The promise of the possibility of noninvasive glucose monitors is to relieve the pain and infections caused by the invasive nature of commercial and mainstream glucose meters used by diabetes patients. Some of these technologies detect blood sugar levels through sweat or contact lenses, using optical sensors to beam glucose data to an attached application. Whereas other systems under development for home use and in-clinic and hospital settings use types of spectroscopy, a technique that identifies chemicals based on the interaction of molecules with electromagnetic radiation. Spectroscopy using lasers is able to produce accurate data on blood glucose levels without piercing the skin. However, the technology works better in hospital or clinic scenarios where it can be a tabletop system, rather than a wearable.
The possible market for noninvasive glucose monitors is large and growing. This has previously brought attempts to develop these devices from companies such as Google and Novartis, both which attempted to develop contact lenses capable of using fluorescent light to monitor sugars through the eye. The GlucoWatch, originally developed in the early 2000s, was eventually recalled by the FDA. C-8 MediSensors were devices using light to identify and analyze glucose molecules under the skin via interstitial fluid; the device managed to obtain an European CE mark in 2012, but the company went under a year later without developing a complete device.
There are several challenges to developing noninvasive glucose monitors. Some of these challenges are technological, and some are non-technological. Accuracy is a concern and requires more sensitive technology and sensors; with increased sensitivity, sensors can struggle with the amount of noise in detection. This is where algorithms, which work to convert sensor signals into glucose values, gain their importance. Especially as most methods work to measure changes in the blood glucose by measuring related changes that can be detected by these sensors. However, the differences between individuals, and the individual response to changes in blood glucose, can create problems for these algorithms in commercial uses.
Smaller components, desired for comfort and for wearable technology, reduce the accuracy of a system, while larger and bulkier devices offer increased accuracy but can reduce their portability and make them unsuitable for wearable technology. As well, even in a bulkier device, they cannot always provide the accuracy to justify the size. Further, understanding how to convert bloodless measurements to glucose values can be difficult. And for physicians, understanding these glucose measurements, and the glucose measurements in fluids other than blood, can difficult to establish.
Non-technological barriers include the cost of these products, the ease of use for wearers, and acceptance from people to wearing a continuous device.
The methods for monitoring and detecting glucose can generally be divided into three overarching methods, including into optical methods, microwave methods, and electrochemical methods. As of August 2021, there were at least sixty noninvasive and minimally invasive glucose monitoring products, either in development or available. However, researchers found, in studying these devices and their methods, that few have received FDA approval or a CE mark. In the reviewed products, twenty-eight were optical products, six were noninvasive fluid-sampling devices, and thirty-one were minimally invasive glucose monitoring devices.
Optical methods for noninvasive glucose monitoring include near-infrared reflectance spectroscopy, polarized optical rotation, Raman spectroscopy, fluorescence, and optical coherence tomography. These methods work to measure the specific response of various light waves and translate these responses to glucose concentrations in the body, rather than directly measure the glucose levels. The difference between individuals and the need to refine any algorithms or models to the individual response means the technology has not been considered a commercial possibility. And in some cases, dependent on the individual, the methods do not work as intended.
Near- and mid-infrared spectroscopy have a relatively ideal detection spectrum, with light having a relatively strong ability in the mid-infrared range to penetrate biofluids and soft tissues while scattering less than ultraviolet and visible light, and the sensing and measurement can be achieved by reflection and transmission. And this offers a possibility for developing a device for noninvasive glucose monitoring that is hampered by the complexity of the internal structure of human tissues and the spectral information of various substances that interfere with the accuracy of the results. This leaves the resulting accuracy to some algorithmic processing and modelling; if those models are not accurate, they further hamper the potential usefulness of this technology.
This was one of the earliest noninvasive blood glucose detection method. The method uses the optically active glucose which has a stable optical rotation, and can be illuminated with a polarized beam of light that can read the level of glucose from the deflection angle based on the glucose rotation. There are two devices for measuring this: a birefringence compensator and dual-wavelength polarizer. But the technologies, in studies, have difficulty with the accuracy of the data in clinical measurements, and the complexity of devices are not considered conducive to home-based blood glucose monitoring.
Raman spectroscopy, based on the Raman scattering effect that is a special effect in the scattering phenomenon of light, can be used to measure the scatter of light at specific frequencies. And unlike other optical methods, Raman spectroscopy detects the fundamental vibration of atomic groups with less overlap and interference and, accordingly, yields more accurate results. However, in measuring for glucose, the resulting Raman signals could be weak and masked by the strong background noise of the surrounding environment.
