# Deep Learning Approaches to Automatic Chronic Venous Disease Classification

^{1}

^{2}

^{3}

^{*}

## Abstract

**:**

## 1. Introduction

## 2. Materials and Methods

#### 2.1. Data Mining

#### 2.1.1. Scrapy Data Mining

#### 2.1.2. Selenium Data Mining

#### 2.1.3. Datasets

#### 2.2. Neural Networks

#### 2.2.1. Filter “Legs–No Legs”

#### 2.2.2. Multi-Classification Problem

- Precision = TruePositive/(TruePositive + FalsePositive)
- Recall = TruePositive/(TruePositive + FalseNegative)
- F-Measure = (2 × Precision × Recall)/(Precision + Recall)
- Logistic Loss curve.

Predicted | |||

Positive | Negative | ||

Actual | Positive | Rated TP = TruePositive/ActualPositive | Rated FN = FalseNegative/ActualPositive |

Negative | Rated FP = FalsePositive/ActualNegative | Rated TN = TrueNegative/ActualNegative |

## 3. Results

#### 3.1. Resnet50 for Filter “Legs–No Legs”

#### 3.2. Resnet50 for the Multi-Classification Problem

#### 3.3. ViT Transformers

#### 3.4. DeiT Multi-Classification Problem

## 4. Discussion

## 5. Conclusions

## Author Contributions

## Funding

## Data Availability Statement

## Conflicts of Interest

## References

- Vuylsteke, M.E.; Colman, R.; Thomis, S.; Guillaume, G.; van Quickenborne, D.; Staelens, I. An epidemiological survey of venous disease among general practitioner attendees in different geographical regions on the globe: The final results of the vein consult program. Angiology
**2018**, 69, 779–785. [Google Scholar] [CrossRef] - Feodor, T.; Baila, S.; Mitea, I.-A.; Branisteanu, D.-E.; Vittos, O. Epidemiology and clinical characteristics of chronic venous disease in Romania. Exp. Ther. Med.
**2019**, 17, 1097–1105. [Google Scholar] [CrossRef] [PubMed] - Carlton, R.; Mallick, R.; Campbell, C.; Raju, A.; O’Donnell, T.; Eaddy, M. Evaluating the expected costs and budget impact of interventional therapies for the treatment of chronic venous disease. Am. Health Drug Benefits
**2015**, 8, 366. [Google Scholar] [CrossRef] [PubMed] - Epstein, D.; Gohel, M.; Heatley, F.; Davies, A.H. Cost-effectiveness of treatments for superficial venous reflux in patients with chronic venous ulceration. BJS Open
**2018**, 2, 203–212. [Google Scholar] [CrossRef] - Vlajinac, H.D.; Radak, Ð.J.; Marinković, J.M.; Maksimović, M.Ž. Risk factors for chronic venous disease. Phlebology
**2012**, 27, 416–422. [Google Scholar] [CrossRef] - Bailey, M.; Solomon, C.; Kasabov, N.; Greig, S. Hybrid systems for medical data analysis and decision making-a case study on varicose vein disorders. In Proceedings of the 1995 Second New Zealand International Two-Stream Conference on Artificial Neural Networks and Expert Systems, Dunedin, New Zealand, 20–23 November 1995; pp. 265–268. [Google Scholar]
- Shad, R.; Cunningham, J.P.; Ashley, E.A.; Langlotz, C.P.; Hiesinger, W. Designing clinically translatable artificial intelligence systems for high-dimensional medical imaging. Nat. Mach. Intell.
