# Investigation into the utility of ultra-high-resolution mode on dynamic-ventilation computed tomography using sponge phantoms

## TL;DR

Ultra-high-resolution CT scans provide better detail of lung structures during simulated breathing compared to normal-resolution scans.

## Contribution

This study is the first to evaluate ultra-high-resolution CT in dynamic-ventilation CT using sponge phantoms.

## Key findings

- UHR-CT detected more simulated peripheral air spaces than NR-CT using binarization and homology-based analysis.
- UHR-CT showed smaller deviations in SPAS measurements compared to smartphone reference images.
- UHR DVCT accurately reflected changes in SPAS with compression, while NR DVCT did not.

## Abstract

Ultra-high-resolution (UHR) computed tomography (CT) provides more detailed information on peripheral lung structures than normal-resolution (NR) CT for static images. However, its potential advantage in four-dimensional dynamic-ventilation CT (DVCT) has not been investigated. This experimental study used sponge phantoms and a compression machine to simulate respiratory changes and assess the superiority of the UHR mode on DVCT. We obtained results that demonstrated: (1) regardless of the compression percentage, the number of simulated peripheral air spaces (SPAS) detected on UHR-CT was greater than on NR-CT, as assessed using a simple binarization method. Similarly, peak b0 values as a representative of total SPAS count obtained by homology-based image analysis were higher on UHR-CT than on NR-CT. (2) The deviation in measured longitudinal diameter and area of the SPAS relative to the reference standard—smartphone images—was smaller on UHR-CT than on NR-CT. (3) UHR DVCT demonstrated a gradual reduction in the longitudinal diameter and area of SPAS with increasing compression percentages, consistent with smartphone images, whereas NR DVCT did not. These findings suggested that UHR mode is superior to NR mode in depicting peripheral lung structures for DVCT as well as static images.

The online version contains supplementary material available at 10.1038/s41598-025-30102-5.