Automatic Segmentation of Fluorescence Lifetime Microscopy Images of Cells Using Multi-Resolution Community Detection

TL;DR

This paper introduces an automatic segmentation method for fluorescence lifetime microscopy images of cells using multi-resolution community detection, outperforming spectral clustering by providing more accurate and consistent segmentation results.

Contribution

The paper presents a novel multi-resolution community detection approach for FLIM image segmentation, demonstrating improved accuracy and consistency over existing spectral clustering methods.

Findings

MCD method outperforms spectral clustering in FLIM segmentation

MSE decreases with increasing network resolution in MCD

Proposed method produces less noisy segments at high resolution

Abstract

We have developed an automatic method for segmenting fluorescence lifetime (FLT) imaging microscopy (FLIM) images of cells inspired by a multi-resolution community detection (MCD) based network segmentation method. The image processing problem is framed as identifying segments with respective average FLTs against a background in FLIM images. The proposed method segments a FLIM image for a given resolution of the network composed using image pixels as the nodes and similarity between the pixels as the edges. In the resulting segmentation, low network resolution leads to larger segments and high network resolution leads to smaller segments. Further, the mean-square error (MSE) in estimating the FLT segments in a FLIM image using the proposed method was found to be consistently decreasing with increasing resolution of the corresponding network. The proposed MCD method outperformed a popular spectral clustering based method in performing FLIM image segmentation. The spectral segmentation method introduced noisy segments in its output at high resolution. It was unable to offer a consistent decrease in MSE with increasing resolution.