**2021**, 3, 929–935. [Google Scholar] [CrossRef] - Bharati, S.; Mondal, M.R.H.; Podder, P.; Prasath, V.B.S. Deep learning for medical image registration: A comprehensive review. Int. J. Comput. Inf. Syst. Ind. Manag. Appl.
**2022**, 14, 173–190. [Google Scholar] - Zhou, S.K.; Greenspan, H.; Davatzikos, C.; Duncan, J.S.; Ginneken, B.V.; Madabhushi, A.; Prince, J.L.; Rueckert, D.; Summers, R.M. A review of deep learning in medical imaging: Imaging traits, technology trends, case studies with progress highlights, and future promises. Proc. IEEE
**2021**, 109, 820–838. [Google Scholar] [CrossRef] - Gergenreter, Y.S.; Zakharova, N.B.; Barulina, M.A.; Maslyakov, V.V.; Fedorov, V.E. Analysis of the cytokine profile of blood serum and tumor supernatants in breast cancer. Acta Biomed. Sci.
**2022**, 7, 134–146. [Google Scholar] [CrossRef] - Wang, Y.B.; You, Z.H.; Yang, S.; Yi, H.-C.; Chen, Z.-H.; Zheng, K. A deep learning-based method for drug-target interaction prediction based on long short-term memory neural network. BMC Med. Inf. Decis. Mak.
**2020**, 20 (Suppl. S2), 49. [Google Scholar] [CrossRef] - Qu, R.; Wang, Y.; Yang, Y. COVID-19 detection using CT image based on YOLOv5 network. In Proceedings of the 2021 3rd International Academic Exchange Conference on Science and Technology Innovation (IAECST), Guangzhou, China, 10–12 December 2021. [Google Scholar]
- Saxena, A.; Singh, S.P. A deep learning approach for the detection of COVID-19 from chest X-Ray images using convolutional neural networks. arXiv
**2022**, arXiv:2201.09952. [Google Scholar] - Gromov, M.S.; Rogacheva, S.M.; Barulina, M.A.; Reshetnikov, A.A.; Prokhozhev, D.A.; Fomina, A.Y. Analysis of some physiological and biochemical indices in patients with COVID-19 pneumonia using mathematical methods. J. Evol. Biochem. Physiol.
**2021**, 57, 1394–1407. [Google Scholar] [CrossRef] [PubMed] - Mohammed, M.A.; Al-Khateeb, B.; Yousif, M.; Mostafa, S.A.; Kadry, S.; Abdulkareem, K.H.; Garcia-Zapirain, B. Novel crow swarm optimization algorithm and selection approach for optimal deep learning COVID-19 diagnostic model. Comput. Intell. Neuralsci.
**2022**, 2022, 1307944. [Google Scholar] [CrossRef] - Wang, X.; Peng, Y.; Lu, L.; Lu, Z.; Bagheri, M.; Summers, R.M. ChestX-Ray8: Hospital-scale Chest X-ray database and benchmarks on weakly-supervised classification and localization of common thorax diseases. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, HI, USA, 21–26 July 2017; pp. 3462–3471. [Google Scholar]
- Ali, O.; Ali, H.; Shah, S.A.A.; Shahzad, A. Implementation of a modified U-Net for medical image segmentation on edge devices. IEEE Trans. Circuits Syst. II Express Briefs 2022, in press. [CrossRef]
- Intel® Neural Compute Stick 2 (Intel® NCS2). Available online: https://www.intel.com/content/www/us/en/developer/tools/neural-compute-stick/overview.html (accessed on 16 September 2022).
- Antonelli, M.; Reinke, A.; Bakas, S.; Farahani, K.; Landman, B.A.; Litjens, G.; Menze, B.; Ronneberger, O.; Summers, R.M.; Ginneken, B.; et al. The medical segmentation decathlon. arXiv
