# 2D and 3D Segmentation of uncertain local collagen fiber orientations in SHG microscopy

## TL;DR

This paper compares classical and deep learning methods for segmenting collagen fiber orientations in SHG microscopy, addressing uncertainty in borders and demonstrating a 3D neural network approach with human-level accuracy.

## Contribution

It introduces a novel method for 3D segmentation of collagen fibers using transferred 2D ImageNet weights in neural networks, achieving human-comparable accuracy.

## Key findings

- Deep neural networks outperform classical methods in fiber orientation classification.
- Transferring 2D weights to 3D networks is effective for segmentation.
- The approach achieves accuracy comparable to human performance.

## Abstract

Collagen fiber orientations in bones, visible with Second Harmonic Generation (SHG) microscopy, represent the inner structure and its alteration due to influences like cancer. While analyses of these orientations are valuable for medical research, it is not feasible to analyze the needed large amounts of local orientations manually. Since we have uncertain borders for these local orientations only rough regions can be segmented instead of a pixel-wise segmentation. We analyze the effect of these uncertain borders on human performance by a user study. Furthermore, we compare a variety of 2D and 3D methods such as classical approaches like Fourier analysis with state-of-the-art deep neural networks for the classification of local fiber orientations. We present a general way to use pretrained 2D weights in 3D neural networks, such as Inception-ResNet-3D a 3D extension of Inception-ResNet-v2. In a 10 fold cross-validation our two stage segmentation based on Inception-ResNet-3D and transferred 2D ImageNet weights achieves a human comparable accuracy.