Periodic Reporting for period 1 - MultiBioScan (Multimodal hyperspectral system for imaging of biological tissues glycation)
Okres sprawozdawczy: 2020-02-12 do 2022-02-11
According to the International Diabetes Federation, the problem of early diagnosis and monitoring the effectiveness of diabetes mellitus (DM) treatment is one of the most acute in modern healthcare. In 2021, there were 537M patients with DM in the world, and by 2045 this figure is to increase to 783M. DM is one of the ten deadliest diseases with a yearly mortality rate of over 6M. One of the consequences of DM is the appearance of foot tissue lesions (diabetic foot syndrome). The main factor in the pathogenesis of foot tissue lesions at DM is a violation of blood microcirculation. Recent studies have revealed that microvascular lesions in various organs, including the feet, are registered already in the first years of diabetes and even at prediabetic conditions long before the clinical manifestations of vascular complications. Therefore, increasing the effectiveness of prevention and treatment of diabetic angiopathies is possible only with timely diagnosis and correction of vascular disorders at early stages. To date, a common disadvantage of the methods available to clinicians for the diagnosis of complications of DM is the inability to noninvasively assess the degree of tissue glycation and monitor its metabolism, as well as to determine the location of the skin areas most probably exposed to the development of trophic ulcers.
In this project, utilizing emerging photonics-based technology, innovative solutions in machine learning, and definitive physiological characteristics, we introduce a diagnostic approach capable of evaluating the skin complications of diabetes mellitus at the earlier stage. The results of the feasibility studies, as well as the actual tests in patients with diabetes and healthy volunteers, clearly show the ability of the approach to differentiate the diabetic and control groups. Furthermore, the developed in-house polarization-based hyperspectral imaging technique accomplished with the implementation of the artificial neural network provides new horizons in the study and diagnosis of different diseases.
The developed hyperspectral system has shown the ability to sense alterations in blood content, blood oxygenation, and collagen structure in human skin in vivo. Proposed concept was successfully demonstrated in preclinical experiments and clinical studies and satisfied the major requirements for modern medical imaging techniques by providing high FOV, high spectral resolution, close to real-time data acquisition and processing, high level of sensitivity and specificity.
The Fellowship has successfully reached its ultimate goal of training a talented researcher, Dr Viktor Dremin, through a research project in the area of developing photonic devices for biomedical research and clinical diagnostics. According to the results of the project, 18 works were published, including 13 peer-reviewed articles, 1 book chapter and 4 conference proceedings. The results of the work were reported at 6 conferences.
In this project, utilizing emerging photonics-based technology, innovative solutions in machine learning, and definitive physiological characteristics, we introduce a diagnostic approach capable of evaluating the skin complications of diabetes mellitus at the earlier stage. The results of the feasibility studies, as well as the actual tests in patients with diabetes and healthy volunteers, clearly show the ability of the approach to differentiate the diabetic and control groups. Furthermore, the developed in-house polarization-based hyperspectral imaging technique accomplished with the implementation of the artificial neural network provides new horizons in the study and diagnosis of different diseases.
The developed hyperspectral system has shown the ability to sense alterations in blood content, blood oxygenation, and collagen structure in human skin in vivo. Proposed concept was successfully demonstrated in preclinical experiments and clinical studies and satisfied the major requirements for modern medical imaging techniques by providing high FOV, high spectral resolution, close to real-time data acquisition and processing, high level of sensitivity and specificity.
The Fellowship has successfully reached its ultimate goal of training a talented researcher, Dr Viktor Dremin, through a research project in the area of developing photonic devices for biomedical research and clinical diagnostics. According to the results of the project, 18 works were published, including 13 peer-reviewed articles, 1 book chapter and 4 conference proceedings. The results of the work were reported at 6 conferences.
WP1: The overall training activities and knowledge transfer. Fellow received a repertoire of scientific skills which can be applied to solve a range of different biomedical engineering and biophotonics problems aimed at translational research from the bench to clinical practice. Courses attended: COMSOL Light-Matter Interaction Workshop (03 June 2020); Research Professional Workshop (21 July 2020); EPSRC Open Fellowship webinar (03 December 2020); Nano and integrated photonics workshop (19-22 April 2021); Workshop on EPSRC Grants & Fellowships for Early Career Researchers (07 September 2021); Transferable Skills Workshops (2020-2022).
WP2: Numerical modelling and Monte Carlo simulations. Different types of tissue models were developed. A database of model diffuse reflection spectra has been created. The developed models made it possible to train a neural network for processing hyperspectral images, predict the optimal parameters of experimental installations and explain the obtained measurement results. Dissemination: IEEE Trans. Med. Imaging 40(4), 1207-1216 (2021); Diagnostics 11(1), 77 (2021); Proc. SPIE 11845, 118450U (2021). Conferences: Saratov Fall Meeting 2020; Saratov Fall Meeting 2021 (Invited Talk).
WP3: Development of instrumental implementation of multimodal hyperspectral measurements. An instrumental implementation of a multimodal hyperspectral system has been obtained, which allows measuring various parameters of blood flow, metabolism and structural changes of collagen in the skin. Dissemination: IEEE Trans. Med. Imaging 40(4), 1207-1216 (2021). Conferences: SPIE Defense + Commercial Sensing (Invited Talk).
