CNN Filter Learning from Drawn Markers for the Detection of Suggestive Signs of COVID-19 in CT Images

TL;DR

This paper introduces a novel CNN filter learning method for COVID-19 detection in CT images that does not require large annotated datasets or backpropagation, using user-drawn markers and a support vector machine classifier.

Contribution

It presents a new approach to train CNN filters from minimal data using manual markers, bypassing traditional backpropagation, and includes an intensity standardization step for diverse CT images.

Findings

Achieved 97% accuracy and 93% kappa on a dataset of 117 CT images.

Outperformed traditional CNN training methods in all tested scenarios.

Effective with limited annotated data and diverse imaging sources.

Abstract

Early detection of COVID-19 is vital to control its spread. Deep learning methods have been presented to detect suggestive signs of COVID-19 from chest CT images. However, due to the novelty of the disease, annotated volumetric data are scarce. Here we propose a method that does not require either large annotated datasets or backpropagation to estimate the filters of a convolutional neural network (CNN). For a few CT images, the user draws markers at representative normal and abnormal regions. The method generates a feature extractor composed of a sequence of convolutional layers, whose kernels are specialized in enhancing regions similar to the marked ones, and the decision layer of our CNN is a support vector machine. As we have no control over the CT image acquisition, we also propose an intensity standardization approach. Our method can achieve mean accuracy and kappa values of $0.97$ and $0.93$, respectively, on a dataset with 117 CT images extracted from different sites, surpassing its counterpart in all scenarios.