Dynamic Conditional Encoding and Feature Frequency Parsing in Diffusion Probabilistic Models for Diabetic Foot Ulcer Detection Using Thermographic Imaging
Kousar Nikhath A
(Login required)
VNR Vignana Jyothi Institute of Engienering and Technology, Hyderabad, Telangana, India
Paper #9193 received 4 Dec 2024; revised manuscript received 10 Jan 2025; accepted for publication 22 Jan 2025; published online 26 Mar 2025.
DOI: 10.18287/JBPE25.11.010304
Abstract
Diabetic patients’ odds of survival can be significantly increased by early identification of diabetic foot ulcers (DFU) Thermography is utilised to identify changes in planter temperature because diabetic foot causes planter ulcers. This study uses publicly accessible thermographic imaging data from patients in the diabetes group as to ll as the control group. For DFU detection, both image-level and patch-level thermograms are used. The work suggests an optimizer-based Diffusion Probabilistic Segmentation model (Op-DPS) for DFU identification. The work suggests Dynamic Conditional Encoding, which creates state-adaptive conditions for every sampling step, to progress the step-wise regional attention in DPS for segmentation tasks. In moderate the detrimental impact of high-frequency noise components in this approach, to also suggest the Feature Frequency Parser (FF-Parser). The paper presents a Football Optimisation Algorithm (FbOA) that uses high dimensionality, nonlinearity, and many local optima to fine-tune the parameters of the suggested model. FbOA is based on the strategic balance between exploration and corruption that is seen in football gameplay. The proposed Op-DPS model achieved exceptional performance with a precision of 89.55%, recall of 85.53%, F1 score of 87.50%, Intersection-over-Union (IoU) of 88.96%, and an overall accuracy of 99.27% on diabetic foot ulcer segmentation tasks. These results demonstrate significant improvements over existing methods, highlighting the model’s capability for accurate and reliable DFU detection. In order to efficiently explore and take advantage of the solution space, the algorithm incorporates short passes, long passes, and positional modifications to replicate players’ tactical placement and movement. Image-level, patch-level, and image–patch combination data are used to generate the findings.
Keywords
Full Text:
PDFReferences
1. C C. Kendrick, B. Cassidy, J. M. Pappachan, C. O’Shea, C. J. Fernandez, E. Chacko, K. Jacob, N. D. Reeves, and M. H. Yap, “Translating Clinical Delineation of Diabetic Foot Ulcers into Machine Interpretable Segmentation,” in Diabetic Foot Ulcers Grand Challenge, M. H. Yap, C. Kendrick, and R. Brüngel (Eds.), Springer Nature Switzerland, 15335 (2025).
2. A. Mahbod, G. Schaefer, R. Ecker, and I. Ellinger, “Automatic foot ulcer segmentation using an ensemble of convolutional neural networks,” in 26th International Conference on Pattern Recognition (ICPR), IEEE, 4358−4364 (2021).
3. R. V. Prakash, K. S. Kumar, “Development of automatic segmentation techniques using convolutional neural networks to differentiate diabetic foot ulcers,” International Journal of Advanced Computer Science and Applications 13(11), (2022).
4. J. Reyes-Luévano, J. A. Guerrero-Viramontes, J. Rubén Romo-Andrade, and M. Funes-Gallanzi, “DFU_VIRNet: A novel Visible-InfraRed CNN to improve diabetic foot ulcer classification and early detection of ulcer risk zones,” Biomedical Signal Processing and Control 86, 105341 (2023).
5. D. Kucharski, A. Kostuch, F. Noworolnik, A. Brodzicki, and J. Jaworek-Korjakowska, “DFU-Ens: End-to-End diabetic foot ulcer segmentation framework with vision transformer based detection,” in Diabetic Foot Ulcers Grand Challenge, M. H. Yap, C. Kendrick, and B. Cassidy (Eds.), Springer International Publishing, 13797, 101–112 (2023).
6. T.-Y. Liao, C.-H. Yang, Y.-W. Lo, K.-Y. Lai, P.-H. Shen, and Y.-L. Lin, “HarDNet-DFUS: enhancing backbone and decoder of HarDNet-MSEG for diabetic foot ulcer image segmentation,” in Diabetic Foot Ulcers Grand Challenge, M. H. Yap, C. Kendrick, and B. Cassidy (Eds.), Springer International Publishing, 13797, 21–30 (2023).
7. H.-N. Huang, T. Zhang, C.-T. Yang, Y.-J. Sheen, H.-M. Chen, C.-J. Chen, and M.-W. Tseng, “Image segmentation using transfer learning and Fast R-CNN for diabetic foot wound treatments,” Frontiers in Public Health 10, 969846 (2022).
