The system developed was used to reveal early changes in skin blood microcirculation and skin texture in patients with diabetes. Photo credit: University of Oulu
The multidisciplinary research team recently published its study “Skin Complications in Diabetes Mellitus Due to Polarized Hyperspectral Imaging and Machine Learning” in the IEEE Transactions on Medical Imaging.
In the publication, researchers present a diagnostic approach that can be used to assess the skin complications of diabetes mellitus at a very early stage using new photonics-based technologies, innovative machine learning solutions and definitive physiological properties.
In this work we performed a clinical validation of our optical device and a method that we developed within the project of the Academy of Finland. The method enables the contactless detection of possible skin complications in diabetes at an early stage as well as the implementation of a comprehensive population screening, says the associate professor Alexander Bykov from the University of Oulu, who explains the research in more detail in the following answers.
What is the key result?
We developed and carried out a clinical test of a compact, portable optical device for the non-contact functional characterization of human skin. The device can remotely measure spatial maps of blood oxygen and blood levels and assess changes in the skin’s collagen structure. To achieve this, the hyperspectral imaging and polarization detection technologies are combined and accompanied with the advanced algorithms of signal processing based on the artificial neural networks.
Hyperspectral imaging is a technique that combines conventional imaging and spectroscopy. Originally developed as a complex satellite or aircraft-based system, the technology has eventually evolved into a compact imaging tool that can be used for medical, industrial, and other relevant applications. With this technology, both spatial and spectral information of an object can be recorded. The 3D image obtained (two spatial and one spectral dimensions) consists of approximately one hundred or more spectral bands for each measured pixel of an object. This precise spectral and spatial information enables a detailed analysis of any object or environment. On the other hand, optical polarization detection enables remote assessment of structural changes within the object that are not visible with conventional hyperspectral imaging. The implementation of the algorithms for neural networks enables near real-time image processing based on the advanced numerical models, e.g. B. the seven-layer skin model that we use in our study.
The system developed was used to reveal early changes in skin blood microcirculation and skin texture in patients with diabetes. The back surface of the patient’s feet was imaged. It was observed that the diabetics had an increased skin blood content and at the same time a decreased oxygen content compared to the control group of healthy volunteers. In addition, the diabetic group has an increased polarization index, which is attributed to changes in the skin’s collagen structure. Thus, the results of the feasibility studies as well as the actual tests on patients with diabetes and healthy volunteers clearly show the ability of the developed approach to differentiate diabetic and control groups.
Why is the result important and interesting?
Timely detection of early-stage skin conditions caused by diabetes is crucial. In people with diabetes, high blood sugar levels damage many areas of the body such as the eyes, kidneys, legs, and feet. Metabolic changes in diabetes lead to clogging of large arteries, but also impair blood flow to small vessels in the lower extremities. These changes cause complications, with diabetic foot ulcers being the main cause. It occurs in 2-6% of type 1 and 2 patients with diabetes during their lifetime. If left untreated, the diabetic ulcers can become infected and develop deep tissue necrosis that may require amputation. The loss of limbs from major amputation is probably the most serious complication of diabetes, dramatically deteriorating the quality of life and putting a huge strain on the healthcare system. The economic cost of amputation is also enormous. Given the aging population, the number of at-risk patients will increase in the coming decade.
It is known that microvascular lesions are registered as early as the early years of diabetes and even in prediabetic conditions, long before the clinical symptoms and complications appear. Timely detection of the lesions, followed by appropriate treatment, allows their development to be reversed in the early preclinical stage, saving health, life and money.
Who is affected by the topic and the results? Where and for whom can the results be useful?
Our system may be able to monitor wound healing and treatment processes, including diabetic foot ulcers, skin burns, or post-operative complications associated with inadequate tissue oxygenation.
The present study focused on diabetics. A common drawback of the current methods available to doctors for diagnosing skin diabetic complications is their inability to assess tissue metabolism in a non-invasive and non-contact manner and to determine the location of the areas of skin most likely to be exposed to the development of trophic ulcers .
One way to improve the quality of diagnosis is to use hyperspectral and polarization-sensitive optical methods. Their advantages are associated with non-invasiveness, high resolution and low cost. Using the information obtained through non-invasive optical imaging would make it possible to identify those patients at increased risk for diabetic foot syndrome and to assess the areas of the lower extremities that are most prone to developing ulcerative defects.
Should the results change current practices (for example in healthcare?)
We are not seeing any dramatic change in current practice. However, the proposed technique can help doctors make the diagnosis more objective and make decisions. In the future, the technology can be adapted to patient self-monitoring, which corresponds to the strategy of personalized healthcare.
Did you do the clinical tests on real patients?
Yes, the tests in the clinic were carried out for 20 diabetics and 20 healthy volunteers in collaboration with our colleagues from the University of Latvia who helped organize the tests.
What is the scientific significance of the results?
To the best of our knowledge, the current work is the first to use a hyperspectrally resolved polarization index for the in vivo examination of diabetic skin and artificial neural networks trained by Monte Carlo to process the hyperspectral measurement data. The proposed diagnostic parameters could serve as biomarkers for diabetic complications. They can also be used to evaluate therapeutic procedures aimed at preventing or reversing diabetic complications. Our results can facilitate the development of biomedical applications of hyperspectral imaging and open new avenues in the study of age-related diseases.
Is there a need for further studies?
It would be of interest to test the developed system and data processing approach for other relevant clinical applications mentioned above. There is also great potential for the combination of fluorescence measurement and hyperspectral imaging for metabolic skin imaging that we are planning in the future.
Black people with type 1 diabetes, COVID-19, are four times more likely to be hospitalized for diabetic ketoacidosis
Viktor Dremin et al. Skin complications in diabetes mellitus from polarized hyperspectral imaging and machine learning, IEEE Transactions on Medical Imaging (2021). DOI: 10.1109 / TMI.2021.3049591
Provided by the University of Oulu
Quote: New method reveals the skin complications of diabetes (2021, February 22), accessed on February 22, 2021 from https://medicalxpress.com/news/2021-02-method-reveals-skin-complications-diabetes.html
This document is subject to copyright. Except for fair trade for the purpose of private study or research, no part may be reproduced without written permission. The content is provided for informational purposes only.