Performance Evaluation of Image Segmentation Using Dual-Energy Spectral CT Images with Deep Learning Image Reconstruction: A Phantom Study

<b>Objectives</b>: To evaluate the medical image segmentation performance of monochromatic images in various energy levels. <b>Methods</b>: The low-density module (25 mm in diameter, 6 Hounsfield Unit (HU) in density difference from background) from the ACR464 phantom was sca...

Full description

Saved in:
Bibliographic Details
Main Authors: Haoyan Li, Zhenpeng Chen, Shuaiyi Gao, Jiaqi Hu, Zhihao Yang, Yun Peng, Jihang Sun
Format: Article
Language:English
Published: MDPI AG 2025-04-01
Series:Tomography
Subjects:
Online Access:https://www.mdpi.com/2379-139X/11/5/51
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850126430361354240
author Haoyan Li
Zhenpeng Chen
Shuaiyi Gao
Jiaqi Hu
Zhihao Yang
Yun Peng
Jihang Sun
author_facet Haoyan Li
Zhenpeng Chen
Shuaiyi Gao
Jiaqi Hu
Zhihao Yang
Yun Peng
Jihang Sun
author_sort Haoyan Li
collection DOAJ
description <b>Objectives</b>: To evaluate the medical image segmentation performance of monochromatic images in various energy levels. <b>Methods</b>: The low-density module (25 mm in diameter, 6 Hounsfield Unit (HU) in density difference from background) from the ACR464 phantom was scanned at both 10 mGy and 5 mGy dose levels. Virtual monoenergetic images (VMIs) at different energy levels of 40, 50, 60, 68, 74, and 100 keV were generated. The images at 10 mGy reconstructed with 50% adaptive statistical iterative reconstruction veo (ASIR-V50%) were used to train an image segmentation model based on U-Net. The evaluation set used 5 mGy VMIs reconstructed with various reconstruction algorithms: FBP, ASIR-V50%, ASIR-V100%, deep learning image reconstruction (DLIR) with low (DLIR-L), medium (DLIR-M), and high (DLIR-H) strength levels. U-Net was employed as a tool to compare algorithm performance. Image noise and segmentation metrics, such as the DICE coefficient, intersection over union (IOU), sensitivity, and Hausdorff distance, were calculated to assess both image quality and segmentation performance. <b>Results</b>: DLIR-M and DLIR-H consistently achieved lower image noise and better segmentation performance, with the highest results observed at 60 keV, and DLIR-H had the lowest image noise across all energy levels. The performance metrics, including IOU, DICE, and sensitivity, were ranked in descending order with energy levels of 60 keV, 68 keV, 50 keV, 74 keV, 40 keV, and 100 keV. Specifically, at 60 keV, the average IOU values for each reconstruction method were 0.60 for FBP, 0.67 for ASIR-V50%, 0.68 for ASIR-V100%, 0.72 for DLIR-L, 0.75 for DLIR-M, and 0.75 for DLIR-H. The average DICE values were 0.75, 0.80, 0.82, 0.83, 0.85, and 0.86. The sensitivity values were 0.93, 0.91, 0.96, 0.95, 0.98, and 0.98. <b>Conclusions</b>: For low-density, non-enhancing objects under a low dose, the 60 keV VMIs performed better in automatic segmentation. DLIR-M and DLIR-H algorithms delivered the best results, whereas DLIR-H provided the lowest image noise and highest sensitivity.
format Article
id doaj-art-e5615024fd9e4f508247d876753c8f60
institution OA Journals
issn 2379-1381
2379-139X
language English
publishDate 2025-04-01
publisher MDPI AG
record_format Article
series Tomography
spelling doaj-art-e5615024fd9e4f508247d876753c8f602025-08-20T02:33:55ZengMDPI AGTomography2379-13812379-139X2025-04-011155110.3390/tomography11050051Performance Evaluation of Image Segmentation Using Dual-Energy Spectral CT Images with Deep Learning Image Reconstruction: A Phantom StudyHaoyan Li0Zhenpeng Chen1Shuaiyi Gao2Jiaqi Hu3Zhihao Yang4Yun Peng5Jihang Sun6Department of Radiology, Beijing Children’s Hospital, Capital Medical University, National Center for Children’s Health, No. 56, Nanlishi Road, Xicheng District, Beijing 100045, ChinaQingdao Academy of Chinese Medical Sciences, Shandong University of Traditional Chinese Medicine, Qingdao 266112, ChinaDepartment of Radiology, Beijing Children’s Hospital, Capital Medical University, National Center for Children’s Health, No. 56, Nanlishi Road, Xicheng District, Beijing 100045, ChinaDepartment of Radiology, Beijing Children’s Hospital, Capital Medical University, National Center for Children’s Health, No. 56, Nanlishi Road, Xicheng District, Beijing 100045, ChinaDepartment of Radiology, Beijing Children’s Hospital, Capital Medical University, National Center for Children’s Health, No. 56, Nanlishi Road, Xicheng District, Beijing 100045, ChinaDepartment of Radiology, Beijing Children’s Hospital, Capital Medical University, National Center for Children’s Health, No. 56, Nanlishi Road, Xicheng District, Beijing 100045, ChinaDepartment of