8. A. Kairys, R. Pauliukiene, V. Raudonis, and J. Ceponis, “Towards home-based diabetic foot ulcer monitoring: a systematic review,” Sensors 23(7), 3618 (2023).
9. B. Gunapriya, T. Rajesh, A. Thirumalraj, and B. Manjunatha, “LW-CNN-based extraction with optimized encoder-decoder model for detection of diabetic retinopathy,” Journal of Autonomous Intelligence 7(3), (2023).
10. H. J. Chae, S. Lee, H. Son, S. Han, and T. Lim, “Generating 3D bio-printable patches using wound segmentation and reconstruction to treat diabetic foot ulcers,” Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition 2539–2549 (2022).
11. D. Bouallal, H. Douzi, and R. Harba, “Diabetic foot thermal image segmentation using Double Encoder-ResUnet (DE-ResUnet),” Journal of Medical Engineering & Technology 46(5), 378–392 (2022).
12. H. Yi, W. Xu, Z. Jiang, J. Gao, Q. Kang, Q. Lao, and K. Li, “OCRNet for diabetic foot ulcer segmentation combined with edge loss,” Diabetic Foot Ulcers Grand Challenge 13797, 31–39 (2023).
13. M. Hassib, M. Ali, A. Mohamed, M. Torki, and M. Hussein, “Diabetic foot ulcer segmentation using convolutional and transformer-based models,” in Diabetic Foot Ulcers Grand Challenge, M. H. Yap, C. Kendrick, and B. Cassidy (Eds.), Springer International Publishing, 13797, 83–91 (2023).
14. A. Kairys, V. Raudonis, “Analysis of training data augmentation for diabetic foot ulcer semantic segmentation,” Electronics 12(22), 4624 (2023).
15. P. Xu, X. Wu, Y. Li, E. U. Haq, J. Yin, and K. Li, “Ensemble learning for diabetic foot ulcer segmentation based on DFUC2022 dataset,” Proceedings of the 2022 6th International Conference on Electronic Information Technology and Computer Engineering, 1750–1754 (2022).
16. R. Brüngel, S. Koitka, and C. M. Friedrich, “Unconditionally generated and pseudo-labeled synthetic images for diabetic foot ulcer segmentation dataset extension,” in Diabetic Foot Ulcers Grand Challenge, M. H. Yap, C. Kendrick, and B. Cassidy (Eds.), Springer International Publishing, 13797, 65–79 (2023).
17. J. Zhang, Y. Qiu, L. Peng, Q. Zhou, Z. Wang, and M. Qi, “A comprehensive review of methods based on deep learning for diabetes-related foot ulcers,” Frontiers in Endocrinology 13, 945020 (2022).
18. G. Scebba, J. Zhang, S. Catanzaro, C. Mihai, O. Distler, M. Berli, and W. Karlen, “Detect-and-segment: A deep learning approach to automate wound image segmentation,” Informatics in Medicine Unlocked 29, 100884 (2022).
19. C. Cao, Y. Qiu, Z. Wang, J. Ou, J. Wang, A. H. Hounye, M. Hou, Q. Zhou, and J. Zhang, “Nested segmentation and multi-level classification of diabetic foot ulcer based on mask R-CNN,” Multimedia Tools and Applications 82(12), 18887–18906 (2023).
20. J. P., S. K. B. K., and S. Jayaraman, “Automatic foot ulcer segmentation using conditional generative adversarial network (AFSegGAN): A wound management system,” PLOS Digit Health 2(11), e0000344 (2023).
21. R. Sarmun, M. E. H. Chowdhury, M. Murugappan, A. Aqel, M. Ezzuddin, S. M. Rahman, A. Khandakar, S. Akter, R. Alfkey, and A. Hasan, “Diabetic foot ulcer detection: combining deep learning models for improved localization,” Cognitive Computation 16(3), 1413–1431 (2024).
22. A. M. El-Kady, M. M. Abbassy, H. H. Ali, and M. F. Ali, “Advancing diabetic foot ulcer detection based on resnet and gan integration,” Journal of Theoretical and Applied Information Technology 102(6), 2258–2268 (2024).
23. R. Basiri, K. Manji, P. M. LeLievre, J. Toole, F. Kim, S. S. Khan, and M. R. Popovic, “Protocol for metadata and image collection at diabetic foot ulcer clinics: enabling research in wound analytics and deep learning,” BioMedical Engineering OnLine 23(1), 12 (2024).
24. S. Biswas, R. Mostafiz, B. K. Paul, K. M. M. Uddin, Md. A. Hadi, and F. Khanom, “DFU_XAI: A deep learning-based approach to diabetic foot ulcer detection using feature explainability,” Biomedical Materials & Devices 2(2), 1225–1245 (2024).
