Capillaroscopic diagnostics of diabetic microangiopathy using artificial neural networks in patients with diabetes mellitus

Cover Page


Cite item

Full Text

Open Access Open Access
Restricted Access Access granted
Restricted Access Subscription or Fee Access

Abstract

Background. The widespread prevalence of diabetes and the progressive deterioration of health against the background of this disease substantiate the need for the use of new methods for the early diagnosis of diabetic microangiopathies. The combined use of digital capillaroscopy and deep learning technologies will significantly improve the quality and speed of microvascular disorders diagnostics.

Aim. Development of a system for assessing capillary images based on artificial neural networks, as well as its testing for the early diagnosis of microangiopathy in patients with diabetes mellitus.

Material and methods. 136 patients (59 females and 77 males) with type 1 diabetes mellitus aged 25.82±6.05 years were examined and divided into two groups: the first group included 65 (47.8%) patients who had no symptoms of diabetic microangiopathies, the second group — 71 (52.2%) patients who were diagnosed with one or more diabetic microangiopathies. The control group consisted of 30 practically healthy volunteers. All patients with diabetes melitus, as well as individuals from the control group, underwent сapillaroscopic examination. The resulting images were analyzed using the developed evaluation system based on artificial neural networks. Statistical data processing was performed using the Student and Mann–Whitney tests (U-test), logistic regression analysis, and ROC analysis.

Results. In patients with diabetes mellitus, there was a decrease in the capillary network density in both groups and the diameter of the arterial sections in the second group. Capillaroscopy using the developed system showed a sufficient level of significance (χ2=21, p=0.000), high sensitivity (71.43%) and specificity (85.71%). This method can be used in the diagnosis of microangiopathy in diabetic patients.

Conclusion. The combined use of capillaroscopy and neural networks allows to increase the speed and quality of the examination, as well as simplify the interpretation of the resulting images.

Full Text

Restricted Access

About the authors

Evgeniy S. Krutikov

Crimean Federal University named after V.I. Vernadsky

Email: nephrostar@yandex.ru
ORCID iD: 0000-0002-5754-4418
SPIN-code: 5967-2847

M.D., D. Sci. (Med.), Prof., Head of Depart., Depart. of Propaedeutics of Internal Medicine, S.I. Georgievsky Medical Academy

Russian Federation, Simferopol, Russia

Viktoriya A. Zhitova

Crimean Federal University named after V.I. Vernadsky

Author for correspondence.
Email: mail@onephro.ru
ORCID iD: 0000-0002-1772-6399
SPIN-code: 3406-6010

M.D., Cand. Sci. (Med.), Assistant, Depart. of Basic and Clinical Pharmacology, S.I. Georgievsky Medical Academy

Russian Federation, Simferopol, Russia

Marina A. Rudenko

Crimean Federal University named after V.I. Vernadsky

Email: rudenko.ma@cfuv.ru
ORCID iD: 0000-0002-8334-8453
SPIN-code: 1900-7487

Cand. Sci. (Technic.), Assoc. Prof., Depart. of Computer Engineering and Modeling

Russian Federation, Simferopol, Russia

Rustam O. Akaev

Crimean Federal University named after V.I. Vernadsky

Email: akaevrustam1975@mail.ru
ORCID iD: 0000-0002-3897-8042
SPIN-code: 9564-4620

M.D., Depart. of Propaedeutics of Internal Medicine, S.I. Georgievsky Medical Academy

