A device-dependent auto-segmentation method based on combined generalized and single-device datasets

IF 3.2 2区医学 Q1 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING

Medical physics Pub Date : 2024-12-19 DOI:10.1002/mp.17570

Hyeongjin Lim, Yongha Gi, Yousun Ko, Yunhui Jo, Jinyoung Hong, Jonghyun Kim, Sung Hwan Ahn, Hee-Chul Park, Haeyoung Kim, Kwangzoo Chung, Myonggeun Yoon

{"title":"A device-dependent auto-segmentation method based on combined generalized and single-device datasets","authors":"Hyeongjin Lim, Yongha Gi, Yousun Ko, Yunhui Jo, Jinyoung Hong, Jonghyun Kim, Sung Hwan Ahn, Hee-Chul Park, Haeyoung Kim, Kwangzoo Chung, Myonggeun Yoon","doi":"10.1002/mp.17570","DOIUrl":null,"url":null,"abstract":"<div>\n \n \n <section>\n \n <h3> Background</h3>\n \n <p>Although generalized-dataset-based auto-segmentation models that consider various computed tomography (CT) scanners have shown great clinical potential, their application to medical images from unseen scanners remains challenging because of device-dependent image features.</p>\n </section>\n \n <section>\n \n <h3> Purpose</h3>\n \n <p>This study aims to investigate the performance of a device-dependent auto-segmentation model based on a combined dataset of a generalized dataset and single CT scanner dataset.</p>\n </section>\n \n <section>\n \n <h3> Method</h3>\n \n <p>We constructed two training datasets for 21 chest and abdominal organs. The generalized dataset comprised 1203 publicly available multi-scanner data. The device-dependent dataset comprised 1253 data, including the 1203 multi-CT scanner data and 50 single CT scanner data. Using these datasets, the generalized-dataset-based model (GDSM) and the device-dependent-dataset-based model (DDSM) were trained. The models were trained using nnU-Net and tested on ten data samples from a single CT scanner. The evaluation metrics included the Dice similarity coefficient (DSC), the Hausdorff distance (HD), and the average symmetric surface distance (ASSD), which were used to assess the overall performance of the models. In addition, DSC<sub>diff</sub>, HD<sub>ratio</sub>, and ASSD<sub>ratio</sub>, which are variations of the three existing metrics, were used to compare the performance of the models across different organs.</p>\n </section>\n \n <section>\n \n <h3> Result</h3>\n \n <p>For the average DSC, the GDSM and DDSM had values of 0.9251 and 0.9323, respectively. For the average HD, the GDSM and DDSM had values of 10.66 and 9.139 mm, respectively; for the average ASSD, the GDSM and DDSM had values of 0.8318 and 0.6656 mm, respectively. Compared with the GDSM, the DDSM showed consistent performance improvements of 0.78%, 14%, and 20% for the DSC, HD, and ASSD metrics, respectively. In addition, compared with the GDSM, the DDSM had better DSC<sub>diff</sub> values in 14 of 21 tested organs, better HD<sub>ratio</sub> values in 13 of 21 tested organs, and better ASSD<sub>ratio</sub> values in 14 of 21 tested organs. The three averages of the variant metrics were all better for the DDSM than for the GDSM.</p>\n </section>\n \n <section>\n \n <h3> Conclusion</h3>\n \n <p>The results suggest that combining the generalized dataset with a single scanner dataset resulted in an overall improvement in model performance for that device image.</p>\n </section>\n </div>","PeriodicalId":18384,"journal":{"name":"Medical physics","volume":"52 4","pages":"2375-2383"},"PeriodicalIF":3.2000,"publicationDate":"2024-12-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Medical physics","FirstCategoryId":"3","ListUrlMain":"https://onlinelibrary.wiley.com/doi/10.1002/mp.17570","RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING","Score":null,"Total":0}

引用次数: 0

Abstract

Background

Although generalized-dataset-based auto-segmentation models that consider various computed tomography (CT) scanners have shown great clinical potential, their application to medical images from unseen scanners remains challenging because of device-dependent image features.

Purpose

This study aims to investigate the performance of a device-dependent auto-segmentation model based on a combined dataset of a generalized dataset and single CT scanner dataset.