25. N. F. Almufadi, H. F. Alhasson, and S. S. Alharbi, “E-DFu-Net: An efficient deep convolutional neural network models for diabetic foot ulcer classification,” Biomolecules and Biomedicine 25(2), 445–460 (2025).
26. C. Thota, D. Jackson Samuel, M. Musa Jaber, M. M. Kamruzzaman, R. V. Ravi, L. J. Gnanasigamani, and R. Premalatha, “Image smart segmentation analysis against diabetic foot ulcer using internet of things with virtual sensing,” Big Data 12(2), 155–172 (2024).
27. S. K. Das, S. Namasudra, and A. K. Sangaiah, “HCNNet: hybrid convolution neural network for automatic identification of ischaemia in diabetic foot ulcer wounds,” Multimedia Systems 30(1), 36 (2024).
28. M. K. Dhar, T. Zhang, Y. Patel, S. Gopalakrishnan, and Z. Yu, “FUSegNet: A deep convolutional neural network for foot ulcer segmentation,” Biomedical Signal Processing and Control 92, 106057 (2024).
29. S. Biswas, R. Mostafiz, M. S. Uddin, and B. K. Paul, “XAI-FusionNet: Diabetic foot ulcer detection based on multi-scale feature fusion with explainable artificial intelligence,” Heliyon 10(10), e31228 (2024).
30. N. Zhao, L. Yu, X. Fu, W. Dai, H. Han, J. Bai, J. Xu, J. Hu, and Q. Zhou, “Application of a Diabetic Foot Smart APP in the measurement of diabetic foot ulcers,” International Journal of Orthopaedic and Trauma Nursing 54, 101095 (2024).
31. F. M. Lusendi, A.-S. Vanherwegen, F. Nobels, and G. A. Matricali, “A multidisciplinary Delphi consensus to define evidence-based quality indicators for diabetic foot ulcer care,” European Journal of Public Health 34(2), 253–259 (2024).
32. S. Hong, Y. Chen, Y. Lin, X. Xie, G. Chen, H. Xie, and W. Lu, “Personalized prediction of diabetic foot ulcer recurrence in elderly individuals using machine learning paradigms,” Technology and Health Care 32, 265–276 (2024).
33. M. Shiraishi, H. Lee, K. Kanayama, Y. Moriwaki, and M. Okazaki, “Appropriateness of artificial intelligence chatbots in diabetic foot ulcer management,” The International Journal of Lower Extremity Wounds 15347346241236811 (2024).
34. M.G Sumithra, C. Venkatesan, “SwinDFU-Net: Deep learning transformer network for infection identification in diabetic foot ulcer,” Technology and Health Care 33(1), 601–618 (2025).
35. C. Wu, C. Xu, S. Ou, X. Wu, J. Guo, Y. Qi, and S. Cai, “A novel approach for diabetic foot diagnosis: Deep learning-based detection of lower extremity arterial stenosis,” Diabetes Research and Clinical Practice 207, 111032 (2024).
36. D. A. Hernandez-Contreras, H. Peregrina-Barreto, J. de J. Rangel-Magdaleno, and F. J. Renero-Carrillo, “Plantar thermogram database for the study of diabetic foot complications,” IEEE Access 7, 161296–161307 (2019).
37. I. Khosa, A. Raza, M. Anjum, W. Ahmad, and S. Shahab, “Automatic diabetic foot ulcer recognition using multi-level thermographic image data,” Diagnostics 13(16), 2637 (2023).
38. E.-S. M. El-Kenawy, F. H. Rizk, A. M. Zaki, M. E. Mohamed, A. Ibrahim, A. A. Abdelhamid, N. Khodadadi, E. M. Almetwally, and M. M. Eid, “Football Optimization Algorithm (FbOA): A novel metaheuristic inspired by team strategy dynamics,” Journal of Artificial Intelligence and Metaheuristics 8(1), 21–38 (2024).
39. R. Aluvalu, T. Sharma, U. M. Viswanadhula, A. D. Thirumalraj, M. V. V. Prasad Kantipudi, and S. Mudrakola, “Komodo dragon mlipir algorithm-based CNN model for detection of illegal tree cutting in smart iot forest area,” Recent Advances in Computer Science and Communications 17(6), e260124226370 (2024)
40. V. Uma Maheswari, S. Stephe, R. Aluvalu, A. Thirumalraj, and S. N. Mohanty, “Chaotic satin bowerbird optimizer based advanced ai techniques for detection of COVID-19 diseases from CT scans images,” New Generation Computing 42(5), 1065–1087 (2024).
Сontact
34 Moskovskoe shosse, Samara, 443086, Russian Federation
Email: j-bpe@ssau.ru
Phone: +7-846-267-4550
© 2014-2025 J-BPE