Three-Dimensional Magnetic Resonance Imaging Bone Models of the Hip Joint Using Deep Learning: Dynamic Simulation of Hip Impingement for Diagnosis of Intra- and Extra-articular Hip Impingement

Femoroacetabular Impingement: Preoperative Planning and Postoperative MR Imaging Evaluation

Imaging plays a critical role in the preoperative and postoperative evaluation of patients with femoroacetabular impingement. Non-contrast MR imaging and direct magnetic resonance arthrography of the hip are the modalities of choice for the preoperative assessment of chondrolabral lesions. The MRI protocol should include radial images for detailed analysis of the cam deformity and fast sequences covering the pelvis and knee for measurement of femoral torsion. In patients with postoperative pain, a comprehensive analysis of residual deformities and signs of osseous overcorrection should be performed as they can cause residual impingement or iatrogenic instability.

Validation of a computerized model for a new biomechanical concept- the fossa-foveolar mismatch- the answer to lesions of the ligamentous fossa-foveolar complex in the hip?

BACKGROUND

Hip-preserving surgery in young patients frequently reveals lesions of the ligamentum teres (LT). Histological and clinical evidence supports that those lesions could be source of intraarticular hip pain. It has been hypothesized that LT degeneration could be linked to the abnormal positioning of the fovea outside the lunate surface during various daily motions. We introduce the "fossa-foveolar mismatch" (FFM) by determining the trajectory of the fovea in the fossa during hip motions, enabling a comparison across diverse hip-pathomorphologies.

AIMS

to determine (1) intraobserver reliability and (2) interobserver reproducibility of our computer-assisted 3-dimensional (3D) model of the FFM.

MATERIALS AND METHODS

All patients with joint preserving surgery for femoroacetabular impingement syndrome (FAIS) or developmental dysplasia of the hip (DDH) at our institution (11. 2015-08.2019)were initially eligible. We employed a simple random sampling technique to select 15 patients for analysis. Three-dimensional surface models based on preoperative computed tomography (CT) scans were built, the fossa virtually excised, the fovea capitis marked. Models were subjected to physiological range of motion with validated 3D collision detection software. Using a standardized medial view on the resected fossa and the transparent lunate surface, the FFM-index was calculated for 17 motions. It was obtained by dividing the surface occupied by the fovea outside of the fossa by the total foveolar tracking surface. Three observers independently performed all analyses twice. (1) Intraobserver reliability and (2) interobserver reproducibility were calculated using intraclass correlation coefficients (ICCs).

RESULTS

(1) We obtained excellent intraobserver ICCs for the FFM-index averaging 0.92 with 95% CI 0.77-0.9 among the three raters for all motions. (2) Interobserver reproducibility between raters was good to excellent, ranging from 0.76 to 0.98.

CONCLUSIONS

The FFM-index showed excellent intraobserver reliability and interobserver reproducibility for all motions. This innovative approach deepens our understanding of biomechanical implications, providing valuable insights for identifying patient populations at risk.

MR-based Bony 3D models enable radiation-free preoperative patient-specific analysis and 3D printing for SCFE patients

Objectives

Slipped capital femoral epiphyses (SCFE) is a common pediatric hip disease with the risk of osteoarthritis and impingement deformities, and 3D models could be useful for patient-specific analysis. Therefore, magnetic resonance imaging (MRI) bone segmentation and feasibility of 3D printing and of 3D ROM simulation using MRI-based 3D models were investigated.

Methods

A retrospective study involving 22 symptomatic patients (22 hips) with SCFE was performed. All patients underwent preoperative hip MR with pelvic coronal high-resolution images (T1 images). Slice thickness was 0.8-1.2 mm. Mean age was 12 ± 2 years (59% male patients). All patients underwent surgical treatment. Semi-automatic MRI-based bone segmentation with manual corrections and 3D printing of plastic 3D models was performed. Virtual 3D models were tested for computer-assisted 3D ROM simulation of patients with knee images and were compared to asymptomatic contralateral hips with unilateral SCFE (15 hips, control group).

