07 Feb 2024
…and how skin monitoring can help diabetic patients track blood sugar.
By Mike HatcherCancer Research UK Cambridge Institute have found that photoacoustic imaging can be impacted by the color and tone of a subject’s skin.
Presenting during last week’s SPIE BiOS conference Photons plus Ultrasound: Imaging and Sensing 2024, Thomas Else outlined the group’s study of 42 healthy volunteers – including six from each of the so-called “Fitzpatrick” skin types, where the lightest skin type is designated “Type 1”, and the darkest “Type 6”.
Else reminded the audience that the inherent bias in optical pulse oximetry hardware impacting people with darker skin emerged as a major issue during the Covid-19 pandemic, with dangerously low oxygen saturation levels missed and significantly worse healthcare outcomes resulting from the fact that melanin absorption in the near-infrared causes oxygen saturation to be overestimated by standard pulse oximetry devices.
For the PA imaging study, which used a multispectral optoacoustic tomography (MSOT) system from iThera to image various parts of the volunteers’ bodies, he reported increased acoustic “clutter” for Type 6 skin tones - with the similar light absorption profiles of melanin and hemoglobin again thought to be the root cause.
Darker skin tones also generated certain imaging artifacts that appeared as a bright reflection - an effect that could potentially confound a clinical diagnosis, for example when looking to identify blood vessels resulting from the presence of a tumor.
“Biases have the potential to affect patient care,” Else concluded in his outlook, suggesting that Fitzpatrick skin types should be reported at the very least, and that new hardware and software might be needed for certain applications of the imaging technology - for example by using longer wavelengths. “Photoacoustic imaging is confounded by skin color,” Else summarized. “We must do more.”
QCLs show promise for glucose monitoring
A team in France is developing a non-invasive glucose monitor based around quantum cascade lasers (QCLs) and photoacoustic spectroscopy of human skin to help diabetic patients track blood sugar levels.
Speaking during the BiOS conference Biophotonics in Exercise Science, Sports Medicine, Health Monitoring Technologies, and Wearables V, Jean-Guillaume Coutard from the startup company Eclypia described early clinical trials using the mid-infrared technology that have shown the ability to predict glycemia in real life conditions with Type-1 diabetics.
Called “Neogly”, the prototype system featured both an upper armband and a wristband to collect data, which could then be used to develop a numerical “twin” and train a deep-learning algorithm.
Coutard said that trials on ten individuals using the system for one week showed “pretty good” results, although there was still plenty of room for improvement in future trials on more people. The next steps for the Grenoble company, which was established in 2019 and is supported by BPI France, include shrinking the setup using integrated photonics and a miniature photoacoustic cell with a MEMS microphone, alongside an integrated battery and electronics.