Exploring Deep Neural Network Architectures in Human Karyotype Image Classification

• Francisco das C. Imperes Filho UFPI
• Vinicius P. Machado UFPI
• Semiramis J. H. do Monte UFPI
• Alan R. F. dos Santos UFPI
• Leonardo P. de Sousa UFPI
• Adalberto S. da Silva UFPI
• Ester M. Pereira UFPI
• Rodrigo de M. S. Veras UFPI

DOI: https://doi.org/10.5753/sbcas.2024.2798

Abstract

Chromosome analysis, a clinically crucial practice traditionally performed by geneticists, can be susceptible to fatigue over time, affecting the quality of diagnoses. This paper explored the automated classification of chromosome images using various deep-learning architectures. We evaluated 23 pairs of human chromosomes in a multiclass task, revealing promising results. The superior performances of the DenseNet169 architecture are highlighted, achieving an accuracy, precision, recall, and F1-Score of 98.77%. The Kappa agreement index reached an ”Excellent”level (0.99), while a low standard deviation (0.002) highlighted the consistency of the metrics, providing reliability and predictability to the proposed model.

References

Abid, F. and Hamami, L. (2018). A survey of neural network based automated systems for human chromosome classification. Artificial Intelligence Review, 49:41–56.

Anh, L., Thanh, V., Son, N., Phuong, D. K., Anh, L., Ram, T., Minh, B., Tung, H., Thinh, N., Ha, V., and Ha, M. (2022). Efficient type and polarity classification of chromosome images using cnns: a primary evaluation on multiple datasets. In 2022 IEEE Ninth International Conference on Communications and Electronics (ICCE), pages 400–405.

Chollet, F. (2017). Xception: Deep learning with depthwise separable convolutions.

He, K., Zhang, X., Ren, S., and Sun, J. (2016). Deep residual learning for imagere cognition. In 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), IEEE Computer Society:770–778.

Hsu, C.-Y. and Li, W. (2023). Explainable geoai: can saliency maps help interpret artificial intelligence’s learning process? an empirical study on natural feature detection. 37(5):963–987.

Huang, G., Liu, Z., and Van Der Maaten, L. and0 andWeinberger, K. Q. (2017). Densely connected convolutional networks. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, page 4700–4708.

Izadyyazdanabadi, M., Belykh, E., Mooney, M., Martirosyan, N., Eschbacher, J., Nakaji, P., Preul, M. C., and Yang, Y. (2018). Convolutional neural networks: ensemble modeling, fine-tuning and unsupervised semantic localization for neurosurgical cle images. 54.

Kornblith, S., Shlens, J., and Le, Q. V. (2019). Do better imagenet models transfer better? Computer Vision and Pattern Recognition (cs.CV), 1:2661–2671.

Lin, C., Chen, H., Huang, J., Peng, J., Guo, L., Yang, Z., Du, J., Li, S., Yin, A., and Zhao, G. (2022). Chromosomenet: A massive dataset enabling benchmarking and building basedlines of clinical chromosome classification. Computational Biology and Chemistry, 100:107731.

Moon, W. K., Lee, Y.-W., Ke, H.-H., Lee, S. H., Huang, C.-S., and Chang, R.-F. (2020). Computer-aided diagnosis of breast ultrasound images using ensemble learning from convolutional neural networks. Computer Methods and Programs in Biomedicine, 190:105361.

Perez, L. and Wang, J. (2017). The effectiveness of data augmentation in image classification using deep learning. Computer Vision and Pattern Recognition (cs.CV).

Russakovsky, O., Deng, J., Su, H., Krause, J., Satheesh, S., Ma, S., Huang, Z., Karpathy, A., Khosla, A., and Bernstein, M. (2015). Imagenet large scale visual recognition challenge. International Journal of Computer Vision, 115(3), page 211–252.

Santos, F., Veras, R., Santos, E., Claro, M., Vogado, L. H., Ito, M., and Bianchi, A. (2021). Uma avaliação de arquiteturas de aprendizado profundo para a classificação de Úlceras do pé diabético. In Anais do XXI Simpósio Brasileiro de Computação Aplicada à Saúde, pages 323–334, Porto Alegre, RS, Brasil. SBC.

Saranya, S. and Lakshmi, S. (2023). Classification of chromosomes to diagnose chromosomal abnormalities using cnn. In 2023 International Conference on Artificial Intelligence and Knowledge Discovery in Concurrent Engineering (ICECONF), pages 1–5.

Simonyan, K. and Zisserman, A. (2014). Very deep convolutional networks for large-scale image recognition. Computer Vision and Pattern Recognition (cs.CV), 1.

Sokolova, M. and Lapalme, G. (2009). A systematic analysis of performance measures for classification tasks. Information Processing Management, 45(4):427–437.

Szegedy, C., Vanhoucke, V., Ioffe, S., Shlens, J., and Wojna, Z. (2016). Rethinking the inception architecture for computer vision. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, page 2818–2826.

Tan, M. and Le, Q. V. (2021). Efficientnetv2: Smaller models and faster training. Computer Vision and Pattern Recognition (cs.CV) (ICML 2021), 1.

Vogado, L., Veras, R., Aires, K., A. F., Silva, R., Ponti, M., and Tavares, J. M. R. (2021). Diagnosis of leukaemia in blood slides based on a fine-tuned and highly generalisable deep learning model. Sensors, 21(9), page 2989–2989.

Xia, C., Wang, J., Qin, Y., Wen, J., Liu, Z., Song, N., Wu, L., Chen, B., Gu, Y., and Yang, J. (2023). Karyonet: Chromosome recognition with end-to-end combinatorial optimization network. IEEE Transactions on Medical Imaging, 42(10):2899–2911.