Um novo método baseado em detector de dois estágios para segmentação de instância de lesões retinianas usando o modelo Mask R-CNN e a biblioteca Detectron2
Resumo
O desenvolvimento de métodos computacionais para realizar a segmentação de instância de lesões retinianas como microaneurismas, hemorragias, exsudatos algodonosos e exsudatos duros desempenha um papel fundamental no diagnóstico precoce da Retinopatia Diabética. Este artigo propõe um método baseado no modelo Mask R-CNN para realizar a segmentação de instância de lesões associadas à Retinopatia Diabética. A arquitetura de rede neural profunda foi treinada, ajustada e testada usando diferentes conjuntos de dados de retinopatia diabética disponíveis publicamente. Além disso, a rede neural foi implementada usando as bibliotecas Detectron2 e OpenCV. O método proposto alcançou no DDR (Dataset for Diabetic Retinopathy) e com otimizador SGD (Stochastic Gradient Descent) o mAP (mean Averegare Precision) de 0,2664 para o limite de Interseção sobre União de 0,5 na etapa de validação.Referências
Alyoubi, W. L., Abulkhair, M. F., and Shalash, W. M. (2021). Diabetic retinopathy fundus image classification and lesions localization system using deep learning. Sensors, 21(11).
Amerikanos, P. and Maglogiannis, I. (2022). Image analysis in digital pathology utilizing machine learning and deep neural networks. Journal of Personalized Medicine, 12(9).
Dai, L., Wu, L., Li, H., Cai, C., Wu, Q., Kong, H., Liu, R., Wang, X., Hou, X., Liu, Y., Long, X., Wen, Y., Lu, L., Shen, Y., Chen, Y., Shen, D., Yang, X., Zou, H., Sheng, B., and Jia, W. (2021). A deep learning system for detecting diabetic retinopathy across the disease spectrum. Nature Communications, 12(1).
El abbadi, N. and Hammod, E. (2014). Automatic early diagnosis of diabetic retinopathy using retina fundus images enas hamood al-saadi-automatic early diagnosis of diabetic retinopathy using retina fundus images. EUROPEAN ACADEMIC RESEARCH, 2.
Goodfellow, I., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, A., and Bengio, Y. (2020). Generative adversarial networks. Communications of the ACM, 63(11):139–144.
He, K., Gkioxari, G., Dollár, P., and Girshick, R. (2020). Mask R-CNN. IEEE Transactions on Pattern Analysis and Machine Intelligence, 42(2):386–397.
Hong, A., Lee, G., Lee, H., Seo, J., and Yeo, D. (2021). Deep learning model generalization with ensemble in endoscopic images. CEUR Workshop Proceedings, 2886:80–89.
Li, T., Gao, Y., Wang, K., Guo, S., Liu, H., and Kang, H. (2019). Diagnostic assessment of deep learning algorithms for diabetic retinopathy screening. Information Sciences, 501:511–522.
Lin, T. Y., Maire, M., Belongie, S., Hays, J., Perona, P., Ramanan, D., Dollár, P., and Zitnick, C. L. (2014). Microsoft COCO: Common objects in context. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 8693 LNCS(PART 5):740–755.
Porwal, P., Pachade, S., Kokare, M., Deshmukh, G., Son, J., Bae, W., Liu, L., Wang, J., Liu, X., Gao, L., Wu, T. B., Xiao, J., Wang, F., Yin, B., Wang, Y., Danala, G., He, L., Choi, Y. H., Lee, Y. C., Jung, S. H., Li, Z., Sui, X., Wu, J., Li, X., Zhou, T., Toth, J., Baran, A., Kori, A., Chennamsetty, S. S., Safwan, M., Alex, V., Lyu, X., Cheng, L., Chu, Q., Li, P., Ji, X., Zhang, S., Shen, Y., Dai, L., Saha, O., Sathish, R., Melo, T., Araújo, T., Harangi, B., Sheng, B., Fang, R., Sheet, D., Hajdu, A., Zheng, Y., Mendonça, A. M., Zhang, S., Campilho, A., Zheng, B., Shen, D., Giancardo, L., Quellec, G., and Mériaudeau, F. (2020). IDRiD: Diabetic Retinopathy – Segmentation and Grading Challenge. Medical Image Analysis, 59.
Ren, S., He, K., Girshick, R., and Sun, J. (2017). Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks. IEEE Transactions on Pattern Analysis and Machine Intelligence, 39(6):1137–1149.
Ronneberger, O., Fischer, P., and Brox, T. (2015). U-net: Convolutional networks for biomedical image segmentation. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 9351:234–241.
Santos, C., Aguiar, M., Welfer, D., Dias, M., Pereira, A., Ribeiro, M., and Belloni, B. (2023). A new approach for fundus lesions instance segmentation based on mask r-cnn x101-fpn pre-trained architecture. IEEE Access, 11:43603–43618.
Shenavarmasouleh, F., Mohammadi, F. G., Amini, M. H., Taha, T., Rasheed, K., and Arabnia, H. R. (2021). Drdrv3: Complete lesion detection in fundus images using mask r-cnn, transfer learning, and lstm.
Unel, F. O., Ozkalayci, B. O., and Cigla, C. (2019). The power of tiling for small object detection. In 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), pages 582–591.
Vocaturo, E. and Zumpano, E. (2020). The contribution of AI in the detection of the Diabetic Retinopathy. Proceedings 2020 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2020, pages 1516–1519.
Wu, Y., Kirillov, A., Massa, F., Lo, W.-Y., and Girshick, R. (2019). Detectron2. https://github.com/facebookresearch/detectron2.
