Paradigms don’t shift easily. It will take a lot of research and work to break out of the traditional, centralised understanding of healthcare – not to mention new materials, techniques, devices, data processing approaches and business models. Put all of that together, though, and great things can happen. Already, recent developments in optical biosensor technology and advanced research in the non-invasive monitoring of biomarkers are reshaping approaches to diagnosis. New tools are now appearing that make use of miniaturised, wearable and implantable biosensors.

One of the main advantages of wearable biosensors is their ability to non-invasively provide continuous, real-time information about physiological parameters. Integrated into medical wearables, these sensors are already helping to address diabetes, as well as cardiovascular, respiratory and many other diseases. Combining the biomarker monitoring used in these examples with technology for tracking parameters – such as activity levels, weather conditions and pollutants in the air – will enable physicians to more accurately assess individual health risks.

The list of possible applications for wearable biosensors is nearly endless. They can be used for monitoring vital signs in outpatients, premature infants, the elderly and people without easy access to medical services. These sensors can also provide valuable information and assistance to athletes, psychiatric patients and people who need long-term care. As such, wearable biosensors have the potential to not only improve our daily lives, but also to change the classical concepts of medical diagnostics and health monitoring. The industry might be on the edge of a shift from the established hospital-based care system towards a more personalised and home-based approach. While making medicine more convenient, this personalisation should also help to cut the time it takes to achieve diagnoses for individual patients, as well as the cost of healthcare more generally.

Hardware made easy

The first attempts to bring biosensors into the hands (or on to the wrists) of consumers began in the early 2000s. The goal then was to capture vital biological information in order to monitor health and fatigue. As a result, the first products were sport-focused, using electrocardiogram heart-rate electrodes to measure pulse rate via a body-clip or wristband. Not long after that, pulse-oximetry devices were integrated into these and similar systems in order to monitor blood oxygen levels. Most recently, thermal IR sensors have been added to measure body temperature.

“With such a range of applications, wearable biosensors have the potential not only improve our daily lives, but also to bring a change to the classical concepts of medical diagnostics and health monitoring.”

Now wearable biosensors come in all forms and shapes. There are accessories such as watches, rings, bracelets and necklaces; smart glasses and contact lenses; and even augmented hats, shirts, belts, shoes and socks. Needless to say, integrating biosensors across all these different forms is a very challenging task, particularly as they are often being adapted from a very well-defined medical environment to a broad range of products with very different sensor configurations. This change requires significant software development.

As far as the hardware is concerned, however, assembly technologies for biosensors are one of the important areas of expertise at Philips Innovation Services. Due to its unique combination of a Microelectromechanical systems (MEMS) foundry and micro-device assembly facilities, Philips Innovation Services is well-suited for the development of innovative devices at the smallest scale, and is further helped by its extensive background in product industrialisation and optoelectronics manufacturing. As a part of Philips, Philips Innovation Services focuses on medical applications, closely collaborating with a range of partners to develop various probes and sensors for both diagnostic and treatment purposes. At the same time, the company also works with a broad range of high-tech customers across manufacturing and consumer electronics.

The MEMS and micro-devices department of Philips Innovation services is especially experienced in manufacturing photonic sensors, whether they’re fibre-based or free-space, spectroscopic or interferometric. One such microscale sensor example is a fibre-based device used for imaging the respiratory tract to diagnose sleep apnoea, as pictured on the right.


Many of the wearable devices on the market today are equipped with optical biosensors. In most cases these are used to monitor users’ heart rate and blood oxygen levels. This non-invasive technique works by measuring changes in how light is absorbed, scattered or reflected. Oximetry is a good example of this because the optical properties of deoxygenated and oxygenated haemoglobin are starkly different.

Photoplethysmography (PPG) is another optical technique that can be utilised to quantify blood volume changes in the microvessels of tissues. Although PPG is a mature sensing technology and is used in many fitness trackers to measure heart rate, there is still room for improvement. The Interuniversity Microelectronics Centre (Imec) is working on using PPG to measure new parameters while making the technology more robust and reliable. While PPG usually operates at a close infrared or red, Imec is combining different wavelengths into its optical sensors, using red and IR light for oxygen saturation and green light for heart rate, to take two examples. Its researchers are also trialling combinations of different measurement modalities and different measurement points within the body. Key to this ‘multimodal’ application of PPG is a small form factor, low-energy consumption and development of smart algorithms to translate the data into actionable insights.

