Effort-free Automated Skeletal Abnormality Detection of Rat Fetuses on Whole-body Micro-CT Scans

TL;DR

This paper introduces an automated method for detecting skeletal abnormalities in rat fetuses using micro-CT scans, employing bone feature engineering to minimize manual effort and achieve high accuracy despite limited data.

Contribution

It presents a novel bone feature engineering approach for automated skeletal analysis in fetal micro-CT scans with minimal data annotation.

Findings

Achieved 90% accuracy in skeletal labeling.

Achieved 81% accuracy in abnormality detection.

Demonstrated effectiveness with only 49 training samples.

Abstract

Machine Learning-based fast and quantitative automated screening plays a key role in analyzing human bones on Computed Tomography (CT) scans. However, despite the requirement in drug safety assessment, such research is rare on animal fetus micro-CT scans due to its laborious data collection and annotation. Therefore, we propose various bone feature engineering techniques to thoroughly automate the skeletal localization/labeling/abnormality detection of rat fetuses on whole-body micro-CT scans with minimum effort. Despite limited training data of 49 fetuses, in skeletal labeling and abnormality detection, we achieve accuracy of 0.900 and 0.810, respectively.