Deep learning based ultra-low dose fan-beam computed tomography image enhancement algorithm: Feasibility study in image quality for radiotherapy.

IF 2 4区医学 Q3 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

Journal of Applied Clinical Medical Physics Pub Date : 2024-11-14 DOI:10.1002/acm2.14560

Hua Jiang, Songbing Qin, Lecheng Jia, Ziquan Wei, Weiqi Xiong, Wentao Xu, Wei Gong, Wei Zhang, Liqin Yu

{"title":"Deep learning based ultra-low dose fan-beam computed tomography image enhancement algorithm: Feasibility study in image quality for radiotherapy.","authors":"Hua Jiang, Songbing Qin, Lecheng Jia, Ziquan Wei, Weiqi Xiong, Wentao Xu, Wei Gong, Wei Zhang, Liqin Yu","doi":"10.1002/acm2.14560","DOIUrl":null,"url":null,"abstract":"Objective: We investigated the feasibility of deep learning-based ultra-low dose kV-fan-beam computed tomography (kV-FBCT) image enhancement algorithm for clinical application in abdominal and pelvic tumor radiotherapy.Methods: A total of 76 patients of abdominal and pelvic tumors were prospectively selected. The Catphan504 was acquired with the same conditions as the standard phantom test set. We used a CycleGAN-based model for image enhancement. Normal dose CT (NDCT), ultra-low dose CT (LDCT) and deep learning enhanced CT (DLR) were evaluated by subjective and objective analyses in terms of imaging quality, HU accuracy, and image signal-to-noise ratio (SNR).Results: The image noise of DLR was significantly reduced, and the contrast-to-noise ratio (CNR) was significantly improved compared to the LDCT. The most significant improvement was the acrylic which represented soft tissue in CNR from 1.89 to 3.37, improving by 76%, nearly approaching the NDCT, and in low-density resolution from 7.64 to 12.6, improving by 64%. The spatial frequencies of MTF10 and MTF50 in DLR were 4.28 and 2.35 cycles/mm in DLR, respectively, which are higher than LDCT 3.87 and 2.12 cycles/mm, and even slightly higher than NDCT 4.15 and 2.28 cycles/mm. The accuracy and stability of HU values of DLR were similar to NDCT. The image quality evaluation of the two doctors agreed well with DLR and NDCT. A two-by-two comparison between groups showed that the differences in image scores of LDCT compared with NDCT and DLR were all statistically significant (p < 0.05), and the subjective scores of DLR were close to NDCT.Conclusion: The image quality of DLR was close to NDCT with reduced radiation dose, which can fully meet the needs of conventional image-guided adaptive radiotherapy (ART) and achieve the quality requirements of clinical radiotherapy. The proposed method provided a technical basis for LDCT-guided ART.","PeriodicalId":14989,"journal":{"name":"Journal of Applied Clinical Medical Physics","volume":null,"pages":null},"PeriodicalIF":2.0000,"publicationDate":"2024-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Applied Clinical Medical Physics","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1002/acm2.14560","RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING","Score":null,"Total":0}

引用次数: 0

Abstract

Objective: We investigated the feasibility of deep learning-based ultra-low dose kV-fan-beam computed tomography (kV-FBCT) image enhancement algorithm for clinical application in abdominal and pelvic tumor radiotherapy.

Methods: A total of 76 patients of abdominal and pelvic tumors were prospectively selected. The Catphan504 was acquired with the same conditions as the standard phantom test set. We used a CycleGAN-based model for image enhancement. Normal dose CT (NDCT), ultra-low dose CT (LDCT) and deep learning enhanced CT (DLR) were evaluated by subjective and objective analyses in terms of imaging quality, HU accuracy, and image signal-to-noise ratio (SNR).

