Automatic Fine-grained Glomerular Lesion Recognition in Kidney Pathology
This work addresses the challenge of diagnosing kidney diseases through automated lesion recognition, which is incremental as it builds on existing detection methods with specific enhancements.
The paper tackles the problem of fine-grained glomerular lesion recognition in kidney pathology by introducing a scheme that uses a focal instance structural similarity loss and an Uncertainty Aided Apportionment Network, achieving an 8-22% improvement in mean Average Precision compared to existing detection methods.
Recognition of glomeruli lesions is the key for diagnosis and treatment planning in kidney pathology; however, the coexisting glomerular structures such as mesangial regions exacerbate the difficulties of this task. In this paper, we introduce a scheme to recognize fine-grained glomeruli lesions from whole slide images. First, a focal instance structural similarity loss is proposed to drive the model to locate all types of glomeruli precisely. Then an Uncertainty Aided Apportionment Network is designed to carry out the fine-grained visual classification without bounding-box annotations. This double branch-shaped structure extracts common features of the child class from the parent class and produces the uncertainty factor for reconstituting the training dataset. Results of slide-wise evaluation illustrate the effectiveness of the entire scheme, with an 8-22% improvement of the mean Average Precision compared with remarkable detection methods. The comprehensive results clearly demonstrate the effectiveness of the proposed method.