Samsung’s open source project facilitates digital health research

Samsung Electronics recently announced the Samsung Health Stack 1.0 open source project focused on digital health research. The goal is to facilitate the creation of applications, backend services and analytics tools for wearable devices based on Android and Wear OS operating systems.

According to Samsung, smartphones and wearables have the potential to revolutionise health research and the development of data-driven biomarkers. However, significant amounts of time and money need to be invested in creating mobile apps and services in healthcare, an activity often unknown to medical researchers. The Samsung Health Stack aims to reduce the barriers between digital technology and medicine.

“The Samsung Health Stack, which wants to make digital health data acquisition easier and safer, will lower the barriers between medical researchers and digital technology experts, serving as a foundation for innovation in the digital health sector,” said Yunsu Lee, vice president and head of Samsung’s data intelligence team. “By promoting the development of digital health services, we hope to improve the quality of life for people around the world. We also hope that Samsung Health Stack, being an open-source project, will be useful to many people in this field.”

Wearable devices used in healthcare typically have a mobile app that stores and analyses the collected data. The Samsung Health Stack aims to provide a common framework to facilitate the joint development of these apps.

This tech stack brings together tools, applications and services often used to develop an end product. The Samsung Health Stack is just such a package presented as an open-source solution that collects and analyzes data coming from wearable devices with Android and Wear OS systems. And, as Samsung reinforces, as an open-source project, the Samsung Health Stack is open to contributions that seek to advance the project.

The technology stack includes:

• A software development kit (SDK);
• A web portal for survey creation, team member management, participant tracking and data analysis;
• Backend services and a data engine available via application programming interfaces (APIs).

With this package, developers can quickly and easily create applications, while survey respondents can be sure that the data they provide is being securely managed.

For example, when conducting studies with wearables on conditions such as diabetes and hypertension, the Samsung Health Stack allows researchers to simplify the data collection process by entering only relevant information into the portal. Apps developed with the App SDK, meanwhile, can facilitate frequent communication between researchers and study participants, leading to better results and making the development process faster for researchers to take on more varied and proactive health projects.

Since the alpha version was released at the Samsung Developer Conference (SDC) 2022, Samsung Health Stack has moved forward and gained a beta version in March. Version 1.0 is scheduled to be released in June.

Samsung Health Stack 1.0 will have improvements in survey input, participant integration, data collection from wearable devices, and basic survey functions. Security and stability have been enhanced to ensure more reliable surveys. The release also offers a feature that tracks clinical participants’ progress on specific survey-related tasks. The poll feature, which allows researchers to acquire data directly from participants, has also been enhanced with a dynamic function that can reflect the research objective and researchers’ intentions based on participants’ responses, allowing follow-up actions to be taken at a later date.

With easier-to-use research management tools, Samsung Health Stack 1.0 allows researchers to focus more on research design. It also features various methods for participants to provide relevant information, avoiding delays. The documentation provided to developers has also been significantly improved to make it easier to build research systems.

Errors in the development process

Developing an app for wearable devices can be a daunting task. If it has an interface that doesn’t provide an engaging user experience, such an app could be doomed to fail in its very first version, warns the Forbes Tech Council.

The group highlights ERRORS that should be avoided to keep these development projects on budget and more likely to succeed:

1. Forget cross-platform syncing – Cross-platform syncing allows you to use products from different vendors, for example Google Pixel headsets and a Mac.

2. Not including ML technology in the device – Wearable devices can benefit significantly from machine learning technology installed in the device. Initiatives such as TinyML are already supported by a significant number of processor manufacturers and allows building low-cost sensors that consume little power.

3. Neglecting battery optimization – Wearable devices can have reduced battery life when connected to multiple apps, so optimizing power consumption is crucial.

4. Not considering limitations – Wearable devices have limited screens, battery life and processing power, so you need to take this into account when designing apps.

5. Not securing the app – Wearable technologies are also exposed to security vulnerabilities and flaws, so it is important to prioritize app security on issues such as credentials or secure data transfer.

6. Forgetting user experience – You can’t forget, when including every feature in the app, to ask yourself how it will help deliver more value to those who use it.

7. Ignore privacy – Wearable devices work with personal data that cannot be exposed to privacy breaches.

8. Crowding the wearable screen with information – The screen on mobile devices is generally quite small. Adding too many features can tire the user with scrolling and too much unnecessary information.

9. Not paying attention to design – Prioritise user experience (UX) by paying attention to design, navigation and usability.

10. Add multistep processes – Don’t add features that will be little used. Focus on good notifications, quick, single button actionable actions.

11. Copy another model – Applications for wearable devices should have something unique, so it’s important to decide which features are most essential and focus on creating them well.

12. Not prioritizing safety – It is worth remembering that safety must come first!

13. Not considering when how wearable devices will be used – Remember the situations in which the app will be used, as often the user can only take a quick glance while on the move.

14. Force user to give too much information – Ideally, the user should be able to use the app even when they choose not to provide personal information.

15. Optimize the default mode parameters – Optimizing these fields can improve the end-user experience.