Results: The image noise of DLR was significantly reduced, and the contrast-to-noise ratio (CNR) was significantly improved compared to the LDCT. The most significant improvement was the acrylic which represented soft tissue in CNR from 1.89 to 3.37, improving by 76%, nearly approaching the NDCT, and in low-density resolution from 7.64 to 12.6, improving by 64%. The spatial frequencies of MTF10 and MTF50 in DLR were 4.28 and 2.35 cycles/mm in DLR, respectively, which are higher than LDCT 3.87 and 2.12 cycles/mm, and even slightly higher than NDCT 4.15 and 2.28 cycles/mm. The accuracy and stability of HU values of DLR were similar to NDCT. The image quality evaluation of the two doctors agreed well with DLR and NDCT. A two-by-two comparison between groups showed that the differences in image scores of LDCT compared with NDCT and DLR were all statistically significant (p < 0.05), and the subjective scores of DLR were close to NDCT.

Conclusion: The image quality of DLR was close to NDCT with reduced radiation dose, which can fully meet the needs of conventional image-guided adaptive radiotherapy (ART) and achieve the quality requirements of clinical radiotherapy. The proposed method provided a technical basis for LDCT-guided ART.

查看原文本刊更多论文

基于深度学习的超低剂量扇形光束计算机断层扫描图像增强算法：放疗图像质量的可行性研究。

目的研究基于深度学习的超低剂量kV-扇形束计算机断层扫描（kV-FBCT）图像增强算法在腹部和盆腔肿瘤放疗中临床应用的可行性：前瞻性地选择了76例腹部和盆腔肿瘤患者。Catphan504的采集条件与标准模型测试集相同。我们使用基于 CycleGAN 的模型进行图像增强。通过主观和客观分析，从成像质量、HU 精确度和图像信噪比（SNR）等方面对正常剂量 CT（NDCT）、超低剂量 CT（LDCT）和深度学习增强 CT（DLR）进行了评估：与 LDCT 相比，DLR 的图像噪声明显降低，对比度-噪声比（CNR）显著提高。改善最明显的是代表软组织的丙烯酸，CNR 从 1.89 提高到 3.37，提高了 76%，几乎接近 NDCT；低密度分辨率从 7.64 提高到 12.6，提高了 64%。DLR的MTF10和MTF50空间频率分别为4.28和2.35周期/毫米，高于LDCT的3.87和2.12周期/毫米，甚至略高于NDCT的4.15和2.28周期/毫米。DLR HU 值的准确性和稳定性与 NDCT 相似。两位医生对 DLR 和 NDCT 的图像质量评价一致。组间两两比较显示，LDCT 与 NDCT 和 DLR 相比，图像评分差异均有统计学意义（P 结论：LDCT 与 NDCT 的图像质量相近：DLR的图像质量接近NDCT，辐射剂量降低，完全可以满足常规影像引导自适应放疗（ART）的需要，达到临床放疗的质量要求。该方法为 LDCT 引导的 ART 提供了技术基础。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Journal of Applied Clinical Medical Physics 医学-核医学

CiteScore

3.60

自引率

19.00%

发文量

331

审稿时长

3 months

期刊介绍： Journal of Applied Clinical Medical Physics is an international Open Access publication dedicated to clinical medical physics. JACMP welcomes original contributions dealing with all aspects of medical physics from scientists working in the clinical medical physics around the world. JACMP accepts only online submission. JACMP will publish: -Original Contributions: Peer-reviewed, investigations that represent new and significant contributions to the field. Recommended word count: up to 7500. -Review Articles: Reviews of major areas or sub-areas in the field of clinical medical physics. These articles may be of any length and are peer reviewed. -Technical Notes: These should be no longer than 3000 words, including key references. -Letters to the Editor: Comments on papers published in JACMP or on any other matters of interest to clinical medical physics. These should not be more than 1250 (including the literature) and their publication is only based on the decision of the editor, who occasionally asks experts on the merit of the contents. -Book Reviews: The editorial office solicits Book Reviews. -Announcements of Forthcoming Meetings: The Editor may provide notice of forthcoming meetings, course offerings, and other events relevant to clinical medical physics. -Parallel Opposed Editorial: We welcome topics relevant to clinical practice and medical physics profession. The contents can be controversial debate or opposed aspects of an issue. One author argues for the position and the other against. Each side of the debate contains an opening statement up to 800 words, followed by a rebuttal up to 500 words. Readers interested in participating in this series should contact the moderator with a proposed title and a short description of the topic