scholarly journals Automating classification of osteoarthritis according to Kellgren-Lawrence in the knee using deep learning in an unfiltered adult population

2021 ◽  
Vol 22 (1) ◽  
Author(s):  
Simon Olsson ◽  
Ehsan Akbarian ◽  
Anna Lind ◽  
Ali Sharif Razavian ◽  
Max Gordon
Keyword(s):  
Deep Learning ◽  
Knee Osteoarthritis ◽  
Adult Population ◽  
Weight Bearing ◽  
Correct Diagnosis ◽  
University Hospital ◽  
Knee Oa ◽  
Orthopedic Surgeons ◽  
Visual Disturbances

Abstract Background Prevalence for knee osteoarthritis is rising in both Sweden and globally due to increased age and obesity in the population. This has subsequently led to an increasing demand for knee arthroplasties. Correct diagnosis and classification of a knee osteoarthritis (OA) are therefore of a great interest in following-up and planning for either conservative or operative management. Most orthopedic surgeons rely on standard weight bearing radiographs of the knee. Improving the reliability and reproducibility of these interpretations could thus be hugely beneficial. Recently, deep learning which is a form of artificial intelligence (AI), has been showing promising results in interpreting radiographic images. In this study, we aim to evaluate how well an AI can classify the severity of knee OA, using entire image series and not excluding common visual disturbances such as an implant, cast and non-degenerative pathologies. Methods We selected 6103 radiographic exams of the knee taken at Danderyd University Hospital between the years 2002-2016 and manually categorized them according to the Kellgren & Lawrence grading scale (KL). We then trained a convolutional neural network (CNN) of ResNet architecture using PyTorch. We evaluated the results against a test set of 300 exams that had been reviewed independently by two senior orthopedic surgeons who settled eventual interobserver disagreements through consensus sessions. Results The CNN yielded an overall AUC of more than 0.87 for all KL grades except KL grade 2, which yielded an AUC of 0.8 and a mean AUC of 0.92. When merging adjacent KL grades, all but one group showed near perfect results with AUC > 0.95 indicating excellent performance. Conclusion We have found that we could teach a CNN to correctly diagnose and classify the severity of knee OA using the KL grading system without cleaning the input data from major visual disturbances such as implants and other pathologies.

scholarly journals Automated Classification of Osteoarthritis in the Knee Using Deep Learning

2021 ◽  
Author(s):  
Simon Olsson ◽  
Ehsan Akbarian ◽  
Anna Lind ◽  
Ali Sharif Razavian ◽  
Max Gordon
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Weight Bearing ◽  
Operative Management ◽  
University Hospital ◽  
Automated Classification ◽  
Knee Oa ◽  
Orthopedic Surgeons ◽  
Grading Scale ◽  
Visual Disturbances

Abstract BACKGROUNDPrevalence for knee osteoarthritis is rising in Sweden and globally due to an ageing and more obese population. This has subsequently led to an increasing demand for knee arthroplasties. Correctly diagnosing, classifying, follow-up and planning for either conservative or operative management of knee OA is therefore of a great interest. Most orthopedic surgeons rely on standard weight bearing radiographs, improving the reliability and reproducibility of these interpretations could thus be hugely beneficial. Recently deep learning, a form of artificial intelligence (AI), has shown promising results for interpreting radiographs. In this study, we aim to evaluate how well an AI can classify knee OA severity using entire image series and not excluding common visual disturbances such as implants, casts and other pathologies.MethodsWe selected 6103 radiograph exams taken at Danderyd University Hospital between the years 2002-2016 and manually categorized them according to the Kellgren & Lawrence grading scale (KL). We then trained a convolutional neural network that we evaluated against a test set of 300 exams. These exams had been reviewed independently by two senior orthopedic surgeons who settled exams with disagreement through a consensus session. ResultsOur network yielded an overall high AUC of >0.87 for all KL grades except KL grade 2 and a mean AUC of 0.92. When merging adjacent KL grades together, all but one group showed near perfect results with AUC > 0.95 indicating excellent performance. ConclusionWe found that we could teach a neural network classify knee OA severity and laterality using the KL grading scale without cleaning the input data from major visual disturbances such as implants and other pathologies.

scholarly journals Individuals with medial knee osteoarthritis show neuromuscular adaptation when perturbed during walking despite functional and structural impairments

