Improved CT-based Osteoporosis Assessment with a Fully Automated Deep Learning Tool.

IF 8.1 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Radiology-Artificial Intelligence Pub Date : 2022-08-31 eCollection Date: 2022-09-01 DOI:10.1148/ryai.220042

Perry J Pickhardt, Thang Nguyen, Alberto A Perez, Peter M Graffy, Samuel Jang, Ronald M Summers, John W Garrett

{"title":"Improved CT-based Osteoporosis Assessment with a Fully Automated Deep Learning Tool.","authors":"Perry J Pickhardt, Thang Nguyen, Alberto A Perez, Peter M Graffy, Samuel Jang, Ronald M Summers, John W Garrett","doi":"10.1148/ryai.220042","DOIUrl":null,"url":null,"abstract":"Purpose: To develop, test, and validate a deep learning (DL) tool that improves upon a previous feature-based CT image processing bone mineral density (BMD) algorithm and compare it against the manual reference standard.Materials and methods: This single-center, retrospective, Health Insurance Portability and Accountability Act-compliant study included manual L1 trabecular Hounsfield unit measurements from abdominal CT scans in 11 035 patients (mean age, 58 years ± 12 [SD]; 6311 women) as the reference standard. Automated level selection and L1 trabecular region of interest (ROI) placement were then performed in this CT cohort with both a previously validated feature-based image processing tool and a new DL tool. Overall technical success rates and agreement with the manual reference standard were assessed.Results: The overall success rate of the DL tool in this heterogeneous patient cohort was significantly higher than that of the older image processing BMD algorithm (99.3% vs 89.4%, P < .001). Using this DL tool, the closest median Hounsfield unit values for single-, three-, and seven-slice vertebral ROIs were within 5% of the manual reference standard Hounsfield unit values in 35.1%, 56.9%, and 85.8% of scans; within 10% in 56.6%, 75.6%, and 92.9% of scans; and within 25% in 76.5%, 89.3%, and 97.1% of scans, respectively. Trade-offs in sensitivity and specificity for osteoporosis assessment were observed from the single-slice approach (sensitivity, 39.4%; specificity, 98.3%) to the minimum value of the multislice approach (for seven contiguous slices; sensitivity, 71.3% and specificity, 94.6%).Conclusion: The new DL BMD tool demonstrated a higher success rate than the older feature-based image processing tool, and its outputs can be targeted for higher specificity or sensitivity for osteoporosis assessment.Keywords: CT, CT-Quantitative, Abdomen/GI, Skeletal-Axial, Spine, Deep Learning, Machine Learning Supplemental material is available for this article. © RSNA, 2022.","PeriodicalId":29787,"journal":{"name":"Radiology-Artificial Intelligence","volume":"4 5","pages":"e220042"},"PeriodicalIF":8.1000,"publicationDate":"2022-08-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9530763/pdf/ryai.220042.pdf","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Radiology-Artificial Intelligence","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1148/ryai.220042","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2022/9/1 0:00:00","PubModel":"eCollection","JCR":"Q1","JCRName":"COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE","Score":null,"Total":0}

引用次数: 0

Abstract

Purpose: To develop, test, and validate a deep learning (DL) tool that improves upon a previous feature-based CT image processing bone mineral density (BMD) algorithm and compare it against the manual reference standard.

Materials and methods: This single-center, retrospective, Health Insurance Portability and Accountability Act-compliant study included manual L1 trabecular Hounsfield unit measurements from abdominal CT scans in 11 035 patients (mean age, 58 years ± 12 [SD]; 6311 women) as the reference standard. Automated level selection and L1 trabecular region of interest (ROI) placement were then performed in this CT cohort with both a previously validated feature-based image processing tool and a new DL tool. Overall technical success rates and agreement with the manual reference standard were assessed.

Results: The overall success rate of the DL tool in this heterogeneous patient cohort was significantly higher than that of the older image processing BMD algorithm (99.3% vs 89.4%, P < .001). Using this DL tool, the closest median Hounsfield unit values for single-, three-, and seven-slice vertebral ROIs were within 5% of the manual reference standard Hounsfield unit values in 35.1%, 56.9%, and 85.8% of scans; within 10% in 56.6%, 75.6%, and 92.9% of scans; and within 25% in 76.5%, 89.3%, and 97.1% of scans, respectively. Trade-offs in sensitivity and specificity for osteoporosis assessment were observed from the single-slice approach (sensitivity, 39.4%; specificity, 98.3%) to the minimum value of the multislice approach (for seven contiguous slices; sensitivity, 71.3% and specificity, 94.6%).

Conclusion: The new DL BMD tool demonstrated a higher success rate than the older feature-based image processing tool, and its outputs can be targeted for higher specificity or sensitivity for osteoporosis assessment.Keywords: CT, CT-Quantitative, Abdomen/GI, Skeletal-Axial, Spine, Deep Learning, Machine Learning Supplemental material is available for this article. © RSNA, 2022.

查看原文本刊更多论文

利用全自动深度学习工具改进基于 CT 的骨质疏松症评估。

目的：开发、测试和验证一种深度学习（DL）工具，该工具改进了以前基于特征的 CT 图像处理骨矿密度（BMD）算法，并将其与人工参考标准进行比较：这项符合《健康保险可携性与责任法案》（Health Insurance Portability and Accountability Act）的单中心回顾性研究将 11 035 名患者（平均年龄为 58 岁 ± 12 [SD]；6311 名女性）腹部 CT 扫描的 L1 小梁 Hounsfield 单位手动测量值作为参考标准。然后，利用以前验证过的基于特征的图像处理工具和新的 DL 工具，在该 CT 队列中进行了自动水平选择和 L1 小梁感兴趣区 (ROI) 放置。评估了总体技术成功率以及与人工参考标准的一致性：结果：在这一异质性患者队列中，DL 工具的总体成功率明显高于旧版图像处理 BMD 算法（99.3% vs 89.4%，P < .001）。使用该 DL 工具，在 35.1%、56.9% 和 85.8% 的扫描中，单片、三片和七片椎体 ROI 的最接近中位 Hounsfield 单位值与手动参考标准 Hounsfield 单位值的比值在 5%以内；在 56.6%、75.6% 和 92.9% 的扫描中，最接近中位 Hounsfield 单位值的比值在 10%以内；在 76.5%、89.3% 和 97.1% 的扫描中，最接近中位 Hounsfield 单位值的比值在 25%以内。从单片方法（灵敏度 39.4%；特异性 98.3%）到多片方法的最小值（七个连续切片；灵敏度 71.3%；特异性 94.6%），骨质疏松症评估的灵敏度和特异性之间存在权衡：结论：新的 DL BMD 工具比旧的基于特征的图像处理工具显示出更高的成功率，其输出结果可为骨质疏松症评估提供更高的特异性或灵敏度：CT、CT 定量、腹部/GI、骨骼轴向、脊柱、深度学习、机器学习本文有补充材料。© RSNA, 2022.

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

Radiology-Artificial Intelligence

CiteScore

16.20

自引率

1.00%

发文量

期刊介绍： Radiology: Artificial Intelligence is a bi-monthly publication that focuses on the emerging applications of machine learning and artificial intelligence in the field of imaging across various disciplines. This journal is available online and accepts multiple manuscript types, including Original Research, Technical Developments, Data Resources, Review articles, Editorials, Letters to the Editor and Replies, Special Reports, and AI in Brief.