Image Segmentation to Distinguish Between Overlapping Human Chromosomes

TL;DR

This paper presents a neural network-based image segmentation method that accurately distinguishes overlapping human chromosomes, facilitating medical diagnostics and biomedical research with high precision.

Contribution

It introduces a customized convolutional neural network for segmenting overlapping chromosomes, achieving high IOU scores and enabling automated, scalable chromosome analysis.

Findings

IOU score of 94.7% for overlapping regions

IOU scores of 88-94% for non-overlapping regions

Effective segmentation of overlapping chromosomes

Abstract

In medicine, visualizing chromosomes is important for medical diagnostics, drug development, and biomedical research. Unfortunately, chromosomes often overlap and it is necessary to identify and distinguish between the overlapping chromosomes. A segmentation solution that is fast and automated will enable scaling of cost effective medicine and biomedical research. We apply neural network-based image segmentation to the problem of distinguishing between partially overlapping DNA chromosomes. A convolutional neural network is customized for this problem. The results achieved intersection over union (IOU) scores of 94.7% for the overlapping region and 88-94% on the non-overlapping chromosome regions.