Intelligent Objective Osteon Segmentation Based on Deep Learning

Zichuan Qin,Congyu Yu,Fangbo Qin,Ying Li

doi:10.3389/feart.2022.783481

Abstract

Histology is key to understand physiology, development, growth and even reproduction of extinct animals. However, the identification and interpretation of certain structures, such as osteons, medullary bone (MB), and Lines of Arrested Growth (LAGs), are not only based on personal judgments, but also require considerable labor for subsequent analysis. Due to the dearth of available specimens, only a few quantitative histological studies have been proceeded for limited dinosaur taxa, most of which focus primarily on their growth, namely, LAGs and other growth lines without much attention to other histological structures. Here we develop a deep convolutional neural network-based method for automated osteohistological segmentation. Raw images are firstly divided into sub-images and the borders are expanded to guarantee the osteon regions integrity. ResNet-50 is employed as feature extractor and atrous spatial pyramid pooling (ASPP) is used to capture multi-scale information. A dual-resolution segmentation strategy is designed to observe the primary and secondary osteon regions from the matrix background. Finally, a segmented map with different osteon regions is obtained. This deep convolutional neural network-based model is tested on a histological dataset derived from various taxa in Alvarezsauria, a highly specialized group of non-avian theropod dinosaurs. The results show that large-scale quantitative histological analysis can be achieved by neural network-based methods, and previously hidden information by traditional methods can be revealed. Phylogenetic mapping of osteon segmentation results suggests a developmental pathway towards miniaturized body sizes in the evolution of Alvarezsauria, which may resemble the transition from non-avian dinosaurs to birds.

Highlights

Histology provides a unique chance to study the physiology, development, growth and even reproduction of extinct animals including non-avian dinosaurs
The earliest histological work on dinosaur fossils can be traced back to Hylaeosaurus and Pelorosaurus by Mantell (1850), Mantell (1851), who provided the first description of microstructures in dinosaur bones, which was revisited by a thorough review on the early history of paleontological histology (Falcon-Lang and Digrius, 2014)
Most of quantitative studies rely on manually collected statistics, for example counting the number of growth lines and calculating areas of osteons, etc

Summary

INTRODUCTION

Histology provides a unique chance to study the physiology, development, growth and even reproduction of extinct animals including non-avian dinosaurs. With the recent development of convolutional neural network (CNN) in image processing, CNN based feature learning methods have shown their great potential in biomedical image segmentation. The network was trained in an end-to-end manner that directly provides segmentation maps based on the raw images, skip connections in its U-shape framework can improve the performance of the expansive path, the U-Net has become a popular architecture for biomedical image segmentation tasks. Gu et al (2019) integrated dense atrous convolution (DAC) and residual multi-kernel pooling (RMP) blocks with encoder-decoder structure for medical image segmentation. Motivated by pursuing automatic objective analysis of osteon distribution in fossil histological images, we utilized the deep learning-based segmentation method to recognize the osteon regions of interest: vascular canal (VC) and circular lamellar bone (CLB). The model performance was tested on a dataset comprising bone thin sections derived from Alvarezsauria (Dinosauria: Theropoda)

MATERIALS AND METHODS

RESULTS

DISCUSSION

DATA AVAILABILITY STATEMENT