Automatic Assessment of Lower-Limb Alignment from Computed Tomography

High variability exists in 3D leg alignment analysis, but underlying principles that might lead to agreement on a universal framework could be identified: A systematic review

PURPOSE

To (1) investigate the hypothesis that there is high variability in the reported methods to derive axes and joint orientations from three-dimensional (3D) bone models to (a) perform 3D knee-related leg alignment analysis and (b) define coordinate systems for the femur, tibia and leg and (2) identify underlying principles that might lead to agreement on a universal 3D leg alignment analysis framework.

METHODS

A systematic review of the literature between January 2006 and June 2024 was performed. Articles explicitly reporting methods to derive axes and joint orientations from CT-based 3D bone models for alignment parameters and/or coordinate systems of the femur, tibia and leg were included. Study characteristics and reported methods were extracted and presented as a qualitative synthesis.

RESULTS

A total of 93 studies were included. There was high variability in the reported methods to derive axes and joint orientations from 3D bone models. Nevertheless, the reported methods could be categorized into four groups, and several underlying principles of the four groups could be identified. Furthermore, the definitions of femoral and tibial coordinate systems were most frequently based on the mechanical axis (femoral, 13/19 [68%]; tibial, 13/26 [50%]) and a central medial-lateral axis (femoral, 16/19 [84%]; tibial, 12/26 [46%]); no leg coordinate system was reported. Interestingly, of the included studies that reported on leg alignment parameters (76/93, 82%), only a minority reported expressing these in a complete coordinate system (25/76, 33%).

CONCLUSION

There is high variability in 3D knee-related leg alignment analysis. Therefore, universal 3D reference values for alignment parameters cannot yet be defined, and comparison of alignment parameter values between different studies is impossible. However, several underlying principles to the reported methods were identified, which could serve to reach more agreement on a future universal 3D framework for leg alignment analysis.

LEVEL OF EVIDENCE

Level I.

Automatic assessment of lower limb deformities using high-resolution X-ray images

BACKGROUND

Planning an osteotomy or arthroplasty surgery on a lower limb requires prior classification/identification of its deformities. The detection of skeletal landmarks and the calculation of angles required to identify the deformities are traditionally done manually, with measurement accuracy relying considerably on the experience of the individual doing the measurements. We propose a novel, image pyramid-based approach to skeletal landmark detection.

METHODS

The proposed approach uses a Convolutional Neural Network (CNN) that receives the raw X-ray image as input and produces the coordinates of the landmarks. The landmark estimations are modified iteratively via the error feedback method to come closer to the target. Our clinically produced full-leg X-Rays dataset is made publically available and used to train and test the network. Angular quantities are calculated based on detected landmarks. Angles are then classified as lower than normal, normal or higher than normal according to predefined ranges for a normal condition.

RESULTS

The performance of our approach is evaluated at several levels: landmark coordinates accuracy, angles' measurement accuracy, and classification accuracy. The average absolute error (difference between automatically and manually determined coordinates) for landmarks was 0.79 ± 0.57 mm on test data, and the average absolute error (difference between automatically and manually calculated angles) for angles was 0.45 ± 0.42°.

CONCLUSIONS

Results from multiple case studies involving high-resolution images show that the proposed approach outperforms previous deep learning-based approaches in terms of accuracy and computational cost. It also enables the automatic detection of the lower limb misalignments in full-leg x-ray images.

A structured framework for standardized 3D leg alignment analysis: an international Delphi consensus study

PURPOSE

To reach consensus among international experts on a structured framework for standardized 3D leg alignment analysis based on 3D bone models, ensuring consistency and improving clinical applicability.

METHODS

A Delphi study was performed in four rounds. Rounds 1 and 2 involved a steering and rating group that developed statements based on principles preserving the 3D complexity of anatomical structures, identified through a systematic review. These statements encompassed approaches for deriving joint centres and joint orientations, and defining coordinate systems using 3D bone models. In Rounds 3 and 4, a panel of 35 international experts, including clinicians (54%) and engineers (46%), with participants from Europe (80%), Oceania (9%), Asia (6%), and the Americas (6%), evaluated these statements. Consensus was defined as ≥80% agreement.

