Consistency of 3D reconstruction measurements of lower limb torsion at different segments with the CT gold standard

Abstract This study aimed to assess the agreement between lower limb segmental torsion measurements obtained via three-dimensional (3D) reconstruction and conventional two-dimensional (2D) CT cross-sections, and to evaluate the reliability and repeatability of 3D model in the measurement of femoral...

Full description

Saved in:
Bibliographic Details
Main Authors: Shouliang Xiong, Yunpeng Luo, Shikuan Zhang, Zhongxuan Wu, Degang Huang, Hongzhou Sun
Format: Article
Language:English
Published: Nature Portfolio 2025-05-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-025-01262-1
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850231270715424768
author Shouliang Xiong
Yunpeng Luo
Shikuan Zhang
Zhongxuan Wu
Degang Huang
Hongzhou Sun
author_facet Shouliang Xiong
Yunpeng Luo
Shikuan Zhang
Zhongxuan Wu
Degang Huang
Hongzhou Sun
author_sort Shouliang Xiong
collection DOAJ
description Abstract This study aimed to assess the agreement between lower limb segmental torsion measurements obtained via three-dimensional (3D) reconstruction and conventional two-dimensional (2D) CT cross-sections, and to evaluate the reliability and repeatability of 3D model in the measurement of femoral and tibial torsion angle. By verifying the comparison results of 3D measurement method and 2D gold standard, the applicability of 3D measurement method in clinical diagnosis was analyzed, which provided data support for the evaluation of lower limb torsion angle in patients with Patellofemoral pain syndrome (PFPS). The CT scan data of lower limbs of 80 patients (56 females, 24 males) diagnosed with PFPS in our hospital from August 2022 to July 2024 were collected. The age range of patients was 10 to 79 years old (average 54 years old). All patients underwent CT plain scan of bilateral lower extremity three joints (hip-knee-ankle) fixation. The scan data were used for the measurement and analysis of 2D and 3D torsion angles. According to the Seitlinger method and Winkler method, the femoral anteversion angle and tibial torsion angle were measured in 2D segments. Mimics software was used to construct a three-dimensional model of the hip, knee and ankle of the patient ‘s lower limbs. Referring to the previous measurement methods of Kim et al.and other researchers, the anatomical axis of each segment of the femur and tibia was defined by the three-dimensional model and the angle was calculated. The 3D model is generated by global threshold segmentation and regional growth to ensure the accurate identification of anatomical markers and the accuracy of axis drawing. Intraclass correlation coefficient (ICC) was used to evaluate the inter-observer consistency of 2D and 3D measurement methods and the consistency between the two methods. Paired t test was used to compare the difference between the results of 2D and 3D methods, and the significance level was set to P < 0.05. For 2D measurements, inter-observer reliability was high, with ICC values (95% CI) for total femoral anteversion (TFA) and middle femoral anteversion (MFA) of 0.981 (95% CI 0.975–0.986) and 0.978 (95% CI 0.971–0.984), respectively. Proximal femoral anteversion (PFA) and distal femoral anteversion (DFA) had ICC values of 0.921 (95% CI 0.894–0.942) and 0.927 (95% CI 0.901–0.946), respectively. Total tibial torsion (TTT), proximal tibial torsion (PTT), and distal tibial torsion (DTT) showed ICC values of 0.940 (95% CI 0.918–0.955), 0.978 (95% CI 0.970–0.984), and 0.985 (95% CI 0.980–0.989), respectively.For 3D measurements, TFA and MFA had ICC values of 0.977 (95% CI 0.968–0.983) and 0.983 (95% CI 0.977–0.988), respectively, while PFA and DFA showed ICC values of 0.928 (95% CI 0.903–0.947) and 0.868 (95% CI 0.824–0.902). TTT, PTT, and DTT exhibited ICC values of 0.925 (95% CI 0.899–0.944), 0.971 (95% CI 0.961–0.979), and 0.981 (95% CI 0.974–0.986), respectively. Observer A’s consistency results for 2D and 3D measurements indicated an ICC of 0.984 (95% CI 0.978–0.988) for TFA and 0.981 (95% CI 0.974–0.986) for MFA, with PFA and DFA having ICC values of 0.925 (95% CI 0.899–0.944) and 0.897 (95% CI 0.862–0.923), respectively. TTT, PTT, and DTT demonstrated ICC values of 0.947 (95% CI 0.929–0.961), 0.969 (95% CI 0.958–0.977), and 0.983 (95% CI 0.976–0.987), respectively. Observer B showed similar consistency, with ICC values of 0.979 (95% CI 0.972–0.985) for TFA and 0.985 (95% CI 0.980–0.989) for MFA, and ICC values of 0.916 (95% CI 0.887–0.938) and 0.888 (95% CI 0.850–0.917) for PFA and DFA, respectively. TTT, PTT, and DTT exhibited ICC values of 0.936 (95% CI 0.914–0.953), 0.965 (95% CI 0.953–0.974), and 0.986 (95% CI 0.981–0.990), respectively. Paired t-tests showed no statistically significant differences between 2D and 3D measurements for all major parameters (P > 0.05). Our findings confirm that 3D measurement methods exhibit high reliability and agreement with the CT gold standard. Given their ability to provide comprehensive segmental torsion analysis with lower observer dependence, 3D methods have strong potential to improve clinical decision-making in PFPS treatment and surgical planning.