Method

We constructed two training datasets for 21 chest and abdominal organs. The generalized dataset comprised 1203 publicly available multi-scanner data. The device-dependent dataset comprised 1253 data, including the 1203 multi-CT scanner data and 50 single CT scanner data. Using these datasets, the generalized-dataset-based model (GDSM) and the device-dependent-dataset-based model (DDSM) were trained. The models were trained using nnU-Net and tested on ten data samples from a single CT scanner. The evaluation metrics included the Dice similarity coefficient (DSC), the Hausdorff distance (HD), and the average symmetric surface distance (ASSD), which were used to assess the overall performance of the models. In addition, DSC_diff, HD_ratio, and ASSD_ratio, which are variations of the three existing metrics, were used to compare the performance of the models across different organs.

Result

For the average DSC, the GDSM and DDSM had values of 0.9251 and 0.9323, respectively. For the average HD, the GDSM and DDSM had values of 10.66 and 9.139 mm, respectively; for the average ASSD, the GDSM and DDSM had values of 0.8318 and 0.6656 mm, respectively. Compared with the GDSM, the DDSM showed consistent performance improvements of 0.78%, 14%, and 20% for the DSC, HD, and ASSD metrics, respectively. In addition, compared with the GDSM, the DDSM had better DSC_diff values in 14 of 21 tested organs, better HD_ratio values in 13 of 21 tested organs, and better ASSD_ratio values in 14 of 21 tested organs. The three averages of the variant metrics were all better for the DDSM than for the GDSM.

Conclusion

The results suggest that combining the generalized dataset with a single scanner dataset resulted in an overall improvement in model performance for that device image.

查看原文本刊更多论文

一种基于综合和单设备数据集的设备相关自动分割方法。

背景：尽管考虑各种计算机断层扫描（CT）扫描仪的基于广义数据集的自动分割模型显示出巨大的临床潜力，但由于设备依赖的图像特征，它们在未见扫描仪的医学图像中的应用仍然具有挑战性。目的：研究基于广义数据集和单个CT扫描仪数据集的组合数据集的设备相关自动分割模型的性能。方法：构建2个胸腹器官训练数据集。广义数据集包含1203个公开可用的多扫描仪数据。设备相关数据集包括1253个数据，其中包括1203个多台CT扫描仪数据和50个单台CT扫描仪数据。利用这些数据集，训练了基于广义数据集的模型（GDSM）和基于设备依赖数据集的模型（DDSM）。这些模型使用nnU-Net进行训练，并在一台CT扫描仪的10个数据样本上进行测试。评估指标包括Dice相似系数（DSC）、Hausdorff距离（HD）和平均对称表面距离（ASSD），用于评估模型的整体性能。此外，DSCdiff、HDratio和ASSDratio是三个现有指标的变体，用于比较模型在不同器官中的性能。结果：对于平均DSC， GDSM和DDSM分别为0.9251和0.9323。平均HD的GDSM和DDSM分别为10.66和9.139 mm；平均ASSD的GDSM和DDSM分别为0.8318和0.6656 mm。与GDSM相比，DDSM在DSC、HD和ASSD指标上的表现分别提高了0.78%、14%和20%。此外，与GDSM相比，DDSM在21个测试器官中有14个器官的DSCdiff值更好，在21个测试器官中有13个器官的hratio值更好，在21个测试器官中有14个器官的assratio值更好。不同指标的三个平均值对于DDSM来说都比GDSM好。结论：结果表明，将广义数据集与单个扫描仪数据集相结合，可以全面提高该设备图像的模型性能。

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

Medical physics 医学-核医学

CiteScore

6.80

自引率

15.80%

发文量

660

审稿时长

1.7 months

期刊介绍： Medical Physics publishes original, high impact physics, imaging science, and engineering research that advances patient diagnosis and therapy through contributions in 1) Basic science developments with high potential for clinical translation 2) Clinical applications of cutting edge engineering and physics innovations 3) Broadly applicable and innovative clinical physics developments Medical Physics is a journal of global scope and reach. By publishing in Medical Physics your research will reach an international, multidisciplinary audience including practicing medical physicists as well as physics- and engineering based translational scientists. We work closely with authors of promising articles to improve their quality.