WP4: Development of algorithms for image processing and analysis. Algorithms and the corresponding numerical codes for image processing and analysis were obtained. Dissemination: Biosensors 10(12), 201 (2020); IEEE Trans. Med. Imaging 40(4), 1207-1216 (2021); PLoS ONE 16(5), e0252296 (2021); Laser Phys. Lett. 17(11), 115605 (2020); J. Biophotonics 14(12), e202100216 (2021) (Front cover paper). Conferences: SPIE Defense + Commercial Sensing (Invited Talk); BioMedEng21; European Conferences on Biomedical Optics 2021; SPIE Photonics West 2021.
WP5: Verification of the technology. New phantoms have been developed for verification of multimodal hyperspectral measurements and image processing algorithms. The measurements of diffuse reflectance and fluorescence were verified using phantoms. Studies on healthy volunteers have demonstrated the effectiveness of wavelet analysis of the temporal dynamics of laser speckle-contrast imaging signals. Dissemination: IEEE Trans. Med. Imaging 40(4), 1207-1216 (2021); Quantum Electron. 21(2), 118-123 (2021); J. of Biomedical Photonics & Eng. 6(1), 010303 (2020) (Front cover paper). Conferences: SPIE Defense + Commercial Sensing (Invited Talk); Saratov Fall Meeting 2021.
WP6: Unification and approbation of the technology. A new method of hyperspectral imaging of DM complications is presented. The experimental system was tested under clinical conditions and demonstrated high sensitivity and specificity. Dissemination: (2020); IEEE Trans. Med. Imaging 40(4), 1207-1216 (2021); J. Biophotonics 13(10), e202000203 (2020) (Front cover paper). Conferences: SPIE Defense + Commercial Sensing (Invited Talk).
WP2: Numerical modelling and Monte Carlo simulations. Different types of tissue models were developed. A database of model diffuse reflection spectra has been created. The developed models made it possible to train a neural network for processing hyperspectral images, predict the optimal parameters of experimental installations and explain the obtained measurement results. Dissemination: IEEE Trans. Med. Imaging 40(4), 1207-1216 (2021); Diagnostics 11(1), 77 (2021); Proc. SPIE 11845, 118450U (2021). Conferences: Saratov Fall Meeting 2020; Saratov Fall Meeting 2021 (Invited Talk).
WP3: Development of instrumental implementation of multimodal hyperspectral measurements. An instrumental implementation of a multimodal hyperspectral system has been obtained, which allows measuring various parameters of blood flow, metabolism and structural changes of collagen in the skin. Dissemination: IEEE Trans. Med. Imaging 40(4), 1207-1216 (2021). Conferences: SPIE Defense + Commercial Sensing (Invited Talk).
WP4: Development of algorithms for image processing and analysis. Algorithms and the corresponding numerical codes for image processing and analysis were obtained. Dissemination: Biosensors 10(12), 201 (2020); IEEE Trans. Med. Imaging 40(4), 1207-1216 (2021); PLoS ONE 16(5), e0252296 (2021); Laser Phys. Lett. 17(11), 115605 (2020); J. Biophotonics 14(12), e202100216 (2021) (Front cover paper). Conferences: SPIE Defense + Commercial Sensing (Invited Talk); BioMedEng21; European Conferences on Biomedical Optics 2021; SPIE Photonics West 2021.
WP5: Verification of the technology. New phantoms have been developed for verification of multimodal hyperspectral measurements and image processing algorithms. The measurements of diffuse reflectance and fluorescence were verified using phantoms. Studies on healthy volunteers have demonstrated the effectiveness of wavelet analysis of the temporal dynamics of laser speckle-contrast imaging signals. Dissemination: IEEE Trans. Med. Imaging 40(4), 1207-1216 (2021); Quantum Electron. 21(2), 118-123 (2021); J. of Biomedical Photonics & Eng. 6(1), 010303 (2020) (Front cover paper). Conferences: SPIE Defense + Commercial Sensing (Invited Talk); Saratov Fall Meeting 2021.
WP6: Unification and approbation of the technology. A new method of hyperspectral imaging of DM complications is presented. The experimental system was tested under clinical conditions and demonstrated high sensitivity and specificity. Dissemination: (2020); IEEE Trans. Med. Imaging 40(4), 1207-1216 (2021); J. Biophotonics 13(10), e202000203 (2020) (Front cover paper). Conferences: SPIE Defense + Commercial Sensing (Invited Talk).
This is the first study to report a label-free, non-invasive polarization-sensitive approach that can discriminate between the skin of DM patients and control participants. It was shown that polarization-enhanced hyperspectral analysis can differentiate the skin tissue with age-related and DM changes.
The proposed system is capable of close to real-time hyperspectral image processing. The use of polarization-sensitive hyperspectral imaging and parameters of blood volume fraction, blood oxygenation, and degree of residual polarization in combination or individually can have important clinical applications, since they can help to identify patients with different degrees of microcirculation impairments in the lower extremities. The proposed parameters could serve as biomarkers of diabetic complications. They can also be used to evaluate therapeutic procedures aimed at preventing or reversing diabetic complications.
The results of the project can facilitate the development of the hyperspectral technique and give a new direction in the studies of different diseases.
The proposed system is capable of close to real-time hyperspectral image processing. The use of polarization-sensitive hyperspectral imaging and parameters of blood volume fraction, blood oxygenation, and degree of residual polarization in combination or individually can have important clinical applications, since they can help to identify patients with different degrees of microcirculation impairments in the lower extremities. The proposed parameters could serve as biomarkers of diabetic complications. They can also be used to evaluate therapeutic procedures aimed at preventing or reversing diabetic complications.
The results of the project can facilitate the development of the hyperspectral technique and give a new direction in the studies of different diseases.