format Article
id doaj-art-c00528221a7c4b978045e59a79c975d2
institution OA Journals
issn 2045-2322
language English
publishDate 2025-05-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-c00528221a7c4b978045e59a79c975d22025-08-20T02:03:35ZengNature PortfolioScientific Reports2045-23222025-05-0115111210.1038/s41598-025-01262-1Consistency of 3D reconstruction measurements of lower limb torsion at different segments with the CT gold standardShouliang Xiong0Yunpeng Luo1Shikuan Zhang2Zhongxuan Wu3Degang Huang4Hongzhou Sun5Department of Joint Surgery, Yijishan Hospital, The First Affliated Hospital of Wannan Medical CollegeDepartment of Joint Surgery, Yijishan Hospital, The First Affliated Hospital of Wannan Medical CollegeDepartment of Joint Surgery, Yijishan Hospital, The First Affliated Hospital of Wannan Medical CollegeDepartment of Joint Surgery, Yijishan Hospital, The First Affliated Hospital of Wannan Medical CollegeDepartment of Joint Surgery, Yijishan Hospital, The First Affliated Hospital of Wannan Medical CollegeThe Second Orthopaedic Department, Changzhou Medical District, No. 904 Hospital, Joint Logistics Support Force, Chinese People ’s Liberation ArmyAbstract This study aimed to assess the agreement between lower limb segmental torsion measurements obtained via three-dimensional (3D) reconstruction and conventional two-dimensional (2D) CT cross-sections, and to evaluate the reliability and repeatability of 3D model in the measurement of femoral and tibial torsion angle. By verifying the comparison results of 3D measurement method and 2D gold standard, the applicability of 3D measurement method in clinical diagnosis was analyzed, which provided data support for the evaluation of lower limb torsion angle in patients with Patellofemoral pain syndrome (PFPS). The CT scan data of lower limbs of 80 patients (56 females, 24 males) diagnosed with PFPS in our hospital from August 2022 to July 2024 were collected. The age range of patients was 10 to 79 years old (average 54 years old). All patients underwent CT plain scan of bilateral lower extremity three joints (hip-knee-ankle) fixation. The scan data were used for the measurement and analysis of 2D and 3D torsion angles. According to the Seitlinger method and Winkler method, the femoral anteversion angle and tibial torsion angle were measured in 2D segments. Mimics software was used to construct a three-dimensional model of the hip, knee and ankle of the patient ‘s lower limbs. Referring to the previous measurement methods of Kim et al.and other researchers, the anatomical axis of each segment of the femur and tibia was defined by the three-dimensional model and the angle was calculated. The 3D model is generated by global threshold segmentation and regional growth to ensure the accurate identification of anatomical markers and the accuracy of axis drawing. Intraclass correlation coefficient (ICC) was used to evaluate the inter-observer consistency of 2D and 3D measurement methods and the consistency between the two methods. Paired t test was used to compare the difference between the results of 2D and 3D methods, and the significance level was set to P < 0.05. For 2D measurements, inter-observer reliability was high, with ICC values (95% CI) for total femoral anteversion (TFA) and middle femoral anteversion (MFA) of 0.981 (95% CI 0.975–0.986) and 0.978 (95% CI 0.971–0.984), respectively. Proximal femoral anteversion (PFA) and distal femoral anteversion (DFA) had ICC values of 0.921 (95% CI 0.894–0.942) and 0.927 (95% CI 0.901–0.946), respectively. Total tibial torsion (TTT), proximal tibial torsion (PTT), and distal tibial torsion (DTT) showed ICC values of 0.940 (95% CI 0.918–0.955), 0.978 (95% CI 0.970–0.984), and 0.985 (95% CI 0.980–0.989), respectively.For 3D