Yuen, H., Princen, J., Illingworth, J., and Kittler, J. (1990). Comparative study of hough transform methods for circle finding. Image and Vision Computing, 8(1):71–77.
Zhao, Z.-Q., Zheng, P., tao Xu, S., and Wu, X. (2019). Object detection with deep learning: A review.
Amerikanos, P. and Maglogiannis, I. (2022). Image analysis in digital pathology utilizing machine learning and deep neural networks. Journal of Personalized Medicine, 12(9).
Dai, L., Wu, L., Li, H., Cai, C., Wu, Q., Kong, H., Liu, R., Wang, X., Hou, X., Liu, Y., Long, X., Wen, Y., Lu, L., Shen, Y., Chen, Y., Shen, D., Yang, X., Zou, H., Sheng, B., and Jia, W. (2021). A deep learning system for detecting diabetic retinopathy across the disease spectrum. Nature Communications, 12(1).
El abbadi, N. and Hammod, E. (2014). Automatic early diagnosis of diabetic retinopathy using retina fundus images enas hamood al-saadi-automatic early diagnosis of diabetic retinopathy using retina fundus images. EUROPEAN ACADEMIC RESEARCH, 2.
Goodfellow, I., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, A., and Bengio, Y. (2020). Generative adversarial networks. Communications of the ACM, 63(11):139–144.
He, K., Gkioxari, G., Dollár, P., and Girshick, R. (2020). Mask R-CNN. IEEE Transactions on Pattern Analysis and Machine Intelligence, 42(2):386–397.
Hong, A., Lee, G., Lee, H., Seo, J., and Yeo, D. (2021). Deep learning model generalization with ensemble in endoscopic images. CEUR Workshop Proceedings, 2886:80–89.
Li, T., Gao, Y., Wang, K., Guo, S., Liu, H., and Kang, H. (2019). Diagnostic assessment of deep learning algorithms for diabetic retinopathy screening. Information Sciences, 501:511–522.
Lin, T. Y., Maire, M., Belongie, S., Hays, J., Perona, P., Ramanan, D., Dollár, P., and Zitnick, C. L. (2014). Microsoft COCO: Common objects in context. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 8693 LNCS(PART 5):740–755.
Porwal, P., Pachade, S., Kokare, M., Deshmukh, G., Son, J., Bae, W., Liu, L., Wang, J., Liu, X., Gao, L., Wu, T. B., Xiao, J., Wang, F., Yin, B., Wang, Y., Danala, G., He, L., Choi, Y. H., Lee, Y. C., Jung, S. H., Li, Z., Sui, X., Wu, J., Li, X., Zhou, T., Toth, J., Baran, A., Kori, A., Chennamsetty, S. S., Safwan, M., Alex, V., Lyu, X., Cheng, L., Chu, Q., Li, P., Ji, X., Zhang, S., Shen, Y., Dai, L., Saha, O., Sathish, R., Melo, T., Araújo, T., Harangi, B., Sheng, B., Fang, R., Sheet, D., Hajdu, A., Zheng, Y., Mendonça, A. M., Zhang, S., Campilho, A., Zheng, B., Shen, D., Giancardo, L., Quellec, G., and Mériaudeau, F. (2020). IDRiD: Diabetic Retinopathy – Segmentation and Grading Challenge. Medical Image Analysis, 59.
Ren, S., He, K., Girshick, R., and Sun, J. (2017). Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks. IEEE Transactions on Pattern Analysis and Machine Intelligence, 39(6):1137–1149.
Ronneberger, O., Fischer, P., and Brox, T. (2015). U-net: Convolutional networks for biomedical image segmentation. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 9351:234–241.
Santos, C., Aguiar, M., Welfer, D., Dias, M., Pereira, A., Ribeiro, M., and Belloni, B. (2023). A new approach for fundus lesions instance segmentation based on mask r-cnn x101-fpn pre-trained architecture. IEEE Access, 11:43603–43618.
Shenavarmasouleh, F., Mohammadi, F. G., Amini, M. H., Taha, T., Rasheed, K., and Arabnia, H. R. (2021). Drdrv3: Complete lesion detection in fundus images using mask r-cnn, transfer learning, and lstm.
Unel, F. O., Ozkalayci, B. O., and Cigla, C. (2019). The power of tiling for small object detection. In 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), pages 582–591.
Vocaturo, E. and Zumpano, E. (2020). The contribution of AI in the detection of the Diabetic Retinopathy. Proceedings 2020 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2020, pages 1516–1519.
Wu, Y., Kirillov, A., Massa, F., Lo, W.-Y., and Girshick, R. (2019). Detectron2. https://github.com/facebookresearch/detectron2.
Yuen, H., Princen, J., Illingworth, J., and Kittler, J. (1990). Comparative study of hough transform methods for circle finding. Image and Vision Computing, 8(1):71–77.
Zhao, Z.-Q., Zheng, P., tao Xu, S., and Wu, X. (2019). Object detection with deep learning: A review.
Publicado
06/08/2023
Como Citar
DIAS, Marcelo; SANTOS, Carlos; AGUIAR, Marilton; WELFER, Daniel; PEREIRA, Alejandro; RIBEIRO, Marcelo.
Um novo método baseado em detector de dois estágios para segmentação de instância de lesões retinianas usando o modelo Mask R-CNN e a biblioteca Detectron2. In: SEMINÁRIO INTEGRADO DE SOFTWARE E HARDWARE (SEMISH), 50. , 2023, João Pessoa/PB.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2023
.
p. 1-12.
ISSN 2595-6205.
DOI: https://doi.org/10.5753/semish.2023.229179.