Results

MRI-based bone segmentation was feasible (all patients, 100%, in 4.5 h, mean 272 ± 52 min). Three-dimensional printing of plastic 3D models was feasible (all patients, 100%) and was considered helpful for deformity analysis by the treating surgeons for severe and moderate SCFE. Three-dimensional ROM simulation showed significantly (p < 0.001) decreased flexion (48 ± 40°) and IR in 90° of flexion (-14 ± 21°, IRF-90°) for severe SCFE patients with MRI compared to control group (122 ± 9° and 36 ± 11°). Slip angle improved significantly (p < 0.001) from preoperative 54 ± 15° to postoperative 4 ± 2°.

Conclusion

MRI-based 3D models were feasible for SCFE patients. Three-dimensional models could be useful for severe SCFE patients for preoperative 3D printing and deformity analysis and for ROM simulation. This could aid for patient-specific diagnosis, treatment decisions, and preoperative planning. MRI-based 3D models are radiation-free and could be used instead of CT-based 3D models in the future.

Femoroacetabular Impingement Measurements Obtained From Two-Dimensional Radiographs Versus Three-Dimensional-Reconstructed Computed Tomography Images Result in Different Values

Purpose

To compare the reliability and accuracy of radiographic measurements obtained from 2-dimensional (2D) radiographs and 3-dimensional (3D)-reconstructed computed tomography (CT) images in the assessment of femoroacetabular impingement syndrome (FAIS).

Methods

Consecutive patients with FAIS from January 2018 to December 2020 were identified and included in this study. Two fellowship-trained surgeons and 2 fellows performed blinded radiographic measurements. Lateral center-edge angle (LCEA) and Tönnis angles were measured on anteroposterior pelvic radiographs, and alpha angles were measured on frog lateral radiographs. Reliability coefficients for individual measurement accuracy were performed using the Cronbach alpha and intra- and inter-rater intraclass correlation coefficients (ICCs). Composite measurements for LCEA, Tönnis angle, and alpha angle were compared with the corresponding 3D value using paired sample t-tests.

Results

Fifty-three patients with FAIS with standardized 2D radiographic and 3D-reconstructed CT imaging were included. All reliability metrics met thresholds for internal reliability. Inter-rater ICCs for LCEA, Tönnis angle, and alpha angle were (0.928, 0.888, 0.857, all P < .001). When we compared 2D radiographic measurements with 3D-reconstructed CT values, there was a significant difference in the LCEA for 2 authors: surgeon 1 (mean [M] = -9.14, standard deviation [SD] = 5.7); t(52) = -11.6, P < .001, and surgeon 2 (M = -5.9°, SD = 4.7); t(52) = -9.2, P < .001. Significant differences were seen for Tönnis angle for 2 authors: fellow 2 (M = 3.9°, SD = 5.6); t(52) = 5.1, P < .001, and surgeon 2 (M = -2.6°, SD = 4.1); t(52) = -4.6, P < .001. Alpha angle measurements compared to the 3D-reconstructed alpha angle at 2 o'clock was significantly different for 3 authors: fellow 1 (M = 11.9°, SD = 16.2); t(52) = 5.3, P < .001; fellow 2 (M = 10.4°, SD = 18.6); t(52) = 4.1, P = .002; and surgeon 2 (M = -6.5°, SD = 16.2); t(52) = -2.9, P = .005. Positive mean values indicate 2D radiographic measurements overestimated 3D reconstruction values and negative mean values indicate underestimation.

Conclusions

The use of 2D radiographs alone for preoperative planning of FAIS may lead to inaccuracies in radiographic measurements.

Level of Evidence

Level, III retrospective cohort study.

The impact of data augmentation and transfer learning on the performance of deep learning models for the segmentation of the hip on 3D magnetic resonance images