measurements, TFA and MFA had ICC values of 0.977 (95% CI 0.968–0.983) and 0.983 (95% CI 0.977–0.988), respectively, while PFA and DFA showed ICC values of 0.928 (95% CI 0.903–0.947) and 0.868 (95% CI 0.824–0.902). TTT, PTT, and DTT exhibited ICC values of 0.925 (95% CI 0.899–0.944), 0.971 (95% CI 0.961–0.979), and 0.981 (95% CI 0.974–0.986), respectively. Observer A’s consistency results for 2D and 3D measurements indicated an ICC of 0.984 (95% CI 0.978–0.988) for TFA and 0.981 (95% CI 0.974–0.986) for MFA, with PFA and DFA having ICC values of 0.925 (95% CI 0.899–0.944) and 0.897 (95% CI 0.862–0.923), respectively. TTT, PTT, and DTT demonstrated ICC values of 0.947 (95% CI 0.929–0.961), 0.969 (95% CI 0.958–0.977), and 0.983 (95% CI 0.976–0.987), respectively. Observer B showed similar consistency, with ICC values of 0.979 (95% CI 0.972–0.985) for TFA and 0.985 (95% CI 0.980–0.989) for MFA, and ICC values of 0.916 (95% CI 0.887–0.938) and 0.888 (95% CI 0.850–0.917) for PFA and DFA, respectively. TTT, PTT, and DTT exhibited ICC values of 0.936 (95% CI 0.914–0.953), 0.965 (95% CI 0.953–0.974), and 0.986 (95% CI 0.981–0.990), respectively. Paired t-tests showed no statistically significant differences between 2D and 3D measurements for all major parameters (P > 0.05). Our findings confirm that 3D measurement methods exhibit high reliability and agreement with the CT gold standard. Given their ability to provide comprehensive segmental torsion analysis with lower observer dependence, 3D methods have strong potential to improve clinical decision-making in PFPS treatment and surgical planning.https://doi.org/10.1038/s41598-025-01262-1KeywordsThree-dimensional reconstructionPatellofemoral pain syndromeSegmental torsionFemoral anteversionTibial torsion
spellingShingle Shouliang Xiong
Yunpeng Luo
Shikuan Zhang
Zhongxuan Wu
Degang Huang
Hongzhou Sun
Consistency of 3D reconstruction measurements of lower limb torsion at different segments with the CT gold standard
Scientific Reports
Keywords
Three-dimensional reconstruction
Patellofemoral pain syndrome
Segmental torsion
Femoral anteversion
Tibial torsion
title Consistency of 3D reconstruction measurements of lower limb torsion at different segments with the CT gold standard
title_full Consistency of 3D reconstruction measurements of lower limb torsion at different segments with the CT gold standard
title_fullStr Consistency of 3D reconstruction measurements of lower limb torsion at different segments with the CT gold standard
title_full_unstemmed Consistency of 3D reconstruction measurements of lower limb torsion at different segments with the CT gold standard
title_short Consistency of 3D reconstruction measurements of lower limb torsion at different segments with the CT gold standard
title_sort consistency of 3d reconstruction measurements of lower limb torsion at different segments with the ct gold standard
topic Keywords
Three-dimensional reconstruction
Patellofemoral pain syndrome
Segmental torsion
Femoral anteversion
Tibial torsion
url https://doi.org/10.1038/s41598-025-01262-1
work_keys_str_mv AT shouliangxiong consistencyof3dreconstructionmeasurementsoflowerlimbtorsionatdifferentsegmentswiththectgoldstandard
AT yunpengluo consistencyof3dreconstructionmeasurementsoflowerlimbtorsionatdifferentsegmentswiththectgoldstandard
AT shikuanzhang consistencyof3dreconstructionmeasurementsoflowerlimbtorsionatdifferentsegmentswiththectgoldstandard
AT zhongxuanwu consistencyof3dreconstructionmeasurementsoflowerlimbtorsionatdifferentsegmentswiththectgoldstandard
AT deganghuang consistencyof3dreconstructionmeasurementsoflowerlimbtorsionatdifferentsegmentswiththectgoldstandard
AT hongzhousun consistencyof3dreconstructionmeasurementsoflowerlimbtorsionatdifferentsegmentswiththectgoldstandard