2014 ◽  
Vol 116 (1) ◽  
pp. 13-23 ◽  
Author(s):  
Deepak Kumar ◽  
Charles (Buz) Swanik ◽  
Darcy S. Reisman ◽  
Katherine S. Rudolph
Keyword(s):  
Knee Osteoarthritis ◽  
Muscle Activation ◽  
Weight Bearing ◽  
Joint Space Width ◽  
Neuromuscular Control ◽  
Emg Activity ◽  
Initial Contact ◽  
Knee Motion ◽  
Quadriceps Weakness ◽  
Knee Oa

Neuromuscular control relies on sensory feedback that influences responses to changing external demands, and the normal response is for movement and muscle activation patterns to adapt to repeated perturbations. People with knee osteoarthritis (OA) are known to have pain, quadriceps weakness, and neuromotor deficits that could affect adaption to external perturbations. The aim of this study was to analyze neuromotor adaptation during walking in people with knee OA ( n = 38) and controls ( n = 23). Disability, quadriceps strength, joint space width, malalignment, and proprioception were assessed. Kinematic and EMG data were collected during undisturbed walking and during perturbations that caused lateral translation of the foot at initial contact. Knee excursions and EMG magnitudes were analyzed. Subjects with OA walked with less knee motion and higher muscle activation and had greater pain, limitations in function, quadriceps weakness, and malalignment, but no difference was observed in proprioception. Both groups showed increased EMG and decreased knee motion in response to the first perturbation, followed by progressively decreased EMG activity and increased knee motion during midstance over the first five perturbations, but no group differences were observed. Over 30 trials, EMG levels returned to those of normal walking. The results illustrate that people with knee OA respond similarly to healthy individuals when exposed to challenging perturbations during functional weight-bearing activities despite structural, functional, and neuromotor impairments. Mechanisms underlying the adaptive response in people with knee OA need further study.

scholarly journals Clinicians’ Perspectives on the Use of Intra-Articular Hyaluronic Acid as a Treatment for Knee Osteoarthritis: A North American, Multidisciplinary Survey

2016 ◽  
Vol 9 ◽  
pp. CMAMD.S34496 ◽  
Author(s):  
Jeffrey Rosen ◽  
Victoria Avram ◽  
Anke Fierlinger ◽  
Faizan Niazi ◽  
Parag Sancheti ◽  
...  
Keyword(s):  
Hyaluronic Acid ◽  
Knee Osteoarthritis ◽  
Patient Outcomes ◽  
Current Literature ◽  
Early Stage ◽  
Sodium Hyaluronate ◽  
Cross Linking ◽  
Product Characteristics ◽  
Knee Oa ◽  
Orthopedic Surgeons

Introduction This study aims to describe the perceptions of orthopedic surgeons on the efficacy of intra-articular hyaluronic acid (IA-HA), the influence of IA-HA product characteristics on its efficacy, and to identify patterns and factors related to the use of IA-HA. Additionally, this study examines factors that influence IA-HA brand selection, focusing on Euflexxa¯ (1% sodium hyaluronate). Methods We developed survey questions by reviewing the current literature and consulting with experts on the use of IA-HA in the management of knee osteoarthritis (OA). The survey included questions on demographics, previous experience with knee OA treatment, opinions on different treatment methods, and where information regarding treatments is obtained. Additionally, questions specific to opinions regarding IA-HA and the reasoning behind these opinions were asked. Results A total of 117 orthopedic surgeons and physicians completed the survey. IA-HA is most frequently prescribed to patients with early-stage (82%) or mid-stage (82.8%) OA, while fewer orthopedic surgeons and physicians use IA-HA for patients with late-stage OA (57.4%). Respondents were generally uncertain of the effects that intrinsic characteristics, such as molecular weight, cross-linking, and production process, had on patient outcomes. Respondents typically use their own clinical experience and results as a deciding factor in utilizing IA-HA treatment, as well as in choosing an IA-HA brand. Conclusion Uncertainty regarding the efficacy of IA-HA treatments is likely due to inconsistency within clinical guidelines and the current literature. Additional research investigating the efficacy of IA-HA treatment and how product characteristics affect outcome and safety is required to provide clarity to the controversy surrounding IA-HA treatment for knee OA.

