Stacked sparse autoencoder networks and statistical shape models for automatic staging of distal femur trochlear dysplasia

Cerveri, Pietro; Belfatto, Antonella; Baroni, Guido; Manzotti, Alfonso

doi:10.1002/rcs.1947

Cited by 15 publications

(16 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Digital bony shapes of distal femur and proximal tibia were extracted from a retrospective dataset of 100 patients (70 males and 30 females) provided in anonymized form by Medacta company (Medacta International SA, Castel S. Pietro, CH), including planning CT scans (acquired in a supine position for all patients) and reconstructed bony 3D surfaces (Cerveri et al, 2017(Cerveri et al, , 2018. The patients, aged 67 ± 10 years, reported localized knee pain associated with mechanical knee instability at staging time.…”

Section: Patient Datamentioning

confidence: 99%

“…In order to construct the SSM embedding femur and tibia shapes, the methodology extensively described in previous papers of our group was adopted, which is based on a pair-wise matching technique (Cerveri et al, 2017(Cerveri et al, , 2018(Cerveri et al, , 2019a. This technique rests on the manual selection of a reference geometry for aligning all the surfaces in the training dataset and computing robust point correspondences.…”

Section: Statistical Shape Modelmentioning

confidence: 99%

“…where the reconstructed surfaceŠ is obtained by summing up the mean model m with the series of all MoVs. For each surface, the weights were computed by projecting the shape pair (distal femur and proximal tibia) on the SSM by means of the scalar product (Cerveri et al, 2018). We retained enough MoVs to describe 95% of the overall shape variation, expressed by the cumulative EV, in the study population.…”

Section: Statistical Shape Modelmentioning

confidence: 99%

“…The magnitude of each MoV outlines the extent to which the morphological aspect it encodes is present in the set. Applications of bony surface SSMs in biomechanics and clinics have spanned anatomical and developmental studies (Li et al, 2010;Zhu and Li, 2011;Mutsvangwa et al, 2015;Baumbach et al, 2017;Wang and Shi, 2017;Zhang and Besier, 2017), shape anomaly staging (Van Haver et al, 2014;Agricola et al, 2015;Zhang et al, 2016;Cerveri et al, 2018;Chan et al, 2018), joint osteoarthritis (Neogi et al, 2013;Van Dijck et al, 2018), surgical planning and intervention (Zheng and Schumann, 2009;Cerveri et al, 2017;Mauler et al, 2017;Youn et al, 2017), and morphology-function relations (Fitzpatrick et al, 2011;Rao et al, 2013;Baka et al, 2014;Peloquin et al, 2014;Smoger et al, 2015;Hollenbeck et al, 2018;Cerveri et al, 2019b;Clouthier et al, 2019). There have, however, been few studies attempting to extensively investigate the relationship between morphological features and the degree of deformity of the tibiofemoral joint affecting the mechanical stability of the knee.…”

Section: Introductionmentioning

confidence: 99%

“…For example, changes in the frontal plane affected the mechanical alignment by at most ±3 • , which is considered the normal range for frontal stability of the knee. Based on such literature and capitalizing on our previous works (Cerveri et al, 2017(Cerveri et al, , 2018(Cerveri et al, , 2019a, in this paper, an SSM of pathological bony shapes in the knee is proposed to investigate the correlation between MoVs and the mechanical deformity of tibia and femur, assumed to induce kinematic instability. The statistical shape model of the tibia-femur bone complex was built using 99 pathological cases.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Predicting Knee Joint Instability Using a Tibio-Femoral Statistical Shape Model

Cerveri

Belfatto

Manzotti

2020

Front. Bioeng. Biotechnol.

Self Cite

View full text Add to dashboard Cite

Statistical shape models (SSMs) are a well established computational technique to represent the morphological variability spread in a set of matching surfaces by means of compact descriptive quantities, traditionally called "modes of variation" (MoVs). SSMs of bony surfaces have been proposed in biomechanics and orthopedic clinics to investigate the relation between bone shape and joint biomechanics. In this work, an SSM of the tibio-femoral joint has been developed to elucidate the relation between MoVs and bone angular deformities causing knee instability. The SSM was built using 99 bony shapes (distal femur and proximal tibia surfaces obtained from segmented CT scans) of osteoarthritic patients. Hip-knee-ankle (HKA) angle, femoral varus-valgus (FVV) angle, internal-external femoral rotation (IER), tibial varus-valgus (TVV) angles, and tibial slope (TS) were available across the patient set. Discriminant analysis (DA) and logistic regression (LR) classifiers were adopted to underline specific MoVs accounting for knee instability. First, it was found that thirty-four MoVs were enough to describe 95% of the shape variability in the dataset. The most relevant MoVs were the one encoding the height of the femoral and tibial shafts (MoV #2) and the one representing variations of the axial section of the femoral shaft and its bending in the frontal plane (MoV #5). Second, using quadratic DA, the sensitivity results of the classification were very accurate, being all >0.85 (HKA: 0.96, FVV: 0.99, IER: 0.88, TVV: 1, TS: 0.87). The results of the LR classifier were mostly in agreement with DA, confirming statistical significance for MoV #2 (p = 0.02) in correspondence to IER and MoV #5 in correspondence to HKA (p = 0.0001), FVV (p = 0.001), and TS (p = 0.02). We can argue that the SSM successfully identified specific MoVs encoding ranges of alignment variability between distal femur and proximal tibia. This discloses the opportunity to use the SSM to predict potential misalignment in the knee for a new patient by processing the bone shapes, removing the need for measuring clinical landmarks as the rotation centers and mechanical axes.

show abstract

Section: Patient Datamentioning

confidence: 99%

Section: Statistical Shape Modelmentioning

confidence: 99%

Section: Statistical Shape Modelmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Predicting Knee Joint Instability Using a Tibio-Femoral Statistical Shape Model

Cerveri

Belfatto

Manzotti

2020

Front. Bioeng. Biotechnol.