Russian Federation, Simferopol, Russia

Daniil V. Burdin

Crimean Federal University named after V.I. Vernadsky

Email: daniil-b96@mail.ru
ORCID iD: 0000-0002-2817-1480
SPIN-code: 8608-6393

Junior Researcher, Depart. of Computer Engineering and Modeling

Russian Federation, Simferopol, Russia

References

  1. Yu KH, Beam AL, Kohane IS. Artificial intelligence in healthcare. Nat Biomed Eng. 2018;2:719–731. doi: 10.1038/s41551-018-0305-z.
  2. Gruson D, Bernardini S, Dabla PK, Gouget B, Stankovic S. Collaborative AI and laboratory medicine integration in precision cardiovascular medicine. Clin Chim Acta. 2020;509:67–71. doi: 10.1016/j.cca.2020.06.001.
  3. Ati M. Diagnosis of diabetes mellitus in an E-health environment based on artificial neural network. Journal of Next Generation Information Technology. 2013;4(5):125–132. doi: 10.4156/jnit.vol4.issue5.12.
  4. Uncini A, Aretusi G, Manganelli F, Sekiguchi Y, Magy L, Tozza S, Tsuneyama A, Lefour S, Kuwabara S, Santoro L, Ippoliti L. Electrodiagnostic accuracy in polyneuropathies: Supervised learning algorithms as a tool for practitioners. Neurol Sci. 2020;41:3719–3727. doi: 10.1007/s10072-020-04499-y.
  5. Sambyal N, Saini P, Syal R. A review of statistical and machine learning techniques for microvascular complications in type 2 diabetes. Curr Diabetes Rev. 2020;17(2):143–155. doi: 10.2174/1573399816666200511003357.
  6. Esteva A, Kuprel B, Novoa RA, Ko J, Swetter SM, Blau HM, Thrun S. Dermatologist-level classification of skin cancer with deep neural networks. Nature. 2017;542(7639):115–118. doi: 10.1038/nature21056.
  7. Reina Reina A, Barrera JM, Valdivieso B, Gas ME, Maté A, Trujillo JC. Machine learning model from a Spanish cohort for prediction of SARS-CoV-2 mortality risk and critical patients. Sci Rep. 2022;12(1):5723. doi: 10.1038/s41598-022-09613-y.
  8. Zou Q, Qu K, Luo Y, Yin D, Ju Y, Tang H. Predicting diabetes mellitus with machine learning techniques. Front Genet. 2018;9:515. doi: 10.3389/fgene.2018.00515.
  9. Alghamdi M, Al-Mallah M, Keteyian S, Brawner C, Ehrman J, Sakr S. Predicting diabetes mellitus using SMOTE and ensemble machine learning approach: The Henry Ford ExercIse Testing (FIT) project. PLoS ONE. 2017;12(7):e0179805. doi: 10.1371/journal.pone.0179805.
  10. Ferreira ACBH, Ferreira DD, Oliveira HC, Resende IC, Anjos A, Lopes MHBM. Competitive neural layer-based method to identify people with high risk for diabetic foot. Comput Biol Med. 2020;120:103744. doi: 10.1016/j.compbiomed.2020.103744.
  11. Padhy SK, Takkar B, Chawla R, Kumar A. Artificial intelligence in diabetic retinopathy: A natural step to the future. Indian J Ophthalmol. 2019;67(7):1004–1009. doi: 10.4103/ijo.IJO_1989_18.
  12. Sové RJ, Goldman D, Fraser GM. A computational model of the effect of capillary density variability on oxygen transport, glucose uptake, and insulin sensitivity in prediabetes. Microcirculation. 2017;24(2):e12342. doi: 10.1111/micc.12342.
  13. Senet P, Fichel F, Baudot N, Gaitz JP, Tribout L, Frances C. La capillaroscopie péri-unguéale en dermatologie. [Nail-fold capillaroscopy in dermatology. Ann Dermatol Venereol. 2014;141(6–7):429–437. doi: 10.1016/j.annder.2014.04.120.
  14. Du-Harpur X, Watt FM, Luscombe NM, Lynch MD. What is AI? Applications of artificial intelligence to dermatology. Br J Dermatol. 2020;183(3):423–430. doi: 10.1111/bjd.18880.
  15. Liyanage H, Liaw ST, Jonnagaddala J, Schreiber R, Kuziemsky C, Terry AL, de Lusignan S. Artificial intelligence in primary health care: Perceptions, issues, and challenges. Yearb Med Inform. 2019;28(1):41–46. doi: 10.1055/s-0039-1677901.
  16. Lui S, Li Y, Zhou J, Hu J, Chen N, Shang Y, Chen Z, Li T. Segmenting nailfold capillaries using an improved U-net network. Microvasc Res. 2020;130:104011. doi: 10.1016/j.mvr.2020.104011.
  17. Wells A, Patel S, Lee JB, Motaparthi K. Artificial intelligence in dermatopathology: Diagnosis, education, and research. J Cutan Pathol. 2021;48(8):1061–1068. doi: 10.1111/cup.13954.
  18. Dedov II, Shestakova MV, Mayorov AYu, Vikulova OK, Galstyan GR, Kuraeva TL, Peterkova VA, Smirnova OM, Starostina EG, Surkova EV, Sukhareva OYu, Tokmakova AYu, Shamkhalova MSh, Jarek-Martynova IR, Artemova EV, Beshlieva DD, Bondarenko ON, Volevodz NN, Gomova IS, Grigoryan OR, Dzhemilova ZN, Esayan RM, Ibragimova LI, Kalashnikov VYu, Kononenko IV, Laptev DN, Lipatov DV, Motovilin OG, Nikonova TV, Rozhivanov RV, Shestakova EA. Standards of specialized diabetes care. 9th edition. Sakharnyy diabet. 2019;22(1S1):1–144. (In Russ.)
  19. Lambova SN. Nailfold capillaroscopy — practical implications for rheumatology practice. Curr Rheumatol Rev. 2020;16(2):79–83. doi: 10.2174/157339711602200415083444.
  20. Anwar SM, Majid M, Qayyum A, Awais M, Alnowami M, Khan MK. Medical image analysis using convolutional neural networks: A review. J Med Syst. 2018;42(11):226. doi: 10.1007/s10916-018-1088-1.
  21. Narkevich AN, Vinogradov KA. Object recognition on digital images of microscopic preparations using artificial neural networks. Vestnik novykh me-ditsinskikh tekhnologiy. 2018;25(4):235–241. (In Russ.) doi: 10.24411/1609-2163-2018-16078.
  22. Maldonado G, Guerrero R, Paredes C, Rios C. Nailfold capillaroscopy in diabetes mellitus. Microvasc Res. 2017;112:41–46. doi: 10.1016/j.mvr.2017.03.001.
  23. Uyar S, Balkarli A, Erol MK, Yeşil B, Tokuç A, Durmaz D, Görar S, Çekin AH. Assessment of the relationship between diabetic retinopathy and nailfold capillaries in type 2 diabetics with a noninvasive method: Nailfold videocapillaroscopy. J Diabetes Res. 2016;7592402. doi: 10.1155/2016/7592402.

Supplementary files

Supplementary Files
Action
1. Рис. 1. Результаты работы системы оценки капилляроскопических изображений в норме и при наличии ­микроангиопатии

Download (24KB)
2. Рис. 2. ROC-кривая соотношения чувствительность/специфичность для системы оценки риска развития микроангиопатии

Download (37KB)

© 2022 Eco-Vector





This website uses cookies

You consent to our cookies if you continue to use our website.

About Cookies