HIERARCHICAL ANALYSIS AND CLASSIFICATION OF ASYMPTOMATIC AND KNEE OSTEOARTHRITIS GAIT PATTERNS USING A WAVELET REPRESENTATION OF KINETIC DATA AND THE NEAREST NEIGHBOR CLASSIFIER

2008 ◽  
Vol 08 (01) ◽  
pp. 45-54 ◽  
Author(s):  
NEILA MEZGHANI ◽  
KARINE BOIVIN ◽  
KATIA TURCOT ◽  
RACHID AISSAOUI ◽  
NICOLA HAGMEISTER ◽  
...  
Keyword(s):  
Knee Osteoarthritis ◽  
Nearest Neighbor ◽  
Reaction Force ◽  
Hierarchical Classification ◽  
Gait Patterns ◽  
Knee Oa ◽  
Nearest Neighbor Rule ◽  
Kellgren And Lawrence ◽  
Using Data

The purpose of this study is twofold: (1) to develop a classification method to distinguish between asymptomatic (AS) and knee osteoarthritis (OA) gait patterns using ground reaction force (GRF) measurements, and (2) to investigate OA severity within OA gait patterns. Features were first extracted from the GRF vectors to be used for classification. We investigated a two-level hierarchical classification and analysis method using the nearest neighbor rule. At the first level, the GRF data were classified into two classes: AS and OA. At the second level, the GRF data of OA patients were classified according to the pathology severity. The OA patients were grouped into two OA severity categories according to the Kellgren and Lawrence (KL) scale: KL 1 and KL 2 for one category, and KL 3 and KL 4 for the other. Experiments were conducted using data of 42 cases, 16 AS and 26 pathological. The method discriminated between AS and OA subjects with an accuracy of 38 of 42 cases, and assessed the severity correctly with an accuracy of 20 of 26 cases. These results demonstrated the validity of both, the feature and the classifier, for automatic classification of AS and knee OA gait patterns and for analysis of OA severity.

FRI0417 IDENTIFICATION OF THE MOST IMPORTANT FEATURES OF KNEE OSTEOARTHRITIS PROGRESSORS USING MACHINE LEARNING METHODS

2020 ◽  
Vol 79 (Suppl 1) ◽  
pp. 807.1-807
Author(s):  
A. Jamshidi ◽  
M. Leclercq ◽  
A. Labbe ◽  
J. P. Pelletier ◽  
F. Abram ◽  
...  
Keyword(s):  
Machine Learning ◽  
Knee Osteoarthritis ◽  
Prediction Models ◽  
Joint Space ◽  
Classification Methods ◽  
Research Support ◽  
Common Features ◽  
Knee Oa ◽  
Hospital Research

Background:Knee osteoarthritis (OA), a leading cause of disability worldwide, can be difficult to define as its development is often insidious and involves different subgroups. We still lack robust prediction models that are able to guide clinical decisions and stratify OA patients according to risk of disease progression.Objectives:This study aimed at identifying the most important features of knee OA progressors. To this end, we used machine learning (ML) algorithms on a large set of subjects and features to develop advanced prediction models that provide high classification and prediction performance.Methods:Participants, features and outcomes were from the Osteoarthritis Initiative. Features were from baseline (1107), including articular knee tissues (135) assessed by quantitative MRI. OA progressors were ascertained by four outcomes: cartilage volume loss in medial plateau at 48 and 96 months (Prop_CV_48M, 96M); Kellgren-Lawrence (KL) grade ≥2; and medial joint space narrowing (JSN) ≥1 at 48 months. Subjects’ numbers were as follows: 1598 for the outcome Prop_CV_96M, 1044 for the Prop_CV_48M, and 1468 for each KL grade ≥2 at 48 months and JSN ≥1 at 48 months. Six feature selection models were used to identify the common features in each outcome. Six classification methods were applied to measure the accuracy of the selected features in classifying the subjects into progressors and non-progressors. Classification of the best features was done using auto-ML interface and the area under the curve (AUC). To prioritize the top features, Sparse Partial Least Square (sPLS) method was used.Results:For the classification of the best common features in each outcome, Multi-Layer Perceptron (MLP) achieved the highest AUC in Prop_CV_96M, KL, and JSN (0.80, 0.88, 0.95), and Gradient Boosting Machine (GBM) for Prop_CV_48M (0.70). sPLS revealed that the baseline top five features to predict knee OA progressors are the joint space width (JSW), mean cartilage thickness of peripheral, medial, and central tibial plateau, and JSN.Conclusion:This is the first time that such a comprehensive study was performed for identifying the best features and classification methods for knee OA progressors. Data revealed that early prediction of knee OA progression can be done with high accuracy and based on only a few features. This study identifies the baseline X-ray and MRI-based features as the most important for predicting knee OA progressors. These results could be used for the development of a tool enabling prediction of knee OA progressors.Acknowledgments:This work was supported in part by the Osteoarthritis Research Unit of the University of Montreal Hospital Research Centre; the Chair in Osteoarthritis, University of Montreal, (both from Montreal, Quebec, Canada); and the Computational Biology Laboratory, Laval University Hospital Research Center, (Québec, Quebec, Canada). A Jamshidi received a bursary from the Canada First Research Excellence Fund through TransMedTech Institute, (Montreal, Quebec, Canada).Disclosure of Interests:Afshin Jamshidi: None declared, Mickaël Leclercq: None declared, Aurelie Labbe: None declared, Jean-Pierre Pelletier Shareholder of: ArthroLab Inc., Grant/research support from: TRB Chemedica, Speakers bureau: TRB Chemedica and Mylan, François Abram Employee of: ArthroLab Inc., Arnaud Droit: None declared, Johanne Martel-Pelletier Shareholder of: ArthroLab Inc., Grant/research support from: TRB Chemedica

