Efficient type and polarity classification of chromosome images using CNNs: a primary evaluation on multiple datasets

This paper develops a CNN architecture for chromosome type and polarity classification. We utilize EfficientNet as the backbone for feature extraction and introduce a weighted classification loss function. The proposed network achieves comparable results with a significantly smaller number of parameters than state-of-the-art methods. The method is embedded in Biochrom software — a karyotyping analysis tool.

Recommended citation: Le Quoc Anh, Vu Duy Thanh, Nguyen Huu Hoang Son, Doan Thi Kim Phuong, Luong Thi Lan Anh, Nguyen Thanh Binh Minh, Tran Hoang Tung, Nguyen Hong Thinh, Luu Manh Ha, et al. "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. IEEE, 2022.
Download Paper