RESULTS

Rounds 1 and 2 resulted in 31 statements to be included in the survey. Of these, 26 achieved consensus in Round 3, with the five remaining statements refined and reaching consensus in Round 4. Experts agreed on utilising all available relevant surface data to define joint centres, joint orientations, and individual femoral and tibial coordinate systems alongside a combined leg coordinate system, and adopting central 3D axes for femoral version and tibial torsion.

CONCLUSIONS

This international Delphi consensus study provides a structured framework for a standardized 3D leg alignment analysis based on 3D bone models. This framework aims to enhance clinical applicability for preoperative planning and execution of uni- and multiplanar correction osteotomies around the knee, reduce the methodological variability in 3D leg alignment analysis literature, and improve cross-study comparability.

LEVEL OF EVIDENCE

Level V.

Comparative impact of high tibial osteotomy and supramalleolar osteotomy on limb alignment and ankle function: a retrospective study

OBJECTIVE

This retrospective study aimed to conduct a comparative analysis of the impact of high tibial osteotomy (HTO) and supramalleolar osteotomy (SMOT) on lower limb alignment and ankle function after surgery.

METHODS

A cohort of patients who underwent either HTO (n = 63) or SMOT (n = 51) for lower limb alignment issues was included in the study. Inclusion criteria comprised individuals who underwent the surgical procedures between June 2018 and June 2021; exclusion criteria encompassed incomplete medical records and inadequate follow-up data. Baseline characteristics, weight-bearing line ratios, ankle joint function, and lower limb lines of force were evaluated before surgery, postoperatively, and at the 6-month follow-up. Statistical analyses were performed to compare the outcomes between the HTO and SMOT groups, as well as between non-deviated and deviated subgroups. Spearman rank correlation analysis was used to reveal correlations between variables.

RESULTS

The preoperative and immediate postoperative weight-bearing line ratios were similar between the HTO and SMOT groups. However, a notable difference emerged at the 6-month follow-up, suggesting distinct impacts of the two procedures on lower limb alignment. Additionally, the HTO group exhibited superior postoperative outcomes in ankle joint function, specifically in pain alleviation and functional improvement, compared to the SMOT group. The analysis of lower limb lines of force demonstrated a significant association between the surgical procedure and alterations in lower limb biomechanics, emphasizing the differential impact of HTO and SMOT. Furthermore, the comparison between non-deviated and deviated subgroups highlighted the potential impact of lower limb alignment on postoperative ankle function.

CONCLUSION

The findings contribute valuable insights into the comparative effectiveness of HTO and SMOT in addressing lower limb alignment and ankle function. This study's results have significant implications for orthopedic treatment and may guide treatment strategies for patients undergoing lower limb realignment surgery, ultimately enhancing the quality of life for affected individuals.

A deep learning-based multi-view approach to automatic 3D landmarking and deformity assessment of lower limb

Anatomical Landmark detection in CT-Scan images is widely used in the identification of skeletal disorders. However, the traditional process of manually detecting anatomical landmarks, especially in three dimensions, is both time-consuming and prone to human errors. We propose a novel, deep-learning-based approach to automatic detection of 3D landmarks in CT images of the lower limb. We generate multiple view renderings of the scanned limb and then integrate them, using a pyramid-style convolutional neural network, to build a 3D model of the bone and to determine the spatial coordinates of the landmarks. Those landmarks are then used to calculate key anatomical indicators that would enable the reliable diagnosis of skeletal disorders. To evaluate the performance of the proposed approach we compare its predicted landmark coordinates and resulting anatomical indicators (both 2D and 3D) with those determined by human experts. The average coordinate error (difference between automatically and manually determined coordinates) of the landmarks was 2.05 ± 1.36 mm on test data, whereas the average angular error (difference between automatically and manually calculated angles in three and two dimensions) on the same dataset was 0.53 ± 0.66° and 0.74 ± 0.87°, respectively. Our proposed deep-learning-based approach not only outperforms the traditional landmark detection and indicator assessment methods in terms of speed and accuracy but also improves the credibility of the ensuing diagnoses by avoiding manual landmarking errors.