Different pathologies of the hip are characterized by the abnormal shape of the bony structures of the joint, namely the femur and the acetabulum. Three-dimensional (3D) models of the hip can be used for diagnosis, biomechanical simulation, and planning of surgical treatments. These models can be generated by building 3D surfaces of the joint's structures segmented on magnetic resonance (MR) images. Deep learning can avoid time-consuming manual segmentations, but its performance depends on the amount and quality of the available training data. Data augmentation and transfer learning are two approaches used when there is only a limited number of datasets. In particular, data augmentation can be used to artificially increase the size and diversity of the training datasets, whereas transfer learning can be used to build the desired model on top of a model previously trained with similar data. This study investigates the effect of data augmentation and transfer learning on the performance of deep learning for the automatic segmentation of the femur and acetabulum on 3D MR images of patients diagnosed with femoroacetabular impingement. Transfer learning was applied starting from a model trained for the segmentation of the bony structures of the shoulder joint, which bears some resemblance to the hip joint. Our results suggest that data augmentation is more effective than transfer learning, yielding a Dice similarity coefficient compared to ground-truth manual segmentations of 0.84 and 0.89 for the acetabulum and femur, respectively, whereas the Dice coefficient was 0.78 and 0.88 for the model based on transfer learning. The Accuracy for the two anatomical regions was 0.95 and 0.97 when using data augmentation, and 0.87 and 0.96 when using transfer learning. Data augmentation can improve the performance of deep learning models by increasing the diversity of the training dataset and making the models more robust to noise and variations in image quality. The proposed segmentation model could be combined with radiomic analysis for the automatic evaluation of hip pathologies.

Open-source navigation system for tracking dissociated parts with multi-registration

PURPOSE

During reconstructive surgery, knee and hip replacements, and orthognathic surgery, small misalignments in the pose of prosthesis and bones can lead to severe complications. Hence, the translational and angular accuracies are critical. However, traditional image-based surgical navigation lacks orientation data between structures, and imageless systems are unsuitable for cases of deformed anatomy. We introduce an open-source navigation system using a multiple registration approach that can track instruments, implants, and bones to precisely guide the surgeon in emulating a preoperative plan.

METHODS

We derived the analytical error of our method and designed a set of phantom experiments to measure its precision and accuracy. Additionally, we trained two classification models to predict the system reliability from fiducial points and surface matching registration data. Finally, to demonstrate the procedure feasibility, we conducted a complete workflow for a real clinical case of a patient with fibrous dysplasia and anatomical misalignment of the right femur using plastic bones.

RESULTS

The system is able to track the dissociated fragments of the clinical case and average alignment errors in the anatomical phantoms of [Formula: see text]  mm and [Formula: see text]. While the fiducial-points registration showed satisfactory results given enough points and covered volume, we acknowledge that the surface refinement step is mandatory when attempting surface matching registrations.

CONCLUSION

We believe that our device could bring significant advantages for the personalized treatment of complex surgical cases and that its multi-registration attribute is convenient for intraoperative registration loosening cases.

Deep learning-based workflow for hip joint morphometric parameter measurement from CT images

Objective.Precise hip joint morphometry measurement from CT images is crucial for successful preoperative arthroplasty planning and biomechanical simulations. Although deep learning approaches have been applied to clinical bone surgery planning, there is still a lack of relevant research on quantifying hip joint morphometric parameters from CT images.Approach.This paper proposes a deep learning workflow for CT-based hip morphometry measurement. For the first step, a coarse-to-fine deep learning model is designed for accurate reconstruction of the hip geometry (3D bone models and key landmark points). Based on the geometric models, a robust measurement method is developed to calculate a full set of morphometric parameters, including the acetabular anteversion and inclination, the femoral neck shaft angle and the inclination, etc. Our methods were validated on two datasets with different imaging protocol parameters and further compared with the conventional 2D x-ray-based measurement method.Main results. The proposed method yields high bone segmentation accuracies (Dice coefficients of 98.18% and 97.85%, respectively) and low landmark prediction errors (1.55 mm and 1.65 mm) on both datasets. The automated measurements agree well with the radiologists' manual measurements (Pearson correlation coefficients between 0.47 and 0.99 and intraclass correlation coefficients between 0.46 and 0.98). This method provides more accurate measurements than the conventional 2D x-ray-based measurement method, reducing the error of acetabular cup size from over 2 mm to less than 1 mm. Moreover, our morphometry measurement method is robust against the error of the previous bone segmentation step. As we tested different deep learning methods for the prerequisite bone segmentation, our method produced consistent final measurement results, with only a 0.37 mm maximum inter-method difference in the cup size.Significance. This study proposes a deep learning approach with improved robustness and accuracy for pelvis arthroplasty planning.

Posterior Hip Impingement at Maximal Hip Extension in Female Patients With Increased Femoral Version or Increased McKibbin Index and Its Effect on Sports Performance

Background

The location of posterior hip impingement at maximal extension in patients with posterior femoroacetabular impingement (FAI) is unclear.

Purpose

To investigate the frequency and area of impingement at maximal hip extension and at 10° and 20° of extension in female patients with increased femoral version (FV) and posterior hip pain.

