How the rise of medical device use in pharma R&D is driving wearable technology in healthcare and revolutionising the medical industry


As wearable technologies continue to rapidly develop, the use cases for wearable devices evolve and expand.

Artificial intelligence (AI) and machine learning (ML) drive these devices and the latest advancements in digital health technology. The medical industry has begun to integrate these technologies in a way that has never been done before. Pharmaceutical companies are at the forefront, advancing medical treatments through their clinical research and development (R&D). By embracing new digital solutions with AI/ML features, pharma can not only cut costs and increase clinical trial efficiencies, but also advance the use of wearable technology on a broader scale in healthcare and revolutionise the medical industry.

Recent developments in wearable technology

In the past five years or so, the biggest change in the wearables industry has not been new, groundbreaking devices with never-before-seen metrics. Instead, it has been a long-anticipated rise in demand from the medical industry and wearables role in the growing “digital health” field1. Many wearable device companies have been working on medical solutions for years, but regulatory approvals2 followed by physician acceptance takes time. In many cases, it is the same devices that have been around on the consumer side for years, like the watches from Apple and Fitbit, that are now being used in more clinical research trials and by physicians to monitor patient health. Insurance companies are even seeing the benefits of individuals monitoring their own health through wearables and backing the development of wearable healthcare technology3.

Of course, there have been several advancements in sensor technology helping to drive the demand for wearables in healthcare. One primary example is wearable ECGs, which emerged about five years ago. While the Apple Watch may be one of the most well-known for their ECG-launch, many other companies, including Samsung, Huawei, Coros, and more, now offer the ability to take an ECG measure at the wrist and send that data to your doctor for review4. My father, who has suffered from atrial fibrillation (Afib), has used this feature many times now. Even when the limitations of the device were unable to detect Afib (i.e., HR too high), he was able to send the tracings to his cardiologist for further review. All of this was done on the same device he uses to track his golf game. This merging of recreational health and wellness with medical-grade biotech features is the hallmark of the wearables industry today.

AI/ML use in wearable development

To address the needs of the medical community, as well as offer additional features for consumers, many new devices are using the same sensor technology that has been around for years, but how the data from the sensors is used continues to evolve.

It should come as no surprise that new features are driven by AI and more specifically, machine learning. AI enables decision making and predictions through algorithms, while ML refers to the branch of artificial intelligence that develops algorithms through training based on data analysis, rather than explicit programming5. Sticking to the ECG example, several wearable ECG monitors, such as the Withings ScanWatch, offer suggestions for when to take a measure based on constant monitoring using the optical HR sensor to detect irregular beats. AI/ML will continue to improve features like these, offering the ability to learn from user data and offer tailored feedback, recommendations, and predictions.

Machine learning can be used to build static or dynamic algorithms. Static ML is based on data previously gathered and used to identify typical behaviours, and the model remains unchanged once implemented6. Examples of applications include the irregular heartbeat notification, fall detection, stress detection, and general fitness tracking7. But, to further improve accuracy on an individual level and enhance user experience, some devices use dynamic ML models. These models are not only based on previous data, but they continuously train based on new data and can learn from the habits, activities, and inputs from the current user. While both methods have their benefits and challenges, dynamic ML has shown great potential for aiding in critical physician decision making and real-time patient monitoring8,9.

Beyond new features such as reminders and improved individual accuracy, AI/ML is being used to generate new metrics based off existing sensors. For example, photoplethysmography (PPG), commonly used for HR monitoring, is now used for additional metrics including SpO2, HRV, stress, and respiratory rate. Vast research efforts and wearable device companies are continuing to invest in using PPG for even more health measures, such as non-invasive continuous glucose monitoring10,11, vascular health12, and calibration-free cuffless blood pressure13-15.

How pharma uses wearables

One of the most significant benefits of digital health devices to pharma clinical trials is the ability to continuously and objectively monitor biological data. Prior to these technologies, trials relied on patient surveys and physician visits at set intervals. While both are still used, wearables offer objective, continuous monitoring of subject health. Let’s use activity monitoring as an example.

Actigraphy devices are widely utilised in clinical research to evaluate activity level and physical functioning, a crucial data point used for study recruitment, protocol adherence, treatment response, and overall health assessment. The traditional six-minute walk test and user-reports of activity level are notoriously unreliable. Devices such as the ActiGraph and the Biostrap are non-invasive and unobtrusive for patients to use, improving both patient experience in trial participation and providing researchers with more data, in real-time. This simplifies trial design and can speed up the time it takes to understand and evaluate outcomes of a given treatment16. Objective data, quicker time to evaluate results, and increased adherence all cut costs for expensive drug trials and contribute to the many reasons pharma companies are rapidly adopting the use of these devices.

Of course, Pharma companies cannot simply trust any new wearable on the market and throw it into a multi-year drug trial. There are thousands of options on the market, offering different features, form factors, and applications. A trial for the treatment of IBS will likely use a different device than one for sleep apnoea, for instance. Pharma companies must invest in evaluating new devices and digital health platforms to determine their accuracy, risk, usability, and data availability. The most accurate device may be cumbersome or uncomfortable. If patients won’t use the device, there is no value in its accuracy.

