Predicting the mechanical hip–knee–ankle angle accurately from standard knee radiographs: a cross-validation experiment in 100 patients

Gielis, Willem Paul; Rayegan, H.; Arbabi, Vahid; Brooghani, Seyed Yousef Ahmadi; Lindner, Claudia; Cootes, T. F.; Jong, Pim A. de; Weinans, Harrie; Custers, Roel J.H.

doi:10.1080/17453674.2020.1779516

Cited by 16 publications

(20 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This is the rst study to our knowledge to present a fully automatic method for measuring coronal knee alignment using knee-only PA radiographs, demonstrating high accuracy for FTA (MAE: 0.8°) and outperforming state-of-the-art 11 for HKA (MAE: 1.7°). This nding is particularly impressive considering no restrictions were imposed on the training dataset and no anatomical annotation was required.…”

Section: Discussionmentioning

confidence: 93%

“…Whilst HKA is the better measure, many centres choose to use only FTA for preop planning or postop follow-up as it is cheaper and more convenient. Enabling HKA measurement from knee-only radiographs would thus be of great clinical use: 11 producing the gold-standard measure with reduced cost and radiation exposure, providing additional information for knee arthroplasty planning and follow-up for surgeons who do not routinely have access to full-limb radiographs.…”

Section: Introductionmentioning

confidence: 99%

“…State-of-the-art for prediction of HKA from knee-only radiographs is based off algorithmic approaches that relate to a clinical understanding of knee anatomy. [10][11][12] An alternative approach could be to leverage recent developments in convolutional neural networks and growth in computing power. 13 Deep learning methods have shown early promise in measuring HKA on full-limb radiographs [14][15][16] and in analysing PA knee radiographs for other applications.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Predicting Hip-Knee-Ankle and Femorotibial Angles from Knee Radiographs with Deep Learning

Arkel

Wang

Hall

et al. 2022

Preprint

View full text Add to dashboard Cite

Knee alignment affects the development and surgical treatment of knee osteoarthritis. This research aimed to improve its automatic measurement from posteroanterior (PA) knee radiographs by using convolutional neural networks. Radiographs from the Osteoarthritis Initiative with corresponding femorotibial angle (FTA, N = 6149) and hip-knee-ankle angle (HKA, N = 2351) measurements, were split into training, validation, and test datasets in a 70:15:15 ratio. Separate models were developed for the prediction of FTA and HKA and their accuracy quantified. Heat maps were then used to identify the anatomical features within each image that most contributed to the predicted angles. High accuracy was achieved for FTA (mean absolute error 0.8°), and better than state-of-the-art accuracy was achieved for HKA (mean absolute error 1.7°). Heat maps for both models were concentrated on the knee anatomy and could prove a valuable tool for assessing prediction reliability in clinical application. Deep learning techniques enable fast, reliable and accurate predictions of both FTA and HKA from plain knee radiographs and could lead to cost savings for healthcare providers and reduced radiation exposure for patients.

show abstract

Section: Discussionmentioning

confidence: 93%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Predicting Hip-Knee-Ankle and Femorotibial Angles from Knee Radiographs with Deep Learning

Arkel

Wang

Hall

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…They also showed the agreement between the FTA on SKR and WLR in the WB position was greater than that in the NWB position (ICC 0.878 vs. 0.657). Additionally, the FTA was often measured on SKR using lines drawn from the knee to a point approximately 10 cm proximal to the center of the knee (17); however, the location the point for the bone shaft also might alter the correlation of FTA to HKA. Sheey's et al (11) compared different versions of FTA and HKA using different reference points and concluded that using shorter FTA shaft lengths to evaluate HKA modestly weakened the correlation of the 2 measurements.…”

Section: Discussionmentioning

confidence: 99%

“…FTA SKR was defined as the intersecting angle of the femoral and tibial anatomical axis on SKR. The measurements were performed using a general technique and validated by Gielis et al (17). To have a better reflection of femoral and tibial anatomical variations (18), we selected the more distant reference points of the femur and tibia (those which were closest to the edge of the image).…”

Section: Alignment Measurementsmentioning

confidence: 99%

Non-weight-bearing short knee radiographs to evaluate coronal alignment before total knee arthroplasty

