Current issue
Archive
Online first
About the Journal
Editorial Office
Editorial Board
Publisher
Editorial Policies
Ethical standards and procedures
Abstracting and indexing
Reviewers
Honorary Reviewers
Subscription
Contact
Most read articles
Instructions for authors
Article processing charge (APC)
Books and Events
Books
Events
ORIGINAL PAPER
Automatic assessment of nailfold capillaroscopy software: a pilot study
1
Department of Rheumatology, Medical University of Lodz, Poland
2
University of Economics and Human Sciences in Warsaw, Poland
Submission date: 2024-07-11
Final revision date: 2024-08-16
Acceptance date: 2024-10-02
Online publication date: 2024-11-09
Corresponding author
Olga Elżbieta Brzezińska
Department of Rheumatology, Medical University of Lodz, 113 Stefana Żeromskiego St., 90-543 Lodz, Poland, e-mail: olga.brzezinska@umed.lodz.pl
Department of Rheumatology, Medical University of Lodz, 113 Stefana Żeromskiego St., 90-543 Lodz, Poland, e-mail: olga.brzezinska@umed.lodz.pl
KEYWORDS
TOPICS
ABSTRACT
Introduction:
Capillaroscopy is a simple method of nailfold capillary imaging, used to diagnose diseases from the systemic sclerosis spectrum. However, the assessment of the capillary image is time-consuming and subjective. This makes it difficult to use for a detailed comparison of studies assessed by various physicians. This pilot study aimed to validate software used for automatic capillary counting and image classification as normal or pathological.
Material and methods:
The study was based on the assessment of 200 capillaroscopic images obtained from patients suffering from systemic sclerosis or scleroderma spectrum diseases and healthy people. Dinolite MEDL4N Pro was used to perform capillaroscopy. Each image was analysed manually and described using working software. The neural network was trained using the fast.ai library (based on PyTorch). The ResNet-34 deep residual neural network was chosen; 10-fold cross-validation with the validation and test set was performed, using the Darknet-YoloV3 state of the art neural network in a GPU-optimized (P5000 GPU) environment. For the calculation of 1 mm capillaries, an additional detection mechanism was designed.
Results:
The results obtained under neural network training were compared to the results obtained in manual analysis. The sensitivity of the automatic tool relative to manual assessment in classification of correct vs. pathological images was 89.0%, specificity 89.4% for the training group, in validation 89.0% and 86.9% respectively. For the average number of capillaries in 1 mm the precision of real images detected within the region of interest was 96.48%.
Conclusions:
The pilot software for fully automatic capillaroscopic image assessment can be a useful tool for the rapid classification of a normal and altered capillaroscopy pattern. In addition, it allows one to quickly calculate the number of capillaries. In the future, the tool will be developed and will make it possible to obtain full imaging characteristics independent of the experience of the examiner.
Capillaroscopy is a simple method of nailfold capillary imaging, used to diagnose diseases from the systemic sclerosis spectrum. However, the assessment of the capillary image is time-consuming and subjective. This makes it difficult to use for a detailed comparison of studies assessed by various physicians. This pilot study aimed to validate software used for automatic capillary counting and image classification as normal or pathological.
Material and methods:
The study was based on the assessment of 200 capillaroscopic images obtained from patients suffering from systemic sclerosis or scleroderma spectrum diseases and healthy people. Dinolite MEDL4N Pro was used to perform capillaroscopy. Each image was analysed manually and described using working software. The neural network was trained using the fast.ai library (based on PyTorch). The ResNet-34 deep residual neural network was chosen; 10-fold cross-validation with the validation and test set was performed, using the Darknet-YoloV3 state of the art neural network in a GPU-optimized (P5000 GPU) environment. For the calculation of 1 mm capillaries, an additional detection mechanism was designed.
Results:
The results obtained under neural network training were compared to the results obtained in manual analysis. The sensitivity of the automatic tool relative to manual assessment in classification of correct vs. pathological images was 89.0%, specificity 89.4% for the training group, in validation 89.0% and 86.9% respectively. For the average number of capillaries in 1 mm the precision of real images detected within the region of interest was 96.48%.
Conclusions:
The pilot software for fully automatic capillaroscopic image assessment can be a useful tool for the rapid classification of a normal and altered capillaroscopy pattern. In addition, it allows one to quickly calculate the number of capillaries. In the future, the tool will be developed and will make it possible to obtain full imaging characteristics independent of the experience of the examiner.
REFERENCES (30)
1.
