Knee menisci segmentation using convolutional neural networks: data from the Osteoarthritis Initiative

Tack, Alexander; Mukhopadhyay, Anirban; Zachow, Stefan

doi:10.1016/j.joca.2018.02.907

Cited by 98 publications

(81 citation statements)

References 27 publications

(28 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…images. DSCs for medial and lateral menisci were also higher in our algorithm than in previous reports [7,8]. This was probably due to the use of an additional PDWI sequence for meniscus segmentation.…”

Section: Discussionmentioning

confidence: 42%

“…For segmentation accuracy, DSCs for bone and cartilage were higher in our algorithm than in previous reports [5][6][7][8]. This was possibly because we used 3D U-Net, a learning algorithm specialized for 3D…”

Section: Discussionmentioning

confidence: 79%

“…However, this promising method is not popular at present because segmentation of cartilage and the meniscus often requires manual operation or correction, which requires time and effort. To solve this problem, automatic segmentation techniques using deep neural networks have been developed [5][6][7][8]. We have also recently developed a system for automatic extraction of cartilage and the meniscus using deep neural networks.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Fully automatic three-dimensional MRI analysis for the cartilage and meniscus of the knee: Kanagawa Knee Study

Aoki¹,

Ozeki²,

Katano³

et al. 2020

Preprint

View full text Add to dashboard Cite

Objective We have developed a fully automatic three-dimensional MRI analysis software that measures the projected cartilage area ratio (PCAR) to allow for the quantification of the cartilage in the knee. Our objectives for this cross-sectional study were to verify our software’s accuracy and to quantify cartilage and meniscus extrusion. We also examined which cartilage quantification was most affected by age and analyzed the relationship between PCAR and meniscus extrusion. Methods 108 subjects were selected for training, and Dice similarity coefficients were determined from 5 other subjects. This study included 561 subjects between 30–70 years of age. From their knee MRI data, we quantified cartilage thickness, cartilage volume, and PCAR (0.0–1.5 mm) in four regions, including the medial tibial (MT) cartilage. Furthermore, each region was divided into nine subregions. The medial central (mc) subregion was also analyzed. As a quantification for meniscus extrusion, the medial meniscus coverage ratio (MMCR) was also investigated. Results Dice similarity coefficients were 0.911 and 0.892 for the femoral and tibial cartilage and 0.916 and 0.891 for the medial and lateral meniscus. Among 48 cartilage quantifications, the highest absolute value of the correlation coefficient with age was mcMT PCAR 1.0 mm in females and mcMT cartilage thickness in males. In females, mcMT PCAR 1.0 mm was correlated with MMCR, although MMCR was not correlated with age. In males, mcMT PCAR 0.0 mm was correlated with MMCR. Conclusions Our software showed high segmentation accuracy and provided numerous quantifications of cartilage related to age and meniscus extrusion.

show abstract

Section: Discussionmentioning

confidence: 42%

Section: Discussionmentioning

confidence: 79%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Fully automatic three-dimensional MRI analysis for the cartilage and meniscus of the knee: Kanagawa Knee Study

Aoki¹,

Ozeki²,

Katano³

et al. 2020

Preprint

View full text Add to dashboard Cite

show abstract

“…Work is underway to utilize image recognition techniques, such as convolution neural networks to automate the image segmentation process (Burton II et al, 2019;Tack et al, 2018), and once that barrier is removed it is hoped that it will be possible to scan a patient's joint, quickly generate a model, and select the optimal implant geometry and surgical alignments to recreate the patient's previously healthy joint kinematic profiles.…”

Section: Computational Simulations (Fem)mentioning

confidence: 99%

Development of a statistical shape-function model of the implanted knee for real-time prediction of joint mechanics