Study Design

Cross-sectional study; Level of evidence, 3.

Methods

Osseous patient-specific 3-dimensional (3D) models were generated of 50 hips (37 female patients, 3D computed tomography) with a positive posterior impingement test and increased FV (defined as >35°). The McKibbin index (combined version) was calculated as the sum of FV and acetabular version (AV). Subgroups of patients with an increased McKibbin index >70° (24 hips) and FV >50° (20 hips) were analyzed. A control group of female participants (10 hips) had normal FV, normal AV, and no valgus deformity (neck-shaft angle, <139°). Validated 3D collision detection software was used for simulation of osseous impingement-free hip extension (no rotation).

Results

The mean impingement-free maximal hip extension was significantly lower in patients with FV >35° compared with the control group (15° ± 15° vs 55° ± 19°; P < .001). At maximal hip extension, 78% of patients with FV >35° had osseous posterior extra-articular ischiofemoral hip impingement. At 20° of extension, the frequency of posterior extra-articular ischiofemoral impingement was significantly higher for patients with a McKibbin index >70° (83%) and for patients with FV >35° (76%) than for controls (0%) (P < .001 for both). There was a significant correlation between maximal extension (no rotation) and FV (r = 0.46; P < .001) as well as between impingement area at 20° of extension (external rotation [ER], 0°) and McKibbin index (0.61; P < .001). Impingement area at 20° of extension (ER, 0°) was significantly larger for patients with McKibbin index >70° versus <70° (251 vs 44 mm²; P = .001).

Conclusion

The limited hip extension found in our study could theoretically affect the performance of sports activities such as running, ballet dancing, or lunges. Therefore, although not examined directly in this study, these activities are not advisable for these patients. Preoperative evaluation of FV and the McKibbin index is important in female patients with posterior hip pain before hip preservation surgery (eg, hip arthroscopy).

[Preoperative MR imaging for hip dysplasia : Assessment of associated deformities and intraarticular pathologies]

BACKGROUND

Developmental dysplasia of the hip (DDH) is a known reason for hip pain for adolescents and young adults. Preoperative imaging is increasingly recognized as an important factor due to the recent advances in MR imaging.

OBJECTIVES

The aim of this article is to give an overview of preoperative imaging for DDH. The acetabular version and morphology, associated femoral deformities (cam deformity, valgus and femoral antetorsion) and intraarticular pathologies (labrum and cartilage damage) and cartilage mapping are described.

METHODS

After an initial evaluation with AP radiographs, CT or MRI represent the methods of choice for the preoperative evaluation of the acetabular morphology and cam deformity, and for the measurement of femoral torsion. Different measurement techniques and normal values should be considered, especially for patients with increased femoral antetorsion because this could lead to misinterpretation and misdiagnosis. MRI allows analysis of labrum hypertrophy and subtle signs for hip instability. 3D MRI for cartilage mapping allows quantification of biochemical cartilage degeneration and yields great potential for surgical decision-making. 3D-CT and, increasingly, 3D MRI of the hip to generate 3D pelvic bone models and subsequent 3D impingement simulation can help to detect posterior extraarticular ischiofemoral impingement.

RESULTS AND DISCUSSION

Acetabular morphology can be divided in anterior, lateral and posterior hip dysplasia. Combined osseous deformities are common, such as hip dysplasia combined with cam deformity (86%). Valgus deformities were reported in 44%. Combined hip dysplasia and increased femoral antetorsion can occur in 52%. Posterior extraarticular ischiofemoral impingement between the lesser trochanter and the ischial tuberosity can occur in patients with increased femoral antetorsion. Typically, labrum damage and hypertrophy, cartilage damage, subchondral cysts can occur in hip dysplasia. Hypertrophy of the muscle iliocapsularis is a sign for hip instability. Acetabular morphology and femoral deformities (cam deformity and femoral anteversion) should be evaluated before surgical therapy for patients with hip dysplasia, considering the different measurement techniques and normal values of femoral antetorsion.

Limited External Rotation and Hip Extension Due to Posterior Extra-articular Ischiofemoral Hip Impingement in Female Patients With Increased Femoral Anteversion: Implications for Sports, Sexual, and Daily Activities

BACKGROUND

Posterior femoroacetabular impingement (FAI) is poorly understood. Patients with increased femoral anteversion (FV) exhibit posterior hip pain.

