for study-specific requirements and features.

Apps produced using status/post deliver scheduled and ad-hoc surveys, and optionally collect sensor data from HealthKit-compatible devices. ResearchKit ActiveTasks can be launched, which lead the user through steps to assess fitness, cognition, gait, and sensorium. Methods for screening, obtaining informed consent, scheduling, messaging, and reports can be included for citizen science studies.

Status/post has been used in over fifty projects at ten academic medical centers in the US and Canada. The most recent project, RN Wellness, is a pilot study conducted at Medical University of South Carolina. In this study, we collected 900,000 heart rate samples and logged 42 million steps walked from forty nurses wearing Apple Watches. We reached 83% compliance with daily and weekly surveys.

Development is now focused on presents/with, a collection of server applications, apps, and FHIR interfaces with electronic health record systems to provide care providers with summarized and actionable reports of collected data. For the purpose of detecting anomalies needing review, personalized machine learning models are trained and injected into the user app for execution on-device.