scholarly journals Association of subchondral bone marrow lesion localisation with weight bearing pain in people with knee osteoarthritis: data from the Osteoarthritis Initiative

2021 ◽  
Author(s):  
Koji Aso ◽  
Seyed Mohsen Shahtaheri ◽  
Daniel F. McWilliams ◽  
David A. Walsh
Keyword(s):  
Bone Marrow ◽  
Knee Osteoarthritis ◽  
Knee Pain ◽  
Subchondral Bone ◽  
Negative Binomial ◽  
Weight Bearing ◽  
The Other ◽  
Knee Oa ◽  
Pain Scores ◽  
Osteoarthritis Initiative

Abstract Background Subchondral bone marrow lesions (BMLs) detected on MRI in knee osteoarthritis (OA) are associated with knee pain. The prevalence and progression of subchondral BMLs are increased by mechanical knee load. However, associations of subchondral BML location with weight-bearing knee pain are currently unknown. In this study, we aim to demonstrate associations of subchondral BML location and size with weight-bearing knee pain in knee OA.Methods We analyzed 1412 and 582 varus knees from cross-sectional and longitudinal Osteoarthritis Initiative datasets, respectively. BML scores were semi-quantitatively analysed with the MRI Osteoarthritis Knee Score for 4 subchondral regions (median and lateral femorotibial, medial and lateral patellofemoral) and subspinous region. Weight-bearing and non-weight-bearing pain scores were derived from WOMAC pain items. Correlation and negative binomial regression models were used for analysis of associations between the BML scores and pain at baseline, and changes in the BML scores and changes in pain after 24-month follow up.Results Greater BML scores at medial femorotibial and lateral patellofemoral compartments were associated with greater weight-bearing pain scores, and statistical significance was retained after adjusting for BML scores at the other 4 joint compartments and other OA features, as well as for non-weight-bearing pain, age, sex and Body Mass Index (BMI) (medial femorotibial; B=0.08, p=0.02. patellofemoral; B=0.13, p=0.01). Subanalysis revealed that greater medial femorotibial BML scores were associated with greater pain on walking and standing (B=0.11, p=0.01, and B=0.10, p=0.04, respectively). Lateral patellofemoral BML scores were associated with pain on climbing, respectively B=0.14, p=0.02. Increases or decreases over 24 months in BML score in the medial femorotibial compartment were significantly associated with increases or decreases in weight-bearing pain severity after adjusting for non-weight-bearing pain, age, sex, baseline weight-bearing pain, BMI, and BML at the other 4 joint compartments (B=0.10, p=0.01). Conclusions Subchondral BML size at the medial femorotibial joint compartment was specifically associated with the severity and the change in weight-bearing pain, independent of non-weight-bearing pain, in knee OA. Specific associations of weight-bearing pain with subchondral BMLs in weight-bearing compartments of the knee indicate that BMLs in subchondral bone contribute to biomechanically-induced OA pain.