PURPOSE

To correlate hip impingement area with FV and with combined version and to investigate frequency of limited external rotation (ER) and hip extension (<40°, <20°, and <0°) due to posterior extra-articular ischiofemoral impingement.

STUDY DESIGN

Cross-sectional study; Level of evidence, 3.

METHODS

Osseous patient-specific three-dimensional (3D) models based on 3D computed tomography scans were generated of 37 female patients (50 hips) with positive posterior impingement test (100%) and increased FV >35° (Murphy method). Surgery was performed in 50% of patients (mean age, 30 years; 100% female). FV and acetabular version (AV) were added to calculate combined version. Subgroups of patients (24 hips) with increased combined version >70° and patients (9 valgus hips) with increased combined version >50° were analyzed. The control group (20 hips) had normal FV, normal AV, and no valgus. Bone segmentation was performed to generate 3D models of every patient. Validated 3D collision detection software was used for simulation of impingement-free hip motion (equidistant method). Impingement area was evaluated in combined 20° of ER and 20° of extension.

RESULTS

Posterior extra-articular ischiofemoral impingement occurred between the ischium and the lesser trochanter in 92% of patients with FV >35° in combined 20° of ER and 20° of extension. Impingement area in combined 20° of ER and 20° of extension was larger with increasing FV and with higher combined version; correlation was significant (P < .001, r = 0.57, and r = 0.65). Impingement area was significantly (P = .001) larger (681 vs 296 mm2) for patients with combined version >70° (vs <70°, respectively) in combined 20° of ER and 20° of extension. All symptomatic patients with increased FV >35° (100%) had limited ER <40°, and most (88%) had limited extension <40°. The frequency of posterior intra- and extra-articular hip impingement of symptomatic patients (100% and 88%, respectively) was significantly (P < .001) higher compared with the control group (10% and 10%, respectively). The frequency of patients with increased FV >35° with limited extension <20° (70%) and patients with limited ER <20° (54%) was significantly (P < .001) higher compared with the control group (0% and 0%, respectively). The frequency of completely limited extension <0° (no extension) and ER <0° (no ER in extension) was significantly (P < .001) higher for valgus hips (44%) with combined version >50° compared with patients with FV >35° (0%).

CONCLUSION

All patients with increased FV >35° had limited ER <40°, and most of them had limited extension <20° due to posterior intra- or extra-articular hip impingement. This is important for patient counselling, for physical therapy, and for planning of hip-preservation surgery (eg, hip arthroscopy). This finding has implications and could limit daily activities (long-stride walking), sexual activity, ballet dancing, and sports (eg, yoga or skiing), although not studied directly. Good correlation between impingement area and combined version supports evaluation of combined version in female patients with positive posterior impingement test or posterior hip pain.

[Treatment strategies for the combination of hip dysplasia, femoroacetabular impingement and malrotation of the proximal femur : How much should be corrected?]

BACKGROUND

Hip dysplasia, FAI and femoral malrotation often occur together, resulting in mixed symptoms and severe biomechanical limitations of the hip.

OBJECTIVES

To report on the current recommendations for the best possible diagnosis and treatment strategies of combination pathologies in hip-preserving surgery.

METHODS

Review and discussion of the relevant literature with consideration of own experience in the treatment of complex combined pathomorphologies of the hip.

RESULTS

Patient history and a thorough clinical examination are key for determining the predominant pathomorphologies causing the symptoms. Standardized conventional ap pelvic and axial images of the hip are the basis for the radiological assessment of the hip, supplemented with MRI, CT and animations of the hip, depending on the case. As the pathologies influence each other functionally, a stepwise approach to treatment is recommended. The functionally most relevant pathology is treated first, followed by further corrections as needed. The primary goal is to achieve a stable hip with normal acetabular coverage, followed by an impingement-free range of motion and normalized musculoskeletal function. Care must be taken in the choice of surgical method to ensure that all pathologies can be adequately treated.

CONCLUSION

Complex, combined pathologies of the hip can be treated sufficiently with hip-preserving surgery. A thorough diagnosis is important in order to recognize the functional interaction of the different pathologies. The goal of the surgical therapy is a correctly covered, stable hip with a normal range of motion.