KerNet: A Novel Deep Learning Approach for Keratoconus and Sub-clinical Keratoconus Detection Based on Raw Data of the Pentacam System.

Keratoconus is one of the most severe corneal diseases, which is difficult to detect at the early stage (i.e., sub-clinical keratoconus) and possibly results in vision loss. In this paper, we propose a novel end-to-end deep learning approach, called KerNet, which processes the raw data of the Pentacam system to detect keratoconus and sub-clinical keratoconus. First, we collect raw data from the Pentacam system. The raw data is of a specific format, that is, each sample consists of five numerical matrices, corresponding to the front and back surface curvature, the front and back surface elevation, and the pachymetry of an eye. Then, we propose a novel convolutional neural network, called KerNet, containing five branches as the backbone with a multi-level fusion architecture. The five branches receive five slices separately and capture effectively the features of different slices by several cascaded residual blocks. The multi-level fusion architecture (i.e., low-level fusion and high-level fusion) moderately takes into account the correlation among five slices and fuses the extracted features for better prediction. Specifically, five spatial attention modules are utilized, each in a branch, to guide the operation of the low-level fusion. The high-level fusion is implemented by simply concatenating the output feature maps of the last residual block in each branch. Experimental results show that: (1) our novel approach outperforms state-of-the-art methods on an in-house dataset, by  ∼ 1\% for keratoconus detection accuracy and  ∼ 4\% for sub-clinical keratoconus detection accuracy; (2) the attention maps visualized by Grad-CAM show that our KerNet places more attention on the inferior temporal part for sub-clinical keratoconus, which has been proved as the identifying regions for ophthalmologists to detect sub-clinical keratoconus in previous clinical studies. To our best knowledge, we are the first to propose an end-to-end deep learning approach utilizing raw data obtained by the Pentacam system for keratoconus and subclinical keratoconus detection. Further, the prediction performance and the clinical significance of our KerNet are well evaluated and proved by two clinical experts. Our code is available at \url{https://github.com/upzheng/Keratoconus}.