Radiological Approach to Assessment of Lower-Limb Alignment-Coronal and Transverse Plane Analysis

Lower-limb alignment deformities constitute a significant clinical concern, as they can lead to serious complications, including progressive degenerative diseases and disabilities. Rotational deformities may give rise to conditions such as joint arthrosis, patellar instability, and the degeneration of the patellofemoral cartilage. Therefore, a comprehensive evaluation of lower-limb alignment is essential for the effective patient management, preoperative planning, and successful correction of these deformities. The primary assessment method employs full-length standing radiographs in the anteroposterior (AP) projection, which facilitates accurate measurements of the anatomical and mechanical axes of the lower limb, including angles and deviations. The outcomes of this analysis are vital for the meticulous planning of osteotomy and total knee arthroplasty (TKA). In addition, computed tomography (CT) provides a specialized approach for the precise evaluation of femoral and tibial rotation. In this area, there are potential opportunities for the implementation of AI-based automated measurement systems.

Automatic identification of radius and ulna bone landmarks on 3D virtual models

BACKGROUND

For bone morphology and biomechanics analysis, landmarks are essential to define position, orientation, and shape. These landmarks define bone and joint coordinate systems and are widely used in these research fields. Currently, no method is known for automatically identifying landmarks on virtual 3D bone models of the radius and ulna. This paper proposes a knowledge-based method for locating landmarks and calculating a coordinate system for the radius, ulna, and combined forearm bones, which is essential for measuring forearm function. This method does not rely on pre-labeled data.

VALIDATION

The algorithm is validated by comparing the landmarks placed by the algorithm with the mean position of landmarks placed by a group of experts on cadaveric specimens regarding distance and orientation.

RESULTS

The median Euclidean distance differences between all the automated and reference landmarks range from 0.4 to 1.8 millimeters. The median angular differences of the coordinate system of the radius and ulna range from -1.4 to 0.6 degrees. The forearm coordinate system's median errors range from -0.2 to 2.0 degrees. The median error in calculating the rotational position of the radius relative to the ulna is 1.8 degrees.

CONCLUSION

The automatic method's applicability depends on the use context and desired accuracy. However, the current method is a validated first step in the automatic analysis of the three-dimensional forearm anatomy.

Who Are the Anatomic Outliers Undergoing Total Knee Arthroplasty? A Computed Tomography-Based Analysis of the Hip-Knee-Ankle Axis Across 1,352 Preoperative Computed Tomographies Using a Deep Learning and Computer Vision-Based Pipeline

BACKGROUND

Dissatisfaction after total knee arthroplasty (TKA) ranges from 15 to 30%. While patient selection may be partially responsible, morphological and reconstructive challenges may be determinants. Preoperative computed tomography (CT) scans for TKA planning allow us to evaluate the hip-knee-ankle axis and establish a baseline phenotypic distribution across anatomic parameters. The purpose of this cross-sectional analysis was to establish the distributions of 27 parameters in a pre-TKA cohort and perform threshold analysis to identify anatomic outliers.

METHODS

There were 1,352 pre-TKA CTs that were processed. A 2-step deep learning pipeline of classification and segmentation models identified landmark images and then generated contour representations. We used an open-source computer vision library to compute measurements for 27 anatomic metrics along the hip-knee axis. Normative distribution plots were established, and thresholds for the 15th percentile at both extremes were calculated. Metrics falling outside the central 70th percentile were considered outlier indices. A threshold analysis of outlier indices against the proportion of the cohort was performed.

RESULTS

Significant variation exists in pre-TKA anatomy across 27 normally distributed metrics. Threshold analysis revealed a sigmoid function with a critical point at 9 outlier indices, representing 31.2% of subjects as anatomic outliers. Metrics with the greatest variation related to deformity (tibiofemoral angle, medial proximal tibial angle, lateral distal femoral angle), bony size (tibial width, anteroposterior femoral size, femoral head size, medial femoral condyle size), intraoperative landmarks (posterior tibial slope, transepicondylar and posterior condylar axes), and neglected rotational considerations (acetabular and femoral version, femoral torsion).

CONCLUSIONS

In the largest non-industry database of pre-TKA CTs using a fully automated 3-stage deep learning and computer vision-based pipeline, marked anatomic variation exists. In the pursuit of understanding the dissatisfaction rate after TKA, acknowledging that 31% of patients represent anatomic outliers may help us better achieve anatomically personalized TKA, with or without adjunctive technology.

