Automatic Area Estimation of Mice Wound Images

• Bruno Uhlmann Marcato UTFPR
• Camila Rodrigues Ferraz University of Maryland Baltimore
• Waldiceu Aparecido Verri Jr UEL
• Rubia Casagrande UEL
• Daniel Prado Campos UTFPR
• José Luis Seixas Junior Eötvös Loránd University
• Rafael Gomes Mantovani UTFPR

DOI: https://doi.org/10.5753/kdmile.2024.241973

Resumo

Image segmentation is a classic computer vision set of techniques that partitions a digital image into discrete groups of pixel-image segments to inform object detection and related tasks. It has been successfully explored in biological studies, such as in the identification of wounds. However, recent approaches towards using black-box deep learning algorithms for image and semantic segmentation of objects have higher computational costs than classic techniques. In this study, we evaluated the effectiveness of thresholding and deep learning techniques for semantic segmentation of wound images of mice. Experiments were performed with a real dataset developed by the Pain, Neuropathy, and Inflammation Laboratory at the State University of Londrina with the approval of the University Ethics Committee on Animal Research and Welfare. The results were promising, showing that deep learning and thresholding were able to recognize wound areas, with an average IoU of 0.75 and 0.72, respectively. However, when estimating the wound areas, deep learning results were the most close to the ground truth.

Palavras-chave: area estimation, image segmentation, thresholding, wound identification

Referências

Aggarwal, C. C. Neural Networks and Deep Learning: A Textbook. Springer International Publishing, Cham, 2018.

Alsahafi, Y. S., Elshora, D. S., Mohamed, E. R., and Hosny, K. M. Multilevel threshold segmentation of skin lesions in color images using coronavirus optimization algorithm. Diagnostics 13 (18), 2023.

Azad, R., Aghdam, E. K., Rauland, A., Jia, Y., Avval, A. H., Bozorgpour, A., Karimijafarbigloo, S., Cohen, J. P., Adeli, E., and Merhof, D. Medical image segmentation review: The success of u-net, 2022.

Breiman, L. Random forests. Machine learning 45 (1): 5–32, 2001.

Csurka, G., Volpi, R., and Chidlovskii, B. Semantic image segmentation: Two decades of research, 2023.

Goyzueta, C. A. R., De la Cruz, J. E. C., and Machaca, W. A. M. Integration of u-net, resu-net and deeplab architectures with intersection over union metric for cells nuclei image segmentation. In 2021 IEEE Engineering International Research Conference (EIRCON). pp. 1–4, 2021.

Hosny, K. M., Khalid, A. M., Hamza, H. M., and Mirjalili, S. Multilevel thresholding satellite image segmentation using chaotic coronavirus optimization algorithm with hybrid fitness function. Neural Computing and Applications 35 (1): 855–886, Jan, 2023.

Huang, Y., Tang, Z., Chen, D., Su, K., and Chen, C. Batching soft iou for training semantic segmentation networks. IEEE Signal Processing Letters vol. 27, pp. 66–70, 2020.

Kang, B. and Nguyen, T. Q. Random forest with learned representations for semantic segmentation. IEEE Transactions on Image Processing 28 (7), 2019.

Kaymak, R., Kaymak, C., and Ucar, A. Skin lesion segmentation using fully convolutional networks: A comparative experimental study. Expert Systems with Applications vol. 161, pp. 113742, 2020.

Long, J., Shelhamer, E., and Darrell, T. Fully convolutional networks for semantic segmentation, 2015.

Manakitsa, N., Maraslidis, G. S., Moysis, L., and Fragulis, G. F. A review of machine learning and deep learning for object detection, semantic segmentation, and human action recognition in machine and robotic vision. Technologies 12 (2), 2024.

Marcato, B. U., Pierotti, S. M., Ritter, P. D., Ferraz, C. R., Verri Jr, W. A., Casagrande, R., Seixas Junior, J. L., and Mantovani, R. G. Semantic segmentation of mice wounds. In Anais do XV Computer on the Beach, 10 a 13 de abril de 2024. pp. 23–29, 2024.

Marsland, S. Machine learning: an algorithmic perspective. CRC press, 2015.

Minaee, S., Boykov, Y., Porikli, F., Plaza, A., Kehtarnavaz, N., and Terzopoulos, D. Image segmentation using deep learning: A survey, 2020.

N, S. and S, V. Image segmentation by using thresholding techniques for medical images. Computer Science & Engineering: An International Journal vol. 6, pp. 1–13, 02, 2016.

Niri, R., Hassan, D., Yves, L., and Treuillet, S. Semantic segmentation of diabetic foot ulcer images: Dealing with small dataset in dl approaches, 2020.

Pare, S., Kumar, A., Singh, G. K., and Bajaj, V. Image segmentation using multilevel thresholding: A research review. Iranian Journal of Science and Technology, Transactions of Electrical Engineering 44 (1): 1–29, Mar, 2020.

Punn, N. S. and Agarwal, S. Modality specific u-net variants for biomedical image segmentation: a survey. Artificial Intelligence Review vol. 55, pp. 5845–5889, 2022.

Simonyan, K. and Zisserman, A. Very deep convolutional networks for large-scale image recognition. In International Conference on Learning Representations. CoRR, San Diego, CA, USA, 2015.

Van der Walt, S., Schönberger, J. L., Nunez-Iglesias, J., Boulogne, F., Warner, J. D., Yager, N., Gouillart, E., and Yu, T. scikit-image: image processing in python. PeerJ vol. 2, pp. e453, 2014.

Wang, C., Anisuzzaman, D., Williamson, V., Dhar, M. K., Rostami, B., Niezgoda, J., Gopalakrishnan, S., and Yu, Z. Fully automatic wound segmentation with deep convolutional neural networks, 2020.

Zhang, R., Tian, D., Xu, D., Qian, W., and Yao, Y. A survey of wound image analysis using deep learning: Classification, detection, and segmentation. IEEE Access vol. 10, pp. 79502–79515, 2022.