Frontal plane mechanical leg alignment estimation from knee x-rays using deep learning

Kenneth Chen , Christoph Stotter , Christopher Lepenik , Thomas Klestil , Christoph Salzlechner , Stefan Nehrer
{"title":"Frontal plane mechanical leg alignment estimation from knee x-rays using deep learning","authors":"Kenneth Chen ,&nbsp;Christoph Stotter ,&nbsp;Christopher Lepenik ,&nbsp;Thomas Klestil ,&nbsp;Christoph Salzlechner ,&nbsp;Stefan Nehrer","doi":"10.1016/j.ocarto.2024.100551","DOIUrl":null,"url":null,"abstract":"<div><h3>Objective</h3><div>Lower limb malalignment can complicate symptoms and accelerate knee osteoarthritis (OA), necessitating consideration in study population selection. In this study, we develop and validate a deep learning model that classifies leg alignment as “normal” or “malaligned” from knee antero-posterior (AP)/postero-anterior (PA) radiographs alone, using an adjustable hip-knee-ankle (HKA) angle threshold.</div></div><div><h3>Material and methods</h3><div>We utilized 8878 digital radiographs, including 6181 AP/PA full-leg x-rays (LLRs) and 2697 AP/PA knee x-rays (2292 with positioning frame, 405 without). The model's evaluation involved two steps: In step 1, the model's predictions on knee images cropped from LLRs were compared against the ground truth from the original LLRs. In step 2, the model was tested on knee AP radiographs, using corresponding same-day LLRs as a proxy for ground truth.</div></div><div><h3>Results</h3><div>The model effectively classified alignment, with step one achieving sensitivity and specificity of 0.92 for a threshold of 7.5°, and 0.90 and 0.85 for 5°. For positioning frame images, step two showed a sensitivity of 0.85 and specificity of 0.81 for 7.5°, and 0.79 and 0.74 for 5°. For non-positioning frame images, sensitivity and specificity were 0.91 and 0.83 for 7.5°, and 0.9 and 0.86 for 5°.</div></div><div><h3>Conclusion</h3><div>The model developed in this study accurately classifies lower limb malalignment from AP/PA knee radiographs using adjustable thresholds, offering a practical alternative to LLRs. This can enhance the precision of study population selection and patient management.</div></div>","PeriodicalId":74377,"journal":{"name":"Osteoarthritis and cartilage open","volume":"7 1","pages":"Article 100551"},"PeriodicalIF":0.0000,"publicationDate":"2024-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11729668/pdf/","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Osteoarthritis and cartilage open","FirstCategoryId":"1085","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2665913124001183","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

Abstract

Objective

Lower limb malalignment can complicate symptoms and accelerate knee osteoarthritis (OA), necessitating consideration in study population selection. In this study, we develop and validate a deep learning model that classifies leg alignment as “normal” or “malaligned” from knee antero-posterior (AP)/postero-anterior (PA) radiographs alone, using an adjustable hip-knee-ankle (HKA) angle threshold.

Material and methods

We utilized 8878 digital radiographs, including 6181 AP/PA full-leg x-rays (LLRs) and 2697 AP/PA knee x-rays (2292 with positioning frame, 405 without). The model's evaluation involved two steps: In step 1, the model's predictions on knee images cropped from LLRs were compared against the ground truth from the original LLRs. In step 2, the model was tested on knee AP radiographs, using corresponding same-day LLRs as a proxy for ground truth.

Results

The model effectively classified alignment, with step one achieving sensitivity and specificity of 0.92 for a threshold of 7.5°, and 0.90 and 0.85 for 5°. For positioning frame images, step two showed a sensitivity of 0.85 and specificity of 0.81 for 7.5°, and 0.79 and 0.74 for 5°. For non-positioning frame images, sensitivity and specificity were 0.91 and 0.83 for 7.5°, and 0.9 and 0.86 for 5°.

Conclusion

The model developed in this study accurately classifies lower limb malalignment from AP/PA knee radiographs using adjustable thresholds, offering a practical alternative to LLRs. This can enhance the precision of study population selection and patient management.
使用深度学习的膝关节x射线的正面机械腿对齐估计。
目的:下肢错位可使膝关节骨性关节炎(OA)症状复杂化并加速发病,在研究人群选择时需要加以考虑。在本研究中,我们开发并验证了一个深度学习模型,该模型使用可调节的髋关节-膝关节-踝关节(HKA)角度阈值,仅从膝关节前后(AP)/后前(PA) x线片将腿部对齐分类为“正常”或“不对齐”。材料和方法:我们使用8878张数字x线片,包括6181张AP/PA全腿x线片(LLRs)和2697张AP/PA膝关节x线片(2292张带定位架,405张不带定位架)。该模型的评估包括两个步骤:第一步,将模型对从llr中裁剪的膝盖图像的预测与原始llr的真实情况进行比较。在步骤2中,在膝关节AP x线片上测试该模型,使用相应的当日llr作为地面真实值的代理。结果:该模型对对齐进行了有效分类,第一步对7.5°阈值的敏感性和特异性分别为0.92,对5°阈值的敏感性和特异性分别为0.90和0.85。对于定位帧图像,步骤二对于7.5°的灵敏度为0.85,特异度为0.81,对于5°的灵敏度为0.79,特异度为0.74。对于非定位帧图像,7.5°的灵敏度和特异性分别为0.91和0.83,5°的灵敏度和特异性分别为0.9和0.86。结论:本研究中开发的模型使用可调节阈值准确地从AP/PA膝关节x线片中分类下肢错位,为llr提供了一种实用的替代方案。这可以提高研究人群选择和患者管理的准确性。
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
Osteoarthritis and cartilage open
Osteoarthritis and cartilage open Orthopedics, Sports Medicine and Rehabilitation
CiteScore
3.30
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信