Furthermore, Imec is investigating the potential of hyperspectral-based PPG-imaging. In this approach, the wavelength of every individual pixel in an image can be detected, making it possible to monitor differences in pressure or oxygen saturation within a tissue. Miniaturisation of the hardware and low-power consumption are also important considerations here.

Optical and audible

Another parameter measured by wearable devices is body temperature. This approach has gained traction during the pandemic, as changes in skin temperature can be an indicator of infection. Excelitas develops optical biosensor components like photodiodes, essentially LEDs specialised for IR temperature sensing, for just these applications. The ongoing trend is towards smaller devices that require less power while making it possible to cluster IR Sensors with other optical solutions.

Another observable trend moves biosensing towards the ear, with so-called ‘(h)earable’ devices. These technologies may still need some time to evolve and gain traction, but as wireless headphones become more and more suited to functioning as hearing aids, the attraction of combining them with biosensors only grows. This trend also expands the target audience from sporty, fashion-conscious consumers to include older people who aren’t shy about wearing a visible ‘(h)earable’ that both improves their hearing and is able to give audible real-time updates on their health status.

Optical biosensors could also be combined with acoustics in photoacoustic sensors for wearables. Imec is investigating how to miniaturise hardware for sending out light and capturing sound waves to enable new sensing modalities. For example, photoacoustic sensors could be used to measure oxygen and oxidative stress, a biomarker in many chronic diseases. Imec is also applying its expertise in miniaturisation to shrink complex Raman spectroscopes for integration into everyday devices. Collaborations with clinical partners are helping Imec learn how this new technique can be used in clinical practice.

Lighting the way

Obviously, the light source is a central piece of any optical biosensor, and the wavelength of that light source has to be chosen with care. There are a few factors that determine that decision: namely, water absorption, isosbestic wavelength and tissue penetration depth. Water, the major component of most body tissues, has high lightabsorbing characteristics in the longer wavelength regions of infrared and ultraviolet. The isosbestic point is a specific wavelength at which different samples have the same absorptivity. For instance, a contrast in absorption between oxyhaemoglobin (HbO2) and haemoglobin (Hb) can be observed except at the isosbestic wavelengths, which is around 805nm for the near-infrared region. At that point, the signal remains unaffected by changes in blood oxygenation.

As optical biosensors truly become the centrepieces of wearable devices, they need to be comfortable, non-obtrusive, autonomous and compact. Furthermore, the development and production of wearable devices requires a combination of expertise from many different fields.

Even within photonics, the ecosystem is very scattered, which is a major challenge for end-user companies and manufacturers. On top of that, the strict regulations applied to wearable devices with medical classifications slow down the introduction of new solutions. To address these challenges, MedPhab has developed a pilot production line for accelerating the commercialisation of wearable diagnostic and point-of-care devices based on photonics technologies while reducing R&D costs. The pilot line is designed to provide a seamless transition to up-scaled production without a need for changing service providers. MedPhab’s high-quality infrastructure and extensive know-how is complemented with a globally unique ability to meet requirements and regulations in the medical domain.

There’s plenty still to be achieved. Further integration of wearable devices with cloud-based data will help to improve and personalise their performance and diagnostic value. Continuous monitoring, meanwhile, will contribute not only to the measurement of vital parameters following treatments, but can support a personalised approach to preventative medicine. The connectivity of wearable devices should also enable a faster shift to digital medicine, while making healthcare more accessible for patients in remote locations.

Certainly, multiple sensing strategies need to be developed and coordinated to make more biomarkers accessible to monitoring, and validation studies of wearable biosensor performance are required for them to achieve clinical acceptance. But the real-time sensing of physiological information using wearable biosensor technologies is already having a broad impact on our daily lives. The paradigm is shifting.