Hughes M, Herrick AL. Raynaud’s phenomenon. Best Pract Res Clin Rheumatol 2016; 30: 112–132, DOI: 10.1016/j.berh.2016.04.001.
2.
Piette Y, Reynaert V, Vanhaecke A, et al. Standardised interpretation of capillaroscopy in autoimmune idiopathic inflammatory myopathies: a structured review on behalf of the EULAR study group on microcirculation in rheumatic diseases. Autoimmun Rev 2022; 21: 103087, DOI: 10.1016/J.AUTREV.2022.103087.
3.
Ruaro B, Smith V, Sulli A, et al. Methods for the morphological and functional evaluation of microvascular damage in systemic sclerosis. Korean J Intern Med 2015; 30: 1–5, DOI: 10.3904/kjim.2015.30.1.1.
4.
Smith V, Riccieri V, Pizzorni C, et al. Nailfold capillaroscopy for prediction of novel future severe organ involvement in systemic sclerosis. J Rheumatol 2013; 40: 2023–2028, DOI: 10.3899/JRHEUM.130528.
5.
Melsens K, Leone MC, Paolino S, et al. Nailfold capillaroscopy in Sjögren’s syndrome: a systematic literature review and standardised interpretation. Clin Exp Rheumatol 2020; 38 Suppl 126: S150-S157.
6.
Graceffa D, Amorosi B, Maiani E, et al. Capillaroscopy in psoriatic and rheumatoid arthritis: a useful tool for differential diagnosis. Arthritis 2013; 2013, DOI: 10.1155/2013/957480.
7.
Maldonado G, Guerrero R, Paredes C, Ríos C. Nailfold capillaroscopy in diabetes mellitus. Microvasc Res 2017; 112: 41–46, DOI: 10.1016/j.mvr.2017.03.001.
8.
Natalello G, De Luca G, Gigante L, et al. Nailfold capillaroscopy findings in patients with coronavirus disease 2019: broadening the spectrum of COVID-19 microvascular involvement. Microvasc Res 2021; 133, DOI: 10.1016/j.mvr.2020.104071.
9.
Boulon C, Blaise S, Lazareth I, et al. Reproducibility of the scleroderma pattern assessed by wide-field capillaroscopy in subjects suffering from Raynaud’s phenomenon. Rheumatology (Oxford) 2017; 56: 1780–1783, DOI: 10.1093/rheumatology/kex282.
10.
Smith V, Beeckman S, Herrick AL, et al. An EULAR study group pilot study on reliability of simple capillaroscopic definitions to describe capillary morphology in rheumatic diseases. Rheumatology (Oxford) 2016; 55: 883–890, DOI: 10.1093/rheumatology/kev441.
11.
Dinsdale G, Moore T, O’Leary N, et al. Intra-and inter-observer reliability of nailfold videocapillaroscopy – a possible outcome measure for systemic sclerosis-related microangiopathy. Microvasc Res 2017; 112: 1–6, DOI: 10.1016/J.MVR.2017.02.001.
12.
Haug CJ, Drazen JM. Artificial intelligence and machine learning in clinical medicine, 2023. N Engl J Med 2023; 388: 1201–1208, DOI: 10.1056/NEJMRA2302038.
13.
Van Den Hoogen F, Khanna D, Fransen J, et al. 2013 classification criteria for systemic sclerosis: an American college of rheumatology/European league against rheumatism collaborative initiative. Ann Rheum Dis 2013; 72: 1747–1755, DOI: 10.1136/ANNRHEUMDIS-2013-204424.
14.
Eden M, Wilkinson S, Murray A, et al. Nailfold capillaroscopy: a survey of current UK practice and “next steps” to increase uptake among rheumatologists. Rheumatology (Oxford) 2022; 62: 335–340, DOI: 10.1093/rheumatology/keac320.
15.
Howard J, Gugger S. Fastai: a layered API for deep learning. Information 2020; 11: 108, DOI: 10.3390/info11020108.
16.
Redmon J, Farhadi A. YOLOv3: an incremental improvement. 2018. Available at: https://arxiv.org/abs/1804.027... (Access: 26.06.2024).
17.
Hu Q, Mahler F. New system for image analysis in nailfold capillaroscopy. Microcirculation 1999; 6: 227–235, DOI: 10.1111/j.1549-8719.1999.tb00105.x.
18.
Bharathi PG, Berks M, Dinsdale G, et al. A deep learning system for quantitative assessment of microvascular abnormalities in nailfold capillary images. Rheumatology (Oxford) 2023; 62: 2325–2329. DOI: 10.1093/rheumatology/kead026.
