automated-joint-space-detection-improves-bone-segmentation-accuracy

Automated Joint Space Detection Improves Bone Segmentation Accuracy

Quantitative description of complex anatomical structures remains challenging due to the expertise necessary for manual segmentation, labor, and ...

University of Rochester Medical Center

Micro-computed tomography (μCT) is a valuable imaging modality for longitudinal quantification of bone volumes to identify disease or treatment effects for a broad range of conditions that affect bone health. Complex structures, such as the hindpaw with up to 31 distinct bones in mice, have considerable analytic potential, but quantification is often limited to a single bone volume metric due to the intensive effort of manual segmentation. Herein, we introduce a high-throughput, user-friendly, and semi-automated method for segmentation of murine hindpaw μCT datasets.

A high-throughput semi-automated bone segmentation workflow for murine hindpaw micro-CT datasets.

Although treatment options and algorithms for rheumatoid arthritis (RA) have improved remarkably in recent decades, there continues to be no definitive cure or pharmacologic intervention with reliable long-term efficacy. For this reason, the combination of medications and healthy lifestyle modifications are essential for controlling joint disease, and extra-articular manifestations of RA, such as interstitial lung disease (ILD) and other lung pathologies, which greatly impact morbidity and mortality. Generally, exercise has been deemed beneficial in RA patients, and both patients and clinicians are motivated to incorporate effective non-pharmacologic interventions. However, there are limited evidence-based and specific exercise regimens available to support engagement in such activities for RA patients. Here, we provided the continuous opportunity for exercise to mice and implemented automated recording and quantification of wheel running behavior. This allowed us to describe the associated effects on the progression of inflammatory-erosive arthritis and ILD in the tumor necrosis factor transgenic (TNF-Tg) mouse model of RA.

Implementation of automated behavior metrics to evaluate voluntary wheel running effects on inflammatory-erosive arthritis and interstitial lung disease in TNF-Tg mice.

Development of reliable disease activity biomarkers is critical for diagnostics, prognostics, and novel drug development. Although computed tomography (CT) is the gold-standard for quantification of bone erosions, there are no consensus approaches or rationales for utilization of specific outcome measures of erosive arthritis in complex joints. In the case of preclinical models, such as sexually dimorphic tumor necrosis factor transgenic (TNF-Tg) mice, disease severity is routinely quantified in the ankle through manual segmentation of the talus or small regions of adjacent bones primarily due to the ease in measurement. Herein, we sought to determine the particular hindpaw bones that represent reliable biomarkers of sex-dependent disease progression to guide future investigation and analysis.

High-throughput micro-CT analysis identifies sex-dependent biomarkers of erosive arthritis in TNF-Tg mice and differential response to anti-TNF therapy.

Quantitative description of complex anatomical structures remains challenging due to the expertise necessary for manual segmentation, labor, and interobserver variability. To overcome this, automated detection of specific landmarks can be provided by digital image analysis techniques including deep learning (DL) models. To this end, we undertook supervised automated analysis of micro-computed tomography (micro-CT) datasets of murine hindpaws and forepaws (30-33 bones). Advancing beyond previously published semi-automated (SA) marker-based watershed algorithms, we added structure enhancement, tensor voting, and output dilation to identify joint spaces. Segmentation was enhanced by the use of a DL joint space prediction model (3D U-Net architecture, ResNet-18 backbone) using wild-type (WT) hindpaw labels as ground truth. Prediction was then extended to hindpaws (52.4% in test group) and forepaws from WT and tumor necrosis factor transgenic (TNF-Tg) mice with inflammatory-erosive arthritis of both sexes across age. Segmentation accuracy improved dramatically using the DL methodology (WT male: SA 79.39±5.73% vs DL 98.16±1.47%, ; WT female: SA 79.16±4.84% vs DL 99.19±1.63%, ). Accuracy declined with increased disease severity and age in TNF-Tg mice (TNF-Tg male 93.54±4.73%, female 91.81±2.80%, ≤). Subsequent testing in forepaws also displayed progressive reduction in accuracy with increasing arthritic severity (i.e., WT 87.29±2.07%, TNF-Tg male 72.65±11.70%, ). Overall, this supervised automated model outperforms recent SA approaches in healthy joints to enhance investigation of complex bone anatomy. Although flexible application to novel and disease-modified datasets demonstrates deprecated performance, utilization may nonetheless catalyze structure-specific segmentation model development.

Article	Year
A high-throughput semi-automated bone segmentation workflow for murine hindpaw micro-CT datasets. Bone reports\| PubMed ID: 35146075	2022
Implementation of automated behavior metrics to evaluate voluntary wheel running effects on inflammatory-erosive arthritis and interstitial lung disease in TNF-Tg mice. Arthritis research & therapy\| PubMed ID: 36732826	2023
High-throughput micro-CT analysis identifies sex-dependent biomarkers of erosive arthritis in TNF-Tg mice and differential response to anti-TNF therapy. PloS one\| PubMed ID: 38968295	2024
Automated Joint Space Detection Improves Bone Segmentation Accuracy. bioRxiv : the preprint server for biology\| PubMed ID: 40654621	2025

Lindsay Schnur

Biography

JoVE Journal Publications

Other Publications