Radiology, Beijing Children’s Hospital, Capital Medical University, National Center for Children’s Health, No. 56, Nanlishi Road, Xicheng District, Beijing 100045, China<b>Objectives</b>: To evaluate the medical image segmentation performance of monochromatic images in various energy levels. <b>Methods</b>: The low-density module (25 mm in diameter, 6 Hounsfield Unit (HU) in density difference from background) from the ACR464 phantom was scanned at both 10 mGy and 5 mGy dose levels. Virtual monoenergetic images (VMIs) at different energy levels of 40, 50, 60, 68, 74, and 100 keV were generated. The images at 10 mGy reconstructed with 50% adaptive statistical iterative reconstruction veo (ASIR-V50%) were used to train an image segmentation model based on U-Net. The evaluation set used 5 mGy VMIs reconstructed with various reconstruction algorithms: FBP, ASIR-V50%, ASIR-V100%, deep learning image reconstruction (DLIR) with low (DLIR-L), medium (DLIR-M), and high (DLIR-H) strength levels. U-Net was employed as a tool to compare algorithm performance. Image noise and segmentation metrics, such as the DICE coefficient, intersection over union (IOU), sensitivity, and Hausdorff distance, were calculated to assess both image quality and segmentation performance. <b>Results</b>: DLIR-M and DLIR-H consistently achieved lower image noise and better segmentation performance, with the highest results observed at 60 keV, and DLIR-H had the lowest image noise across all energy levels. The performance metrics, including IOU, DICE, and sensitivity, were ranked in descending order with energy levels of 60 keV, 68 keV, 50 keV, 74 keV, 40 keV, and 100 keV. Specifically, at 60 keV, the average IOU values for each reconstruction method were 0.60 for FBP, 0.67 for ASIR-V50%, 0.68 for ASIR-V100%, 0.72 for DLIR-L, 0.75 for DLIR-M, and 0.75 for DLIR-H. The average DICE values were 0.75, 0.80, 0.82, 0.83, 0.85, and 0.86. The sensitivity values were 0.93, 0.91, 0.96, 0.95, 0.98, and 0.98. <b>Conclusions</b>: For low-density, non-enhancing objects under a low dose, the 60 keV VMIs performed better in automatic segmentation. DLIR-M and DLIR-H algorithms delivered the best results, whereas DLIR-H provided the lowest image noise and highest sensitivity.https://www.mdpi.com/2379-139X/11/5/51multi-slice CTdual-energy CTimage enhancementU-Net modeldeep learningimage reconstruction
spellingShingle Haoyan Li
Zhenpeng Chen
Shuaiyi Gao
Jiaqi Hu
Zhihao Yang
Yun Peng
Jihang Sun
Performance Evaluation of Image Segmentation Using Dual-Energy Spectral CT Images with Deep Learning Image Reconstruction: A Phantom Study
Tomography
multi-slice CT
dual-energy CT
image enhancement
U-Net model
deep learning
image reconstruction
title Performance Evaluation of Image Segmentation Using Dual-Energy Spectral CT Images with Deep Learning Image Reconstruction: A Phantom Study
title_full Performance Evaluation of Image Segmentation Using Dual-Energy Spectral CT Images with Deep Learning Image Reconstruction: A Phantom Study
title_fullStr Performance Evaluation of Image Segmentation Using Dual-Energy Spectral CT Images with Deep Learning Image Reconstruction: A Phantom Study
title_full_unstemmed Performance Evaluation of Image Segmentation Using Dual-Energy Spectral CT Images with Deep Learning Image Reconstruction: A Phantom Study
title_short Performance Evaluation of Image Segmentation Using Dual-Energy Spectral CT Images with Deep Learning Image Reconstruction: A Phantom Study
title_sort performance evaluation of image segmentation using dual energy spectral ct images with deep learning image reconstruction a phantom study
topic multi-slice CT
dual-energy CT
image enhancement
U-Net model
deep learning
image reconstruction
url https://www.mdpi.com/2379-139X/11/5/51
work_keys_str_mv AT haoyanli performanceevaluationofimagesegmentationusingdualenergyspectralctimageswithdeeplearningimagereconstructionaphantomstudy
AT zhenpengchen performanceevaluationofimagesegmentationusingdualenergyspectralctimageswithdeeplearningimagereconstructionaphantomstudy
AT shuaiyigao performanceevaluationofimagesegmentationusingdualenergyspectralctimageswithdeeplearningimagereconstructionaphantomstudy
AT jiaqihu performanceevaluationofimagesegmentationusingdualenergyspectralctimageswithdeeplearningimagereconstructionaphantomstudy
AT zhihaoyang performanceevaluationofimagesegmentationusingdualenergyspectralctimageswithdeeplearningimagereconstructionaphantomstudy
AT yunpeng performanceevaluationofimagesegmentationusingdualenergyspectralctimageswithdeeplearningimagereconstructionaphantomstudy
AT jihangsun performanceevaluationofimagesegmentationusingdualenergyspectralctimageswithdeeplearningimagereconstructionaphantomstudy