Artificial intelligence (AI) and machine learning (ML) could be present in nearly every industry, driving what some consider a brand new age of innovation – particularly in healthcare, where it’s estimated the role of AI will grow at a 50% rate annually by 2025. ML is increasingly playing an important role in assisting with diagnoses, imaging, predictive health, and more.
With recent medical devices and wearables available in the market, ML has the aptitude to remodel medical monitoring by collecting, analyzing, and delivering easily accessible information for people to raised manage their very own health – improving the likelihood for the early detection or prevention of chronic diseases. There are several aspects researchers should have in mind when developing these novel technologies to make sure they’re collecting the very best quality data and constructing scalable, accurate, and equitable ML algorithms fit for real-world use cases.
Using ML to scale clinical research and data evaluation
Over the past 25 years, the event of medical devices has accelerated, especially through the COVID-19 pandemic. We’re beginning to see more consumer devices akin to fitness trackers and wearables commoditize, and development shift to medical diagnostic devices. As these devices are delivered to market, their capabilities proceed to evolve. More medical devices means more continuous data and bigger, more diverse data sets that must be analyzed. This processing could be tedious and inefficient when done manually. ML enables extensive datasets to be analyzed faster and with more accuracy, identifying patterns that may result in transformative insights.
With all this data now at our fingertips, we must ensure before everything that we’re processing the information. Data shapes and informs the technology that we use, but not all data provides the identical profit. We’d like high-quality, continuous, unbiased data, with the proper data collection methods supported by gold-standard medical references as a comparative baseline. This ensures we’re constructing protected, equitable, and accurate ML algorithms.
Ensuring equitable system development within the medical device space
When developing algorithms, researchers and developers must consider their intended populations more broadly. It’s not unusual for many corporations to conduct studies and clinical trials in a singular, ideal, non-real-world instance. Nevertheless, it’s critical that developers consider all real-world use cases for the device, and all of the possible interactions their intended population could have with the technology on a day-to-day basis. We ask: who’s the intended population for the device, and are we factoring in the whole population? Does everyone within the targeted audience have equitable access to the technology? How will they interact with the technology? Will they be interacting with the technology 24/7 or intermittently?
When developing medical devices which are going to integrate into someone’s day by day life, or potentially intervene with day by day behaviors, we also must think about the entire person – mind, body, and environment – and the way these components may change over time. Every human presents a novel opportunity, with variations at different points throughout the day. Understanding time as a component in data collection allows us to amplify the insights we generate.
By factoring in these elements and understanding all components of physiology, psychology, background, demographics, and environmental data, researchers and developers can ensure they’re collecting high-resolution, continuous data that permits them to construct accurate and powerful models for human health applications.
How ML can transform diabetes management
These ML best practices might be particularly transformative within the diabetes management space. The diabetes epidemic is rapidly growing across the globe: 537M people worldwide live with Type 1 and Type 2 diabetes and that number is predicted to grow to 643M by 2030. With so many impacted, it’s imperative that patients have access to an answer that shows them what is going on inside their very own body and allows them to effectively manage their conditions.
In recent times, in response to the epidemic, researchers and developers have begun exploring non-invasive methods of measuring blood glucose, akin to optical sensing techniques. These methods, nevertheless, have known limitations as a result of various human aspects akin to melanin levels, BMI levels, or skin thickness.
Radiofrequency (RF) sensing technology overcomes the restrictions of optical sensing and has the potential to remodel the way in which individuals with diabetes and prediabetes manage their health. This technology offers a more reliable solution in relation to non-invasively measuring blood glucose as a result of its ability to generate large amounts of information and safely measure through the total tissue stack.
RF sensor technology allows for data collection across several hundred thousand frequencies, leading to billions of information observations to process and requiring powerful algorithms to administer and interpret such large and novel datasets. ML is crucial in processing and interpreting the large amount of novel data generated from any such sensor technology, enabling faster and more accurate algorithm development – critical to constructing an efficient non-invasive glucose monitor that improves health outcomes across all intended use cases.
Within the diabetes space, we’re also seeing a shift from intermittent to continuous data. Finger pricking, for instance, provides insights into blood glucose levels at select points throughout the day, but a continuous glucose monitor (CGM) provides insights in additional frequent, yet non-continuous increments. These solutions, nevertheless, still require puncturing the skin, often leading to pain and skin sensitivity. A non-invasive blood glucose monitoring solution enables us to capture high-quality continuous data from a broader population with ease and with out a lag time in measurement. Overall, this solution would offer an unquestionably higher user experience and lower cost over time.
As well as, the high volume of continuous data contributes to the event of more equitable and accurate algorithms. As more time series data is collected, together with high resolution data, developers can proceed to construct higher algorithms to extend accuracy in detecting blood glucose over time. This data can fuel continued algorithm improvement because it includes various aspects that reflect how people change day-to-day (and throughout a single day), yielding a highly accurate solution. Non-invasive solutions that monitor different vitals can transform the medical monitoring industry and supply a deeper look into how the human body works through continuous data from diverse patient populations.
Medical devices creating an interconnected system
As technology advances and medical device systems achieve even higher levels of accuracy, patients and consumers are seeing increasingly opportunities to take control of their very own day by day health through advanced and multi-modal data from a wide range of products. But with the intention to see essentially the most impact from medical device and wearables data, there must be an interconnected system to create a smooth exchange of information across multiple devices with the intention to provide a holistic view of a person’s health.
Prioritizing medical device interoperability will unlock the total capability of those devices to assist manage chronic conditions, akin to diabetes. A seamless flow and exchange of data between devices akin to insulin pumps and CGMs will allow individuals to have a greater understanding of their diabetes management system.
High-fidelity data has the potential to remodel the healthcare industry when collected and used appropriately. With the assistance of AI and ML, medical devices could make measurable developments inside distant patient monitoring by treating individuals as individuals, and understanding an individual’s health on a deeper level. ML is the important thing to unlocking insights from data to tell predictive and preventative health management protocols and empower patients with access to information on their very own health, transforming the way in which data is used.