Automating Linear and Angular Measurements for the Hip and Knee After Computed Tomography: Validation of a Three-Stage Deep Learning and Computer Vision-Based Pipeline for Pathoanatomic Assessment

Background

Variability in the bony morphology of pathologic hips/knees is a challenge in automating preoperative computed tomography (CT) scan measurements. With the increasing prevalence of CT for advanced preoperative planning, processing this data represents a critical bottleneck in presurgical planning, research, and development. The purpose of this study was to demonstrate a reproducible and scalable methodology for analyzing CT-based anatomy to process hip and knee anatomy for perioperative planning and execution.

Methods

One hundred patients with preoperative CT scans undergoing total knee arthroplasty for osteoarthritis were processed. A two-step deep learning pipeline of classification and segmentation models was developed that identifies landmark images and then generates contour representations. We utilized an open-source computer vision library to compute measurements. Classification models were assessed by accuracy, precision, and recall. Segmentation models were evaluated using dice and mean Intersection over Union (IOU) metrics. Contour measurements were compared against manual measurements to validate posterior condylar axis angle, sulcus angle, trochlear groove-tibial tuberosity distance, acetabular anteversion, and femoral version.

Results

Classifiers identified landmark images with accuracy of 0.91 and 0.88 for hip and knee models, respectively. Segmentation models demonstrated mean IOU scores above 0.95 with the highest dice coefficient of 0.957 [0.954-0.961] (UNet3+) and the highest mean IOU of 0.965 [0.961-0.969] (Attention U-Net). There were no statistically significant differences for the measurements taken automatically vs manually (P > 0.05). Average time for the pipeline to preprocess (48.65 +/- 4.41 sec), classify/retrieve landmark images (8.36 +/- 3.40 sec), segment images (<1 sec), and obtain measurements was 2.58 (+/- 1.92) minutes.

Conclusions

A fully automated three-stage deep learning and computer vision-based pipeline of classification and segmentation models accurately localized, segmented, and measured landmark hip and knee images for patients undergoing total knee arthroplasty. Incorporation of clinical parameters, like patient-reported outcome measures and instability risk, will be important considerations alongside anatomic parameters.

Automated Landmark Annotation for Morphometric Analysis of Distal Femur and Proximal Tibia

Manual anatomical landmarking for morphometric knee bone characterization in orthopedics is highly time-consuming and shows high operator variability. Therefore, automation could be a substantial improvement for diagnostics and personalized treatments relying on landmark-based methods. Applications include implant sizing and planning, meniscal allograft sizing, and morphological risk factor assessment. For twenty MRI-based 3D bone and cartilage models, anatomical landmarks were manually applied by three experts, and morphometric measurements for 3D characterization of the distal femur and proximal tibia were calculated from all observations. One expert performed the landmark annotations three times. Intra- and inter-observer variations were assessed for landmark position and measurements. The mean of the three expert annotations served as the ground truth. Next, automated landmark annotation was performed by elastic deformation of a template shape, followed by landmark optimization at extreme positions (highest/lowest/most medial/lateral point). The results of our automated annotation method were compared with ground truth, and percentages of landmarks and measurements adhering to different tolerances were calculated. Reliability was evaluated by the intraclass correlation coefficient (ICC). For the manual annotations, the inter-observer absolute difference was 1.53 ± 1.22 mm (mean ± SD) for the landmark positions and 0.56 ± 0.55 mm (mean ± SD) for the morphometric measurements. Automated versus manual landmark extraction differed by an average of 2.05 mm. The automated measurements demonstrated an absolute difference of 0.78 ± 0.60 mm (mean ± SD) from their manual counterparts. Overall, 92% of the automated landmarks were within 4 mm of the expert mean position, and 95% of all morphometric measurements were within 2 mm of the expert mean measurements. The ICC (manual versus automated) for automated morphometric measurements was between 0.926 and 1. Manual annotations required on average 18 min of operator interaction time, while automated annotations only needed 7 min of operator-independent computing time. Considering the time consumption and variability among observers, there is a clear need for a more efficient, standardized, and operator-independent algorithm. Our automated method demonstrated excellent accuracy and reliability for landmark positioning and morphometric measurements. Above all, this automated method will lead to a faster, scalable, and operator-independent morphometric analysis of the knee.