Gibbons

Clary

Rullkoetter

et al. 2019

Journal of Biomechanics

View full text Add to dashboard Cite

Outcomes of total knee arthroplasty (TKA) are dependent on surgical technique, patient variability, and implant design. Non-optimal design or alignment choices may result in undesirable contact mechanics and joint kinematics, including poor joint alignment, instability, and reduced range of motion. Implant design and surgical alignment are modifiable factors with potential to improve patient outcomes, and there is a need for robust implant designs that can accommodate patient variability. Our objective was to develop a statistical shape-function model (SFM) of a posterior stabilized implant knee to instantaneously predict output mechanics in an efficient manner. Finite element methods were combined with Latin hypercube sampling and regression analyses to produce modeling equations relating nine implant design and six surgical alignment parameters to tibiofemoral (TF) joint mechanics outcomes during a deep knee bend. A SFM was developed and TF contact mechanics, kinematics, and soft tissue loads were instantaneously predicted from the model. Average normalized root-mean-square prediction errors were between 2.79% and 9.42%, depending on the number of parameters included in the model. The statistical shape-function model generated instantaneous joint mechanics predictions using a maximum of 130 training simulations, making it ideally suited for integration into a patient-specific design and alignment optimization pipeline. Such a tool may be used to optimize kinematic function to achieve more natural motion or minimize implant wear and may aid the engineering and clinical communities in improving patient satisfaction and surgical outcomes. viii TABLE OF CONTENTS

show abstract

“…However, this promising method is not popular at present because segmentation of cartilage and the meniscus often requires manual operation or correction, which requires time and effort. To solve these problems, automatic segmentation techniques using deep neural networks have been developed [4][5][6][7]. We have also recently developed a system for automatic extraction of cartilage and meniscus using deep neural networks.…”

Section: Read Full License Introductionmentioning

confidence: 99%

Fully automatic three-dimensional MRI analysis for the cartilage and meniscus of the knee: Kanagawa Knee Study

Aoki

Ozeki

Katano

et al. 2020

Preprint

View full text Add to dashboard Cite

ObjectiveWe have developed a fully automatic three-dimensional MRI analysis software that measures the projected cartilage area ratio (PCAR) to allow for the quantification of the cartilage in the knee. Our objectives were to verify our software’s accuracy for segmentation of cartilage and meniscus and to quantify cartilage and meniscus extrusion in our cross-sectional study. We also examined which cartilage quantification was most affected by age and analyzed the relationship between PCAR and meniscus extrusion.MethodsMRI data from 108 subjects were selected for training, and Dice similarity coefficients were determined from 5 other subjects to verify the accuracy. Our cross-sectional study included other 561 subjects between 30–70 years of age. We quantified cartilage thickness, cartilage volume, and PCAR (0.0–1.5 mm) in medial femoral, lateral femoral, medial tibial, and medial lateral regions. Each region was divided into nine subregions, with particular focus on the medial central (mc) subregion. The medial meniscus coverage ratio (MMCR) was investigated as a quantification for meniscus extrusion.ResultsDice similarity coefficients for cartilage and meniscus were both approximately 0.9. Among cartilage quantifications, the highest absolute value of the correlation coefficient with age was mcMT PCAR 1.0 mm in females and mcMT cartilage thickness in males. In females, mcMT PCAR 1.0 mm was correlated with MMCR, although MMCR was not correlated with age. In males, mcMT PCAR 0.0 mm was correlated with MMCR.ConclusionsOur software showed high segmentation accuracy and provided numerous quantifications of cartilage related to age and meniscus extrusion.Trial registration: UMIN, UMIN000032826. Registered 1 September 2018, https://upload.umin.ac.jp/cgi-open-bin/ctr/ctr_view.cgi?recptno=R000037299

show abstract

Knee menisci segmentation using convolutional neural networks: data from the Osteoarthritis Initiative

Cited by 98 publications

References 27 publications

Fully automatic three-dimensional MRI analysis for the cartilage and meniscus of the knee: Kanagawa Knee Study

Fully automatic three-dimensional MRI analysis for the cartilage and meniscus of the knee: Kanagawa Knee Study

Development of a statistical shape-function model of the implanted knee for real-time prediction of joint mechanics

Fully automatic three-dimensional MRI analysis for the cartilage and meniscus of the knee: Kanagawa Knee Study

Contact Info

Product

Resources

About