Software challenges when developing EKG wearables

The past decade has seen the emergence of many new wearable devices that have been widely adopted by consumers and more recently by physicians for remote monitoring during the Covid-19 pandemic.

Modor Intelligence

The impact of Covid-19 on the wearable market is expected to have an ongoing significant effect, resulting in more accessible technology for diagnosis and management of disease being needed.

There are some key software challenges when it comes to the development of remote EKG wearables that leverage new technologies to provide efficient, cost-efficient, medical-grade accurate monitoring devices, including: signal noise, EKG variation and data strategy.

Signal Noise

An EKG records the electrical signal from the heart and checks for different heart conditions. Whether it’s in a hospital or outpatient environment, the EKG signal can be susceptible to additional noise such as powerline noise, muscle, and motion artefact.

This can make it difficult to clearly detect individual waves in the EKG signal, which can lead to inaccuracies in readings and compromise the EKG recording.

Despite the fact that the hardware and electrodes work to remove a level of noise, ultimately the software needs to recognise and remove or ignore the noise to ensure all readings are precise.

EKG Variation

The shape of an individual’s EKG is unique to them. This means EKG software will encounter and have to handle a wide variety of EKG signals.

Some individuals may have EKG shapes that are difficult for software to handle. If an individual has a particularly low amplitude signal, the software needs to be able to detect the wave shape accurately to ensure the algorithm results being provided to the user are correct whilst also being robust to noise.

Data Strategy

The ability to detect changes in the heart accurately and earlier is now available but for many these data points are just that; the translation of this to usable information via algorithms is critical for all of us to understand the consequences and make the relevant changes to improve our health and wellness.

Once the algorithms have been developed, further challenges are faced when they are ready to be validated. It is vital that the algorithms are developed and validated using representative data as algorithms trained using hospital data may not perform well on consumer devices like smartwatches.

Data collected from a hospital environment will be from a 12 lead, wet electrode EKG. However, for the wearable use case it is important that the algorithms are developed and validated on Lead I dry electrodes to ensure the highest possible accuracy in the end device.

Conclusion

When developing EKG software, there are many factors to consider and challenges to overcome. These challenges require years of research and expertise which can be a very timely and expensive process.

To help you overcome these challenges, B-Secur have created the EKG Development Kit, a full solution stack for accelerating your EKG development.

Using our 15 years expertise in EKG technology, we provide you with a Software Library of Algorithms as well as a Software Guide to ensure minimal effort in incorporating EKG technology into your product, saving you time, money and resource.

Ben Carter, CCO