Deconvolving convolution neural network for cell detection

TL;DR

This paper introduces a novel cell detection method using deconvolution of CNN outputs, which improves precision and F1-score over existing deep learning approaches in histology images.

Contribution

The paper proposes a deconvolution-based approach for cell detection, offering an alternative to local maxima detection in deep learning methods.

Findings

Higher precision in cell detection compared to state-of-the-art methods

Improved F1-score demonstrating better overall accuracy

Effective handling of varying cell sizes and stain variations

Abstract

Automatic cell detection in histology images is a challenging task due to varying size, shape and features of cells and stain variations across a large cohort. Conventional deep learning methods regress the probability of each pixel belonging to the centre of a cell followed by detection of local maxima. We present deconvolution as an alternate approach to local maxima detection. The ground truth points are convolved with a mapping filter to generate artifical labels. A convolutional neural network (CNN) is modified to convolve it's output with the same mapping filter and is trained for the mapped labels. Output of the trained CNN is then deconvolved to generate points as cell detection. We compare our method with state-of-the-art deep learning approaches where the results show that the proposed approach detects cells with comparatively high precision and F1-score.