**2021**, arXiv:2106.05735. [Google Scholar] [CrossRef] - Shah, M.I.; Mishra, S.; Yadav, V.K.; Chauhan, A.; Sarkar, M.; Sharma, S.K.; Rout, C. Ziehl–neelsen sputum smear microscopy image database: A resource to facilitate automated bacilli detection for tuberculosis diagnosis. J. Med. Imaging
**2017**, 4, 027503. [Google Scholar] [CrossRef] [PubMed] - Peng, J.; Wang, Y. Medical image segmentation with limited supervision: A review of deep network models. IEEE Access
**2021**, 9, 36827–36851. [Google Scholar] [CrossRef] - Chi, W.; Ma, L.; Wu, J.; Chen, M.; Lu, W.; Gu, X. Deep learning-based medical image segmentation with limited labels. Phys. Med. Biol.
**2020**, 65, 235001. [Google Scholar] [CrossRef] - Butova, X.; Shayakhmetov, S.; Fedin, M.; Zolotukhin, I.; Gianesini, S. Artificial Intelligence Evidence-Based Current Status and Potential for Lower Limb Vascular Management. J. Pers. Med.
**2021**, 11, 1280. [Google Scholar] [CrossRef] - Ryan, L.; Mataraso, S.; Siefkas, A.; Pellegrini, E.; Barnes, G.; Green-Saxena, A. A machine learning approach to predict deep venous thrombosis among hospitalized patients. Clin. Appl. Thromb. Hemost.
**2021**, 27, 1076029621991185. [Google Scholar] [CrossRef] - Nafee, T.; Gibson, C.M.; Travis, R.; Yee, M.K.; Kerneis, M.; Chi, G. Machine learning to predict venous thrombosis in acutely ill medical patients. Res. Pract. Thromb. Haemost.
**2020**, 4, 230–237. [Google Scholar] [CrossRef] [PubMed] - Fong-Mata, M.B.; García-Guerrero, E.E.; Mejía-Medina, D.A.; López-Bonilla, O.R.; Villarreal-Gómez, L.J.; Zamora-Arellano, F.; Inzunza-González, E. An artificial neural network approach and a data augmentation algorithm to systematize the diagnosis of deep-vein thrombosis by using wells’ criteria. Electronics
**2020**, 9, 1810. [Google Scholar] [CrossRef] - Rosenberg, D.; Eichorn, A.; Alarcon, M.; McCullagh, L.; McGinn, T.; Spyropoulos, A.C. External validation of the risk assessment model of the International Medical Prevention Registry on Venous Thromboembolism (IMPROVE) for medical patients in a tertiary health system. J. Am. Heart Assoc.
**2014**, 3, e001152. [Google Scholar] [CrossRef] [PubMed] - Barbar, S.; Noventa, F.; Rossetto, V.; Ferrari, A.; Brandolin, B.; Perlati, M. A risk assessment model for the identification of hospitalized medical patients at risk for venous thromboembolism: The Padua Prediction Score. J. Thromb. Haemost.
**2010**, 8, 2450–2457. [Google Scholar] [CrossRef] - Hippisley-Cox, J.; Coupland, C. Development and validation of risk prediction algorithm (QThrombosis) to estimate future risk of venous thromboembolism: Prospective cohort study. BMJ
**2011**, 343, d4656. [Google Scholar] [CrossRef] - Van Es, N.; Di Nisio, M.; Cesarman, G.; Kleinjan, A.; Otten, H.M.; Mahé, I.; Büller, H.R. Comparison of risk prediction scores for venous thromboembolism in cancer patients: A prospective cohort study. Haematologica
**2017**, 102, 1494. [Google Scholar] [CrossRef] - Ferroni, P.; Zanzotto, F.M.; Scarpato, N.; Riondino, S.; Guadagni, F.; Roselli, M. Validation of a machine learning approach for venous thromboembolism risk prediction in oncology. Dis. Mrk.
**2017**, 2017, 8781379. [Google Scholar] [CrossRef] - Pabinger, I.; van Es, N.; Heinze, G.; Posch, F.; Riedl, J.; Reitter, E.M.; Ay, C. A clinical prediction model for cancer-associated venous thromboembolism: A development and validation study in two independent prospective cohorts. Lancet Haematol.
