Longitudinal Variability Analysis on Low-dose Abdominal CT with Deep Learning-based Segmentation

TL;DR

This study evaluates the stability and variability of low-dose abdominal CT measurements over time using deep learning segmentation, revealing organ-specific variability and the influence of slice position on measurements.

Contribution

It introduces a comprehensive analysis of longitudinal variability in low-dose single slice CT using automated segmentation and clustering, with a focus on stability and sources of measurement variability.

Findings

Segmentation methods are stable with Dice scores from 0.821 to 0.962.

High variability observed in most organs with ICC<0.5.

Variability is highly related to slice position.

Abstract

Metabolic health is increasingly implicated as a risk factor across conditions from cardiology to neurology, and efficiency assessment of body composition is critical to quantitatively characterizing these relationships. 2D low dose single slice computed tomography (CT) provides a high resolution, quantitative tissue map, albeit with a limited field of view. Although numerous potential analyses have been proposed in quantifying image context, there has been no comprehensive study for low-dose single slice CT longitudinal variability with automated segmentation. We studied a total of 1816 slices from 1469 subjects of Baltimore Longitudinal Study on Aging (BLSA) abdominal dataset using supervised deep learning-based segmentation and unsupervised clustering method. 300 out of 1469 subjects that have two year gap in their first two scans were pick out to evaluate longitudinal variability with measurements including intraclass correlation coefficient (ICC) and coefficient of variation (CV) in terms of tissues/organs size and mean intensity. We showed that our segmentation methods are stable in longitudinal settings with Dice ranged from 0.821 to 0.962 for thirteen target abdominal tissues structures. We observed high variability in most organ with ICC<0.5, low variability in the area of muscle, abdominal wall, fat and body mask with average ICC>0.8. We found that the variability in organ is highly related to the cross-sectional position of the 2D slice. Our efforts pave quantitative exploration and quality control to reduce uncertainties in longitudinal analysis.