19.
Berks M, Tresadern P, Dinsdale G, et al. An automated system for detecting and measuring nailfold capillaries. Med Image Comput Comput Assist Interv 2014; 17 (Pt 1): 658–665. DOI: 10.1007/978-3-319-10404-1_82.
20.
Berks M, Dinsdale G, Murray A, et al. Automated structure and flow measurement – a promising tool in nailfold capillaroscopy. Microvasc Res 2018; 118: 173–177, DOI: 10.1016/j.mvr.2018.03.016.
21.
Gronenschild EHBM, Muris DMJ, Schram MT, et al. Semi-automatic assessment of skin capillary density: proof of principle and validation. Microvasc Res 2013; 90: 192–198, DOI: 10.1016/j.mvr.2013.08.003.
22.
Sulli A, Vanhaecke A, Pizzorni C, et al. Automated Nailfold Capillary Counting System (AUTOCAPI) in systemic sclerosis patients with diferent capillaroscopic patterns. Arthritis Rheumatol 2019; 71 (suppl 10): 1642. Available at: https://acrabstracts.org/abstr... (Access: 29.10.2024).
23.
Schaefer G, Krawczyk B, Doshi NP, Merla A. Scleroderma capillary pattern identification using texture descriptors and ensemble classification. Annu Int Conf IEEE Eng Med Biol Soc 2013; 2013: 5473–5476, DOI: 10.1109/EMBC.2013.6610788.
24.
Yin H, Wu Z, Huang A, et al. Automated nailfold capillary density measurement method based on improved YOLOv5. Microvasc Res 2023; 150: 104593, DOI: 10.1016/J.MVR.2023.104593.
25.
Cutolo M, Trombetta AC, Melsens K, et al. Automated assessment of absolute nailfold capillary number on videocapillaroscopic images: proof of principle and validation in systemic sclerosis. Microcirculation 2018; 25: e12447, DOI: 10.1111/MICC.12447.
26.
Murray AK, Feng K, Moore TL, et al. Preliminary clinical evaluation of semi-automated nailfold capillaroscopy in the assessment of patients with Raynaud’s phenomenon. Microcirculation 2011; 18: 440–447, DOI: 10.1111/j.1549-8719.2011.00104.x.
27.
Tello BG, Ibáñez ER, Mateo PF, et al. The challenge of comprehensive nailfold videocapillaroscopy practice: a further contribution. Clin Exp Rheumatol 2022; 40: 1926–1932, DOI: 10.55563/clinexprheumatol/6USCE8.
28.
Garaiman A, Nooralahzadeh F, Mihai C, et al. Vision transformer assisting rheumatologists in screening for capillaroscopy changes in systemic sclerosis: an artificial intelligence model. Rheumatology (Oxford) 2023; 62: 2492–2500, DOI: 10.1093/rheumatology/keac541.
29.
Kassani PH, Ehwerhemuepha L, Martin-King C, et al. Artificial intelligence for nailfold capillaroscopy analyses – a proof of concept application in juvenile dermatomyositis. Pediatr Res 2024; 95: 981–987, DOI: 10.1038/S41390-023-02894-7.
30.
Shah R, Petch J, Nelson W, et al. Nailfold capillaroscopy and deep learning in diabetes. J Diabetes 2023; 15: 145–151, DOI: 10.1111/1753-0407.13354.
Copyright: © Narodowy Instytut Geriatrii, Reumatologii i Rehabilitacji w Warszawie. This is an Open Access journal, all articles are distributed under the terms of the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0) License (https://creativecommons.org/licenses/by-nc-sa/4.0/), allowing third parties to copy and redistribute the material in any medium or format and to remix, transform, and build upon the material, provided the original work is properly cited and states its license.
Share
RELATED ARTICLE
| eISSN: | 2084-9834 |
| ISSN: | 0034-6233 |
We process personal data collected when visiting the website. The function of obtaining information about users and their behavior is carried out by voluntarily entered information in forms and saving cookies in end devices. Data, including cookies, are used to provide services, improve the user experience and to analyze the traffic in accordance with the Privacy policy. Data are also collected and processed by Google Analytics tool (more).
You can change cookies settings in your browser. Restricted use of cookies in the browser configuration may affect some functionalities of the website.
You can change cookies settings in your browser. Restricted use of cookies in the browser configuration may affect some functionalities of the website.