Pan¹,

Huang²,

Zhang³

et al. 2022

Quant Imaging Med Surg

View full text Add to dashboard Cite

Background: Whole leg radiograph (WLR) is the gold standard in assessing lower limb alignment before total knee arthroplasty (TKA) although in practice, non-weight-bearing short knee radiographs (NWB SKRs) are used by most medical institutions. The objective of this study was to determine whether the femorotibial angle (FTA) could be used to evaluate lower limb alignment on limited NWB SKRs. We also investigated whether FTA alignment measurements on NWB SKRs and WLRs differed depending on the direction of knee deformity.Methods: In all, 105 knees which underwent both NWB SKR and WLR were included. Measurement of hip-knee-ankle angle (HKA) was obtained through WLR, while the FTA was found using NWB SKR (FTA SKR ) and WLR (FTA WLR ). All knees were divided into three groups based on the HKA. The Kappa statistic was used to compare the agreement of categorical alignment variables between the HKA and FTA SKR . The agreement of the measurements obtained from the two radiographs was made using Bland-Altman plots and intraclass correlation coefficient (ICC). Pearson correlation coefficient and simple linear regression analysis were also conducted to evaluate the correlation between different angles. Results:The agreement for categories of lower limb deformity was rated excellent (kappa =0.804, P<0.001). The Bland-Altman plot showed that the mean difference for the FTA SKR and FTA WLR was 4.4°.There was an excellent correlation (r=0.861, P<0.001) and good reliability (ICC, 0.607) between the FTA SKR and HKA. For the varus group, there was a good correlation between the FTA SKR and HKA (r=0.650, P<0.001); however, there were no significant correlations between the FTAS KR and HKA in the neutral (r=0.106, P=0.543) and valgus groups (r=0.322, P=0.102).Conclusions: For outpatient follow-up, we found that the FTA on NWB SKRs is an acceptable means for classifying knee alignment (varus, neutral, or valgus). The measurement on NWB SKR also showed excellent correlation and good agreement with the HKA. For varus knees, NWB SKR measurements showed the best correlation with the HKA. However, for neutral and valgus knees, NWB SKR measurements were insufficient for conducting a reliable calculation and quantification of coronal alignment of the lower limb.

show abstract

Femorotibial angle scan–rescan reproducibility: A high‐precision calculation on a large cohort

Hall,

Jones,

van Arkel

2024

Knee surg. sports traumatol. arthrosc.

View full text Add to dashboard Cite

PurposeFemorotibial angle (FTA) is a convenient measure of coronal knee alignment that can be extracted from a short knee radiograph, avoiding the additional radiation exposure and specialist equipment required for full‐leg radiographs. While intra‐ and inter‐reader reproducibility from the same image has been reported, the full scan–rescan reproducibility across images, as calculated in this study, has not.MethodsIn this study, 4589 FTA measurement pairs from 2586 subjects acquired a year apart were used to estimate FTA scan–rescan reproducibility using data from the Osteoarthritis Initiative. Subjects with radiographic progression of osteoarthritis or other conditions that may cause a change in coronal knee alignment were excluded. Measurement pairs were analysed using paired‐samples tests to detect differences and compared to symptomatic changes in Western Ontario and McMaster Universities Arthritis Index scores for joint pain, stiffness and physical function to detect correlations.ResultsThe 95% limit of agreement and the paired‐samples correlation were calculated with high precision to be [−1.76°, +1.78°] and 0.938, considerably worse than the corresponding figures for intra‐ and inter‐reader reproducibility, without relation to symptomatic or radiographic changes in knee condition. This error will weakly attenuate and values from their true values in correlative studies involving FTA. The realistic maximum value for is 87% and for Pearson's is 93%.ConclusionThe scan–rescan reproducibility in FTA is almost double the intra‐ and inter‐reader reliability from a single scan. At almost ±2° accuracy, FTA is inappropriate for surgical use, but it is sufficiently reproducible to produce good correlations in studies predicting disease incidence and progression.Level of EvidenceLevel II, retrospective study.

show abstract

Predicting the mechanical hip–knee–ankle angle accurately from standard knee radiographs: a cross-validation experiment in 100 patients

Cited by 16 publications

References 25 publications

Predicting Hip-Knee-Ankle and Femorotibial Angles from Knee Radiographs with Deep Learning

Predicting Hip-Knee-Ankle and Femorotibial Angles from Knee Radiographs with Deep Learning

Non-weight-bearing short knee radiographs to evaluate coronal alignment before total knee arthroplasty

Femorotibial angle scan–rescan reproducibility: A high‐precision calculation on a large cohort

Contact Info

Product

Resources

About