This method works to measure the decay of a molecule from a higher energy excited state back to a ground state, called a radiation relaxation phenomenon, which involves fluorescence and phosphorescence in the human blood. Fluorescence sensors have an advantage of high sensitivity, and are also able to characterize the spectrum of specific fluorophores to ensure high specificity of the sensor. The stability and wavelength of the fluorophore are factors in the fluorescence sensor that can impact the sensitivity and accuracy of the sensor. As well, the cost to develop such a sensor, which often uses quantum dots and carbon dots, can be prohibitive.
Optical coherence tomography (OCT) is a high-resolution imaging technology that provides depth-oriented tomography capabilities based on low-coherence interference. The contrast of OCT signals results from the scattering properties in biological tissues or materials, and is capable of using this to observe the internal microstructure of biological tissue without destroying said tissue. The distinction of glucose concentration leads to different dermal tissue scattering coefficients, which change the OCT signal slopes. This change in slopes can be used to measure changes in blood glucose concentrations. However, the presence of large scattering components and many scattering components can impede the accuracy of this method.
The interaction between electromagnetic waves and biological tissues offers advantages over optical detection methods for blood glucose monitoring. This is, in part, because microwave radiation has a lower energy per photon and scatters less in the atmosphere, which offers deeper penetration into tissues with more realistic data. The design of these sensors is often based on the dielectric properties of tissues, which have constant variations based on glucose fluctuations, and is closely related to the reflection, transmission, and absorption properties of microwaves.
These sensors are favored by researchers in the field of noninvasive blood glucose and have a broad development prospect due, in part, to the sensors' low cost and portability. However, the sensors face challenges in their sensitivity and selectivity, which many projects are working on. These challenges can cause sensors to miss changes in blood glucose levels.
Transdermal reverse iontophoresis (RI) is a needle-free blood glucose monitoring technique that extracts biomolecules and drugs through intact skin for blood glucose detection. This is based on the between ISF, an extracellular fluid that surrounds tissue cells, and blood that can provide information on the health of an individual. RI involves two mechanisms. The first is electro-migration, which is the direct interaction between charged ions and applied electrical field. The second is electro-osmosis, which is a connective solvent flow from anode to cathode direction. These mechanisms can allow the RI technology to extract the necessary biofluid to measure blood glucose.
This was the mechanism used for the GlucoWatch, the first FDA approved noninvasive glucose monitoring product, which later lost its FDA approval. Devices based on this technology have issues, including the need to use a standard glucometer for calibration, a long warm-up time, the need to charge gaskets every twelve hours, and the device could not be used during sweating while the prolonged electric current can cause skin damage to users. These issues were observed in the GlucoWatch, and the history of that product has impacted the research on RI sensors since.
This category of devices uses different bodily fluids (biofluids) to measure the glucose noninvasively. The sensor depends on the type of bodily fluid that is extracted and then measured. The fluids include saliva, tears, sweat, and interstitial fluids (ISF). Each fluid type is measured by a different type of sensor and comes with different advantages and disadvantages.
For example, saliva is easy to collect, but it often has low sensitivity and correlation to blood glucose while also breeding bacteria. While tears have less impurity, the comfort of measuring tears can be a problem, especially as tears offer poorer precision. Sweat is the easiest of the fluids to collect and can offer continuous monitoring, although it requires the wearer to be continuously sweating. And, as mentioned above, the collection of interstitial fluids (ISF), while offering the greatest sensitivity and accuracy, can cause discomfort and skin irritation in its collection.
However, ISF monitoring devices, with the relative success of the GlucoWatch, tend to draw the most interest, especially as these devices can provide continuous glucose monitoring, which can make up for the lack of accuracy of the method when compared to traditional blood glucose monitoring devices. Some devices that use ISF monitoring include:
- FreeStyle Libre, which was approved by the FDA in 2017
- Eversense CGM, which was approved by the FDA in 2019
- Dexcom G6 CGM, which was approved by the FDA in 2018
- Guardian Connect System, which was approved by the FDA in 2018
39 Potential New Continuous Glucose Monitors for Diabetes
February 20, 2020
7 Great Glucose Monitors and Meters
March 11, 2021
A thermal activated and differential self-calibrated flexible epidermal biomicrofluidic device for wearable accurate blood glucose monitoring
Zhihua Pu, Xingguo Zhang, Haixia Yu, Jiaan Tu, Hailong Chen, Yuncong Liu, Xiao Su, Ridong Wang, Lei Zhang, Dachao Li
Metamaterials Enabling Medical Breakthroughs, Promise Affordable, Accurate Point-of-Care Diagnostics -- Part 1 | INN
October 14, 2020
Non-invasive continuous-time glucose monitoring system using a chipless printable sensor based on split ring microwave resonators - Scientific Reports
July 31, 2020