Detection of Virus and Small Cell Patches in Foci Images Using Switchable Convolution and Feature Pyramid Networks

TL;DR

This paper introduces an enhanced YOLOv2-based model with feature pyramid networks and switchable atrous convolution for improved detection of virus and cell patches in microscopy images, addressing size and density variability.

Contribution

It presents a novel combination of FPN and switchable convolution within YOLOv2 to better detect biomedical targets of varying scales and densities.

Findings

Achieved 40.5% mAP for small cell patches at 25% IoU.

Achieved 68% mAP for virus patches.

Improved multi-scale detection performance in biomedical images.

Abstract

Accurate detection and counting of virus patches in focus-forming unit (FFU) images, also known as foci images, are important for quantifying viral infection and analyzing cellular structures. This task is challenging because biomedical targets often vary substantially in size, density, contrast, and shape. In this paper, we propose an enhanced YOLOv2-based detector that integrates a Feature Pyramid Network (FPN) to improve multi-scale feature representation. We also incorporate a switchable atrous convolution mechanism to adapt the receptive field for fine-grained targets in dense microscopy images. The proposed method is evaluated on biomedical foci image datasets for virus patch and small cell patch detection. For small cell patch detection, the model achieves a mean average precision (mAP) of 40.5% at a 25% Intersection over Union (IoU) threshold. For FFU virus patch detection, the model achieves an mAP of 68%. These results indicate that combining FPN-based feature fusion with switchable convolution improves the suitability of YOLOv2 for specialized biomedical object detection tasks