A Survey of Knee Osteoarthritis Assessment Based on Gait

A Statistical Model for Prediction of Lower Limb Injury of Active Sportsperson

“…And this result also supports the finding of the present study. Many research evidence suggested the difference of knee kinematics between healthy and injured subjects and highlighted the importance of medical application of knee joint kinematics to classify injury among different populations [30,31] however, the association between altered knee kinematics and lower limb injury seems to be conflicting according to Cronström et al [32]. The result of the comparative descriptive statistics about toe out angle is supported by the results of Rosenbaum [33] and Simpson & Jiang [34].…”

A vision‐based clinical analysis for classification of knee osteoarthritis, Parkinson's disease and normal gait with severity based on k‐nearest neighbour

“…In this article, we have presented a simple method to create a low-cost passive marker-based dataset and an approach based on Fractional Order Darwinian Particle Swarm Optimization (FODPSO) and KNN for the identification and classification of KOA, PD, and NM subjects. The proposed work contribution's concerning state-of-literature performed in Kour et al (2020aKour et al ( , 2020b and Kour et al (2019) describing novelty of work can be summarized in the following points:…”

“…Exclusion criteria were any disorder other than KOA and PD such as hip or ankle osteoarthritis, rheumatoid arthritis, and so forth. The current dataset is publicly available and is online under licence CC BY 4.0 on Mendeley (Kour et al, 2020a(Kour et al, , 2020b. The dataset considers two sequences for each gait video (left to right and right to left) in the sagittal plane, the sample frames of which are presented in Figure 5.…”

“…A work by Sharifi Renani et al (2020) experimented with 15 inertial sensor configurations for gait‐based OA analysis using deep neural networks and found more accuracy with sensors positioned on feet‐thigh with a percent error from 2.1% to 73.7%. Concerning KOA and PD research, the detailed state‐of‐art has been also provided in our previous works (Kour et al, 2019; Kour et al, 2020a, 2020b). The deep machine learning algorithms applied to various other application areas for different diseases detection such as a novel PCA‐ whale optimization oriented deep neural network (Gadekallu et al, 2020) for tomato plant disease identification, optimized fuzzy logic with cuckoo search (Gadekallu & Khare, 2017) used for heart and diabetic disease prediction can experiment in the present study for reliable KOA and PD gait analysis.…”

“…In this article, we have presented a simple method to create a low‐cost passive marker‐based dataset and an approach based on Fractional Order Darwinian Particle Swarm Optimization (FODPSO) and KNN for the identification and classification of KOA, PD, and NM subjects. The proposed work contribution's concerning state‐of‐literature performed in Kour et al (2020a, 2020b) and Kour et al (2019) describing novelty of work can be summarized in the following points: The study made an effort to explore the effectiveness of the MB paradigm for KOA and PD gait evaluation on three severity levels which have been ignored in previous works. This research article extends to focus on two different diseases, that is, KOA and PD simultaneously and their gait analysis with an unaffected gait as well. Due to the lack of public MB datasets, the proposed study created the first publicly available low‐cost MB dataset for KOA, PD, and NM subjects with early, moderate and severe stages. The study presented an effective method to detect ROI using colour segmentation with FODPSO and extracted robust body motion features along linear kinematic and SPT gait features, differently from other studies. This study performed statistical as well as machine learning analysis comparing KNN, SVM, LR, and RF to evaluate their classification performance based on various metrics. …”