**2018**, 5, e289–e298. [Google Scholar] [CrossRef] - Zhu, R.; Niu, H.; Yin, N.; Wu, T.; Zhao, Y. Analysis of Varicose Veins of Lower Extremities Based on Vascular Endothelial Cell Inflammation Images and Multi-Scale Deep Learning. IEEE Access
**2019**, 7, 174345–174358. [Google Scholar] [CrossRef] - Atreyapurapu, V.; Vyakaranam, M.I.; Atreya III, S.; Gupta, P.I.; Atturu, G.V. Assessment of Anatomical Changes in Advanced Chronic Venous Insufficiency Using Artificial Intelligence ) and Machine Learning Techniques. J. Vasc. Surg. Venous Lymphat. Disord.
**2022**, 10, 571–572. [Google Scholar] [CrossRef] - Ortega, M.A.; Fraile-Martínez, O.; García-Montero, C.; Álvarez-Mon, M.A.; Chaowen, C.; Ruiz-Grande, F.; Pekarek, L.; Monserrat, J.; Asúnsolo, A.; García-Honduvilla, N.; et al. Understanding chronic venous disease: A critical overview of its pathophysiology and medical management. J. Clin. Med.
**2021**, 10, 3239. [Google Scholar] [CrossRef] [PubMed] - Shi, Q.; Chen, W.; Pan, Y.; Yin, S.; Fu, Y.; Mei, J.; Xue, Z. An Automatic Classification Method on Chronic Venous Insufficiency Images. Sci. Rep.
**2018**, 8, 17952. [Google Scholar] [CrossRef] [PubMed] [Green Version] - Zhang, F.; Song, Y.; Cai, W.; Hauptmann, A.G.; Liu, S.; Pujol, S.; Kikinis, R.; Fulham, M.J.; Feng, D.D.; Chen, M. Dictionary pruning with visual word significance for medical image retrieval. Neuralcomputing
**2016**, 177, 75–88. [Google Scholar] [CrossRef] - Dosovitskiy, A.; Beyer, L.; Kolesnikov, A.; Weissenborn, D.; Zhai, X.; Unterthiner, T.; Houlsby, N. An image is worth 16 × 16 words: Transformers for image recognition at scale. arXiv
**2020**, arXiv:2010.11929. [Google Scholar] - DeiT: Data-Efficient Image Transformers. Available online: https://github.com/facebookresearch/deit/blob/main/README_deit.md (accessed on 17 September 2022).
- Patrício, C.; Neves, J.C.; Teixeira, L.F. Explainable Deep Learning Methods in Medical Imaging Diagnosis: A Survey. arXiv
**2022**, arXiv:2205.04766. [Google Scholar] - Wang, R.; Lei, T.; Cui, R.; Zhang, B.; Meng, H.; Nandi, A.K. Medical Image Segmentation Using Deep Learning: A Survey. arXiv
**2022**, arXiv:2009.13120. [Google Scholar] [CrossRef] - Scrapy. Available online: https://scrapy.org/ (accessed on 25 August 2022).
- Selenium Automates Browsers. That’s it! Available online: https://www.selenium.dev/ (accessed on 25 August 2022).
- He, K.; Zhang, X.; Ren, S.; Sun, J. Deep Residual Learning for Image Recognition. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, USA, 27–30 June 2016; pp. 770–778. [Google Scholar]
- Flickr Image Dataset. Available online: https://www.kaggle.com/datasets/hsankesara/flickr-image-dataset (accessed on 17 September 2022).
- Available online: https://pytorch.org/hub/nvidia_deeplearningexamples_resnet50/ (accessed on 26 August 2022).
- PyTorch. Available online: https://pytorch.org/ (accessed on 25 August 2022).
- Hoens, T.R.; Chawla, N.V. Imbalanced Datasets: From Sampling to Classifiers. In Book Imbalanced Learning: Foundations, Algorithms, and Applications, 1st ed.; Ma, Y., He, H., Eds.; Wiley: Hoboken, NJ, USA, 2013; pp. 43–60. [Google Scholar]