Furthermore, patient safety is of utmost concern in setting up a clinical trial. While wearables can help improve trial safety overall through increased patient monitoring, pharma companies must ensure the selected device(s) themselves are safe and provide reliable data. Digital technology departments within pharma companies are becoming common, with the responsibility to stay on top of the latest digital health trends and devices, as well as understand how to use the data provided. The departments oversee preclinical testing of various devices, which has become a more prominent and required step prior to clinical trial initiation. Many pharma companies outsource to companies, like Valencell’s Biometrics Lab, who offer wearable device expertise and the ability to conduct this type of preclinical testing.

Translating R&D use to real-world patient applications

Physicians look to the research conducted by pharma to not only understand the efficacy of drug treatments and therapies, but also for how to monitor patients both for treatment and prevention. Patients are not one-size-fits all. To improve health outcomes overall, personalised medicine and individual approaches are the way of the future.

Wearables powered by AI/ML fill this need by offering physicians a way to identify potential health concerns sooner, monitor treatments, and adjust quicker. The challenges, however, are both from mistrust by physicians and adoption from consumers. Pharma companies have these same concerns and, thus, are the ones who demonstrate to the medical community which devices are most viable for real-world use.

Wearable devices most likely to succeed must demonstrate efficacy through proven, reliable results, as well as have the features patients want in a form they are familiar with and/or are already using. This is where consumer and medical devices will continue to merge.

What’s next

Innovations on the horizon will continue to be driven by AI/ML. Metrics like blood pressure, continuous glucose monitoring, cough detection, and more could soon be added features to your next smartwatch, likely using the same sensors as your old one. Pharma companies will continue to drive the integration of these wearables into the medical community, demonstrating to physicians how and why they should be used by patients to improve health outcomes. So, if you’re in pharma research and haven’t already, it’s past time to jump on the wearables bandwagon!


  1. N. Kasoju et al., “Digital health: trends, opportunities and challenges in medical devices, pharma and bio-technology,” CSIT, vol. 11, no. 1, pp. 11–30, Apr. 2023, doi: 10.1007/s40012-023-00380-3.
  2. S. LeBoeuf, Ph.D., “Cuffless BP monitoring – not for the faint of heart,” Healthcare IT TODAY, May 03, 2023.
  3. “Wearable Tech in Healthcare: Smart Medical Devices & Trends in 2023.” (accessed Sep. 08, 2023).
  4. “ECG smartwatches explained: How they work and the best on the market – Wareable.” (accessed Sep. 08, 2023).
  5. I. ACCESS, “Framework for the Use of Digital Health Technologies in Drug and Biological Product Development.”
  6. “Decoding AI: Static vs Dynamic and the Role of Generative AI | HireVue.” (accessed Sep. 09, 2023).
  7. F. Sabry, T. Eltaras, W. Labda, K. Alzoubi, and Q. Malluhi, “Machine learning for healthcare wearable devices: the big picture.,” J. Healthc. Eng., vol. 2022, p. 4653923, Apr. 2022, doi: 10.1155/2022/4653923.
  8. G. Mitsi et al., “Implementing digital technologies in clinical trials: lessons learned.,” Innov. Clin. Neurosci., vol. 19, no. 4–6, pp. 65–69, 2022.
  9. “Artificial Intelligence and Machine Learning in Healthcare.” (accessed Sep. 09, 2023).
  10. C.-T. Yen, U.-H. Chen, G.-C. Wang, and Z.-X. Chen, “Non-Invasive Blood Glucose Estimation System Based on a Neural Network with Dual-Wavelength Photoplethysmography and Bioelectrical Impedance Measuring.,” Sensors, vol. 22, no. 12, Jun. 2022, doi: 10.3390/s22124452.
  11. G. Hammour and D. P. Mandic, “An In-Ear PPG-Based Blood Glucose Monitor: A Proof-of-Concept Study.,” Sensors, vol. 23, no. 6, Mar. 2023, doi: 10.3390/s23063319.
  12. D. Castaneda, A. Esparza, M. Ghamari, C. Soltanpur, and H. Nazeran, “A review on wearable photoplethysmography sensors and their potential future applications in health care.,” IJBSBE, vol. 4, no. 4, pp. 195–202, Aug. 2018, doi: 10.15406/ijbsbe.2018.04.00125.
  13. M. Hosanee et al., “Cuffless Single-Site Photoplethysmography for Blood Pressure Monitoring.,” J. Clin. Med., vol. 9, no. 3, Mar. 2020, doi: 10.3390/jcm9030723.
  14. M. Nour, K. Polat, Ü. Şentürk, and M. Arıcan, “A novel cuffless blood pressure prediction: uncovering new features and new hybrid ML models.,” Diagnostics (Basel), vol. 13, no. 7, Mar. 2023, doi: 10.3390/diagnostics13071278.
  15. A. Avolio, “The reality and serendipity of cuffless blood pressure monitoring.,” Hypertens. Res., vol. 46, no. 6, pp. 1609–1611, Jun. 2023, doi: 10.1038/s41440-023-01269-z.
  16. “Acting on actigraphy: Why more clinical trials are leveraging actigraphy measures and endpoints – STAT.” (accessed Sep. 09, 2023).


Leave a Reply

Your email address will not be published. Required fields are marked *