Self Cite

View full text Add to dashboard Cite

show abstract

Three‐dimensional bone morphology is a risk factor for medial postmeniscectomy syndrome: A retrospective cohort study

Grammens,

Van Haver,

Danckaers

et al. 2024

J. exp. orthop.

View full text Add to dashboard Cite

PurposeThe study aims to identify differences in tibiofemoral joint morphology between responders (R group, no pain) to arthroscopic partial medial meniscectomy (APMM) versus medial postmeniscectomy syndrome patients (MPMS group, recurrent pain at 2 years postmeniscectomy) in a clinically neutrally aligned patient population. The second aim was to build a morphology‐based predictive algorithm for response to treatment (RTT) in APMM.MethodsTwo patient groups were identified from a large multicentre database of meniscectomy patients at 2 years of follow‐up: the R group included 120 patients with a KOOS pain score > 75, and the MPMS group included 120 patients with a KOOS pain score ≤ 75. Statistical shape models (SSMs) of distal femur, proximal tibia and tibiofemoral joint were used to compare knee morphology. Finally, a predictive model was developed to predict RTT, with the SSM‐derived morphologic variables as predictors.ResultsNo differences were found between the R and MPMS groups for patient age, sex, height, weight or cartilage status. Knees in the MPMS group were significantly smaller, had a wider femoral notch and a smaller medial femoral condyle. A morphology‐based predictive model was able to predict MPMS at 2 years follow‐up with a sensitivity of 74.9% (95% confidence interval [CI]: 74.4%–75.4%) and a specificity of 81.0% (95% CI: 80.6%–81.5%).ConclusionA smaller tibiofemoral joint, a wider intercondylar notch and smaller medial femoral condyle were observed shape variations related to medial postmeniscectomy syndrome. These promising results are a first step towards a knee morphology‐based clinical decision support tool for meniscus treatment.Study DesignCase–control study.Level of EvidenceLevel IIIb.

show abstract

A landmark-based 3D analysis reveals a narrower tibial plateau and patella in trochlear dysplastic knees

Peeters

Haver

Wyngaert³

et al. 2019

Knee Surg Sports Traumatol Arthrosc

View full text Add to dashboard Cite

Purpose The trochlear dysplastic femur has a specific morphotype previously characterised by not only dysplastic features of the trochlea but also by specific features of the notch and posterior femur. In this study the morphology of the tibia and patella was investigated to gain further insight in the complete geometrical complexity of the trochlear dysplastic knee. Methods Arthro‐CT scan‐based 3D models of 20 trochlear dysplastic and 20 normal knees were uniformly scaled and landmarks and landmark‐based reference planes were created to quantify a series of morphometric characteristics of the tibia and patella. Results In the mediolateral direction, the 3D‐analysis revealed a 3% smaller medial tibial plateau (30.4 ± 1.6 mm vs 31.5 ± 1.6 mm), a 3% smaller overall width of the tibial plateau (73.6 ± 2.0 mm vs 75.7 ± 2.0 mm), a 16% smaller medial facet (17.3 ± 2.2 mm vs 20.1 ± 1.3 mm) and a 4% smaller overall width of the patella (41.7 ± 2.5 mm vs 43.4 ± 2.3 mm) in trochlear dysplastic knees. In the anteroposterior direction, the lateral tibial plateau of trochlear dysplastic knees was 5% larger (37.2 ± 2.3 mm vs 35.5 ± 3.1 mm). A correlation test between the width of the femur and the width of the tibia revealed that trochlear dysplastic knees show less correspondence between the femur and tibia compared to normal knees. Conclusion Significant differences in the morphology of the tibial plateau and patella were detected between trochlear dysplastic and normal knees. Both in the trochlear dysplastic tibial plateau and patella a narrower medial compartment leads to a significant smaller overall mediolateral width. These findings are important for the understanding of knee biomechanics and the design of total knee arthroplasty components. Level of evidence III.

show abstract

Stacked sparse autoencoder networks and statistical shape models for automatic staging of distal femur trochlear dysplasia

Abstract: From a clinical point of view, this paper contributes to support the increasing role of SSM, integrated with deep learning techniques, in diagnostics and therapy definition as quantitative and advanced visualization tools.

Cited by 15 publications

References 44 publications

Predicting Knee Joint Instability Using a Tibio-Femoral Statistical Shape Model

Predicting Knee Joint Instability Using a Tibio-Femoral Statistical Shape Model

Three‐dimensional bone morphology is a risk factor for medial postmeniscectomy syndrome: A retrospective cohort study

A landmark-based 3D analysis reveals a narrower tibial plateau and patella in trochlear dysplastic knees

Contact Info

Product

Resources

About