**Figure 2.**Images from the Flickr Image dataset [45].

**Figure 7.**The multi-classification model based on Resnet50. The color scheme for confusion matrices (

**a**); the confusion matrices for C0 (

**b**), for C1 (

**c**), and C2 (

**d**) classes.

**Figure 8.**The multi-classification model based on Resnet50. The confusion matrices for C3 (

**a**), C4 (

**b**), C5 (

**c**), and C6 (

**d**) classes.

**Figure 10.**The multi-classification model based on vit-base-patch16-224. The confusion matrices for C0 (

**a**), C1 (

**b**), C2 (

**c**), C3 (

**d**), C4 (

**e**), C5 (

**f**), and C6 (

**g**) classes.

**Figure 11.**The multi-classification model based on vit-base-patch16-384. The confusion matrices for C0 (

**a**), C1 (

**b**), C2 (

**c**), C3 (

**d**), C4 (

**e**), C5 (

**f**), and C6 (

**g**) classes.

**Figure 13.**The color scheme for confusion matrices for the multi-classification model based on DeiT (

**a**) and the confusion matrices for the C0 class (

**b**).

**Figure 14.**The multi-classification model based on DeiT. The confusion matrices for C1 (

**a**), C2 (

**b**), C3 (

**c**), C4 (

**d**), C5 (

**e**), and C6 (

**f**) classes.

Parameter | vit-base-patch16-224 | vit-base-patch16-384 |
---|---|---|

hidden_size | 768 | 768 |

image_size | 224 | 384 |

num_hidden_layers | 12 | 12 |

patch_size | 16 | 16 |

NN | Precision | Recall | F1 Score |
---|---|---|---|

Resnet50 | 0.62 | 0.61 | 0.61 |

vit-base-patch16-224 | 0.75 | 0.75 | 0.75 |

vit-base-patch16-384 | 0.79 | 0.79 | 0.79 |

DeiT | 0.77 | 0.77 | 0.77 |

Model | NN | Rated TP | Rated TN | Rated FP | Rated FN |
---|---|---|---|---|---|

C0 | Resnet50 | 0.80 | 0.97 | 0.20 | 0.029 |

vit-base-patch16-224 | 0.61 | 0.98 | 0.39 | 0.019 | |

vit-base-patch16-384 | 0.71 | 0.98 | 0.29 | 0.015 | |

DeiT | 0.76 | 0.99 | 0.24 | 0.001 | |

C1 | Resnet50 | 0.79 | 0.91 | 0.21 | 0.086 |

vit-base-patch16-224 | 0.78 | 0.94 | 0.22 | 0.057 | |

vit-base-patch16-384 | 0.83 | 0.96 | 0.17 | 0.042 | |

DeiT | 0.86 | 0.94 | 0.14 | 0.055 | |

C2 | Resnet50 | 0.52 | 0.90 | 0.48 | 0.098 |

vit-base-patch16-224 | 0.67 | 0.92 | 0.33 | 0.082 | |

vit-base-patch16-384 | 0.71 | 0.93 | 0.29 | 0.07 | |

DeiT | 0.63 | 0.95 | 0.37 | 0.055 | |

C3 | Resnet50 | 0.60 | 0.85 | 0.40 | 0.150 |

vit-base-patch16-224 | 0.84 | 0.89 | 0.16 | 0.11 | |

vit-base-patch16-384 | 0.85 | 0.91 | 0.15 | 0.087 | |

DeiT | 0.83 | 0.90 | 0.17 | 0.099 | |

C4 | Resnet50 | 0.47 | 0.91 | 0.53 | 0.085 |

vit-base-patch16-224 | 0.67 | 0.96 | 0.33 | 0.039 | |

vit-base-patch16-384 | 0.75 | 0.96 | 0.25 | 0.038 | |

DeiT | 0.70 | 0.94 | 0.30 | 0.058 | |

C5 | Resnet50 | 0.29 | 0.91 | 0.71 | 0.030 |

vit-base-patch16-224 | 0.6 | 0.99 | 0.40 | 0.009 | |

vit-base-patch16-384 | 0.59 | 0.99 | 0.41 | 0.007 | |

DeiT | 0.40 | 0.99 | 0.60 | 0.014 | |

C6 | Resnet50 | 0.40 | 0.99 | 0.60 | 0.012 |

vit-base-patch16-224 | 0.60 | 0.99 | 0.40 | 0.005 | |

vit-base-patch16-384 | 0.79 | 1.00 | 0.21 | 0.004 | |

DeiT | 0.55 | 0.99 | 0.45 | 0.009 |

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |

© 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Barulina, M.; Sanbaev, A.; Okunkov, S.; Ulitin, I.; Okoneshnikov, I.
Deep Learning Approaches to Automatic Chronic Venous Disease Classification. *Mathematics* **2022**, *10*, 3571.
https://doi.org/10.3390/math10193571

**AMA Style**

Barulina M, Sanbaev A, Okunkov S, Ulitin I, Okoneshnikov I.
Deep Learning Approaches to Automatic Chronic Venous Disease Classification. *Mathematics*. 2022; 10(19):3571.
https://doi.org/10.3390/math10193571

**Chicago/Turabian Style**

Barulina, Marina, Askhat Sanbaev, Sergey Okunkov, Ivan Ulitin, and Ivan Okoneshnikov.
2022. "Deep Learning Approaches to Automatic Chronic Venous Disease Classification" *Mathematics* 10, no. 19: 3571.
https://doi.org/10.3390/math10193571