scholarly journals Tibial Condyle Valgus Osteotomy for Ipsilateral Knee Osteoarthritis after Hip Arthrodesis

2021 ◽  
Vol 2021 ◽  
pp. 1-5
Author(s):  
Daisuke Fukuhara ◽  
Hiroaki Inoue ◽  
Shuji Nakagawa ◽  
Yuji Arai ◽  
Kenji Takahashi
Keyword(s):  
Knee Joint ◽  
Knee Osteoarthritis ◽  
Mechanical Load ◽  
Weight Bearing ◽  
Valgus Osteotomy ◽  
Hip Arthrodesis ◽  
Knee Oa ◽  
Early Loosening ◽  
Tibial Condyle ◽  
End Stage

We report a case of tibial condylar valgus osteotomy (TCVO) for ipsilateral knee osteoarthritis (OA) after hip arthrodesis. A 58-year-old woman developed right purulent hip arthritis at one month of age and underwent right hip fusion at 16 years old. She visited our department at the age of 57 because her right knee joint pain worsened. The range of motion for her right knee was 80° and -5° of flexion and extension, respectively, and she experienced medial weight-bearing pain. A plain X-ray image showed that the right knee joint had end-stage knee OA with a bone defect inside the tibia, and the tibial plateau shape was the pagoda type. There was a marked instability in her right knee with a valgus of 9° and varus of 7° on stress photography. She underwent TCVO on her right knee and was allowed full load four weeks after surgery. Computed tomography imaging showed bone union nine months after surgery. Two years after the operation, there was no correction loss, and she could walk independently without pain. In general, total knee arthroplasty (TKA) is indicated for end-stage knee OA; however, there are problems, such as early loosening due to the increased mechanical load on the knee after hip OA. In this case, since a good course was obtained, TCVO is considered a treatment option for terminal knee OA after hip arthrodesis.

Association Between Clinical Symptoms and Radiographic Features in Late-Stage Knee Osteoarthritis Using a New Radiographic Parameter

Pain Medicine ◽  
2021 ◽  
Author(s):  
Tengjiao Zhu ◽  
Xing Xin ◽  
Bin Yang ◽  
Chen Liu ◽  
Bolong Kou ◽  
...  
Keyword(s):  
Knee Osteoarthritis ◽  
Correlation Coefficient ◽  
Late Stage ◽  
Tibial Plateau ◽  
Weight Bearing ◽  
Womac Pain ◽  
Radiographic Parameter ◽  
Radiographic Features ◽  
Knee Oa ◽  
The Relationship

Abstract Objective In this study, we proposed a new radiographic parameter, the plateau attrition index (PAI), and the PAI grades (PAIs) to explore the relationship between subchondral attrition of the tibial plateau and symptoms of knee osteoarthritis (OA) in patients with late-stage knee osteoarthritis. Method One hundred nineteen patients with late-stage knee osteoarthritis were enrolled. The Kellgren and Lawrence (K/L) grades and hip-knee-ankle (HKA) angle were used to characterize the radiographic features of knee OA. The bone attrition of the tibial plateau was determined by the PAI and PAIs. The symptoms of knee OA were assessed by the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC), which is composed of the WOMAC pain (WOMP), WOMAC stiffness (WOMS) and WOMAC function (WOMF) subscores. WOMAC pain scores were divided into non-weight-bearing pain (NWBP) and weight-bearing pain (WBP) subcategories. The Pearson correlation coefficient was used to determine the relationship between the PAI, HKA angle, and WOMAC scores. The Spearman rank correlation coefficient was used to evaluate the correlation between the WOMAC score and the PAIs and K/L grades. Results The distribution of the WOMAC scores according to the PAIs was significant (p < 0.01). A positive correlation was identified between the PAI and the WOMAC, WOMP, WOMF and WBP scores (r = 0.29, 0.34, 0.26 and 0.34, p < 0.01, respectively). In addition, the PAIs was also significantly correlated with the WOMAC, WOMP, WOMF and WBP scores (r = 0.37, 0.38, 0.35 and 0.44, p < 0.01, respectively). Conclusion The attrition of tibial subchondral bone determined by the new parameter, the plateau attrition index, was correlated with symptoms, especially weight-bearing pain in late-stage knee OA.

