The early diagnosis of cancer based on histopathology images plays an important role in medical science. Existing techniques generally partition the original histopathology image into small pieces for further classification. However, due to the fact that the number of benign (majority) samples is much larger than that of malignant (minority) samples, the classification is significantly imbalanced which adversely affects classification performance. Undersampling is commonly used to address the class-imbalance problem. However, existing methods are typically time consuming so they are not suitable to handle large-scale and high-dimensional data. In this paper we propose a fast and scalable undersampling method, hashing-based undersampling (HBU), to address class imbalance in large-scale medical image classification. Benign images are hashed and then placed into different buckets according to their locations in the input space. Undersampling is achieved by proportionally selecting benign images from the hash buckets. The HBU method is experimentally evaluated on two real histopathology image datasets, CAMELYON16 and ACDC@LUNGHP, by comparison with existing methods. Experimental results show that the HBU method outperforms six state-of-The-Art methods and is scalable and fast.
|Title of host publication||Proceedings of the 2022 IEEE 21st International Conference on Cognitive Informatics and Cognitive Computing|
|Publication status||Published - Dec 2022|
|Event||2022 IEEE 21st International Conference on Cognitive Informatics & Cognitive Computing - , Canada|
Duration: 8 Dec 2022 → 10 Dec 2022
|Conference||2022 IEEE 21st International Conference on Cognitive Informatics & Cognitive Computing|
|Period||8/12/22 → 10/12/22|
Bibliographical noteThis work was supported in part by the National Natural Science Foundation of China under Grants 62202175, 61876066, 61772344, and 61672443, the Science and Technology Planning Project of Guangzhou (SL2023A04J01464), the 67th Chinese Postdoctoral Science Foundation (2020M672631), the Key-Area Research and Development of Guangdong Province under Grant 2020B010166002, EU Horizon 2020 Programme (700381, ASGARD), the Hong Kong RGC General Research Funds under Grant 9042489 (CityU 11206317), Grant 9042816 (CityU 11209819) and Grant 9042322 (CityU 11200116), and the Hong Kong Innovation and Technology Commission (InnoHK Project CIMDA).
- Cancer diagnosis
- Histopathology image