scholarly journals Association of Cross Linked C-Telopeptide II Collagen and Hyaluronic Acid with Knee Osteoarthritis Severity

2013 ◽  
Vol 5 (3) ◽  
pp. 179
Author(s):  
John Butar Butar ◽  
Zola Wijayanti ◽  
Beatrix Tjahyana ◽  
Veli Sunggono ◽  
Hori Hariyanto
Keyword(s):  
Hyaluronic Acid ◽  
Knee Osteoarthritis ◽  
Metabolic Diseases ◽  
Weight Bearing ◽  
Interleukin 1 ◽  
High Sensitivity ◽  
Type I ◽  
Knee Oa ◽  
Reactive Protein ◽  
Hs Crp

BACKGROUND: This study was carried out to investigate the association of Cross Linked C-Telopeptide Type I & II Collagen (CTX-I and II) and hyaluronic acid (HA) with knee osteoarthritis (OA) severity.METHODS: Sixty menopause women with primary knee OA were enrolled in this study during their visits to the Outpatient Department. Patients with knee pain during weight bearing, active or passive range of motion, or tenderness with Kellgren-Lawrence (KL) grade of more than I were included. Patients with injury, inflammatory and metabolic diseases were excluded. Patients were put in a 10-hour fasting prior to withdrawal of morning blood samples for examinations of HA, CTX-I, interleukin 1 beta (IL-1β), and high sensitivity C reactive protein (hs-CRP) level. Second void morning urine specimens were taken for CTXII assessment. HA, CTX-I and II levels were measured by enzyme-linked immunosorbent assay.RESULTS: Sixty menopausal female patients were included in this study, 35 with KL grade II, 17 grade III, and 8 grade IV. Means of CTX-II were significantly different between subjects KL grade IV and III (p=0.021). Correlation of KL grade was significant with CTX-II (p=0.001, r=0.412) and HA (p=0.0411, r=0.269). KL grades were not significantly associated with CTX-I (p=0.8364, r=-0.0272); IL-1β (p=0.5773, r=0.0853) and hs-CRP (p=0.2625, r=0.1470).CONCLUSION: CTX-II and HA were associated with severity of knee OA, suggesting that CTX-II and HA can be used as marker for knee OA severity.KEYWORDS: CTX-II, hyaluronic acid, otestoarthritis, knee

scholarly journals Emergence of Deep Learning in Knee Osteoarthritis Diagnosis

2021 ◽  
Vol 2021 ◽  
pp. 1-20
Author(s):  
Pauline Shan Qing Yeoh ◽  
Khin Wee Lai ◽  
Siew Li Goh ◽  
Khairunnisa Hasikin ◽  
Yan Chai Hum ◽  
...  
Keyword(s):  
Deep Learning ◽  
Knee Osteoarthritis ◽  
Research Field ◽  
Knee Joints ◽  
Learning Approaches ◽  
Clinical Practices ◽  
Clinical Workflow ◽  
3 Dimensional ◽  
Knee Oa ◽  
3D Cnn

Osteoarthritis (OA), especially knee OA, is the most common form of arthritis, causing significant disability in patients worldwide. Manual diagnosis, segmentation, and annotations of knee joints remain as the popular method to diagnose OA in clinical practices, although they are tedious and greatly subject to user variation. Therefore, to overcome the limitations of the commonly used method as above, numerous deep learning approaches, especially the convolutional neural network (CNN), have been developed to improve the clinical workflow efficiency. Medical imaging processes, especially those that produce 3-dimensional (3D) images such as MRI, possess ability to reveal hidden structures in a volumetric view. Acknowledging that changes in a knee joint is a 3D complexity, 3D CNN has been employed to analyse the joint problem for a more accurate diagnosis in the recent years. In this review, we provide a broad overview on the current 2D and 3D CNN approaches in the OA research field. We reviewed 74 studies related to classification and segmentation of knee osteoarthritis from the Web of Science database and discussed the various state-of-the-art deep learning approaches proposed. We highlighted the potential and possibility of 3D CNN in the knee osteoarthritis field. We concluded by discussing the possible challenges faced as well as the potential advancements in adopting 3D CNNs in this field.

Sign in / Sign up

Export Citation Format

Share Document