Youssef Ouakrim

Context aware adaptable approach for fall detection bases on smart textile

Fall detection is very important to provide adequate interventions for aging people in risk situations. Existing techniques focus on detecting falls using wearable or ambient sensors. However, they do not consider fall orientations. In this paper, we present our novel fall detection system based on smart textiles and machine learning techniques. Using a non-linear support vector machine, we determine the fall orientation which will be helpful to study the impact of a fall according to its orientation. Additionally, we classify falls based on their orientations among 11 classes (moving upstairs, moving downstairs, walking, running, standing, fall forward, fall backward, fall right, fall left, lying, sitting). Results show the reliability of the proposed approach for falls detection (98% of accuracy, 97.5% of sensitivity and 98.5% specificity) and also for fall orientation (98.5% of accuracy).

Data correction to determine a representative pattern of a set of 3D knee kinematic measurements

The purpose of this study is to determine a representative pattern of a set of three dimensional (3D) knee kinematic measurement curves recorded throughout several trials with a patient walking on a treadmill. The measurements are knee angles, (namely joint angles) with respect to the sagittal, frontal, and transverse planes, as a function of time during a gait cycle. Two serious difficulties met while extracting a representative pattern from the trials are that the curves possess phase variability and there are outliers. We propose a scheme which first removes outliers using the modified band depth index method, and follows with phase variability reduction by curve registration. This scheme leads to retaining the mean curve of the corrected set of curves, as the most representative.The purpose of this study is to determine a representative pattern of a set of three dimensional (3D) knee kinematic measurement curves recorded throughout several trials with a patient walking on a treadmill. The measurements are knee angles, (namely joint angles) with respect to the sagittal, frontal, and transverse planes, as a function of time during a gait cycle. Two serious difficulties met while extracting a representative pattern from the trials are that the curves possess phase variability and there are outliers. We propose a scheme which first removes outliers using the modified band depth index method, and follows with phase variability reduction by curve registration. This scheme leads to retaining the mean curve of the corrected set of curves, as the most representative.

Mechanical biomarkers of medial compartment knee osteoarthritis diagnosis and severity grading: Discovery phase

OBJECTIVE To investigate, as a discovery phase, if 3D knee kinematics assessment parameters can serve as mechanical biomarkers, more specifically as diagnostic biomarker and burden of disease biomarkers, as defined in the Burden of Disease, Investigative, Prognostic, Efficacy of Intervention and Diagnostic classification scheme for osteoarthritis (OA) (Altman et al., 1986). These biomarkers consist of a set of biomechanical parameters discerned from 3D knee kinematic patterns, namely, flexion/extension, abduction/adduction, and tibial internal/external rotation measurements, during gait recording. METHODS 100 medial compartment knee OA patients and 40 asymptomatic control subjects participated in this study. OA patients were categorized according to disease severity, by the Kellgren and Lawrence grading system. The proposed biomarkers were identified by incremental parameter selection in a regression tree of cross-sectional data. Biomarker effectiveness was evaluated by receiver operating characteristic curve analysis, namely, the area under the curve (AUC), sensitivity and specificity. RESULTS Diagnostic biomarkers were defined by a set of 3 abduction/adduction kinematics parameters. The performance of these biomarkers reached 85% for the AUC, 80% for sensitivity and 90% for specificity; the likelihood ratio was 8%. Burden of disease biomarkers were defined by a 3-decision tree, with sets of kinematics parameters selected from all 3 movement planes. CONCLUSION The results demonstrate, as part of a discovery phase, that sets of 3D knee kinematic parameters have the potential to serve as diagnostic and burden of disease biomarkers of medial compartment knee OA.

Knee Kinematics Feature Selection for Surgical and Nonsurgical Arthroplasty Candidate Characterization

Keywords: Knee Kinematic, Biomechanical Data, Feature Selection, Complexity Measures, Arthroplasty. Abstract: The purpose of this study is to investigate a method to select a set of knee kinematic data fatures to characterize surgical vs nonsurgical arthroplasty subjects. The kinematic features are generated from 3D knee kinematic data patterns, namely, rotations of flexion-extension, abduction-adduction, and tibial internal-external recorded during a walking task on a dedicated treadmill. The discrimination features are selected using three types of statistical complexity measures: the Fisher discriminant ratio, volume of overlap region, and feature efficiency. The interclass distance measurements which the features thus selected induce demonstrate their effectiveness to characterize surgical and nonsurgical subjects for arthroplasty.

Physiological Data Validation of the Hexoskin Smart Textile

Keywords: Concordance Correlation Coefficient, Intraclass Correlation Coefficient, Bland-Altman, Agreement Analysis, Cardiorespiratory. Abstract: The aim of this study is to validate cardiorespiratory function measurement of a healthy population provided by a wearable textile during a progressive maximal exercise test. The following measurements were collected using embedded sensors to assess three variables: heart rate (HR), breathing rate (BR) and ventilation (Ve). These variables were recorded simultaneously by the wearable textile and using as a reference system for a comparison purpose. The validation was performed based on the two systems agreement estimation by calculating the intraclass correlation coefficient (ICC), the concordance correlation coefficient (CCC) and the Bland-Altman plot for each variable. Twenty-eight healthy volunteers participated in this study. Analysis of each participant under exercise condition by the two measurement systems revealed high CCC values (rc between 0.91 and 0.99), no deviation from the 45� line (Cb between 0.96 and 0.99) and significant ICC values (r between 0.91 and 0.99, p < 0.05) for HR and BR. The Bland Altman plot for HR and BR indicated no deviation of the mean difference from zero and a small variability with tight agreement limits. However, the analysis of the estimated ventilation Ve of each participant revealed doubtful values for the CCC (rc between 0.2 and 0.99) and ICC (r between 0.11 and 0.99). In summary, the Hexoskin presented good agreement for HR and BR. However, for ventilation, it is difficult to conclude from the results due to variability

A variational method to determine the most representative shape of a set of curves and its application to knee kinematic data for pathology classification

The purpose of this study is to investigate a variational method to determine the most representative shape of a family of curves and its application to three-dimensional knee kinematic data for knee pathology classification. High variability and the presence of outliers are characteristic of the data in this application. This method determines the most representative shape by averaging the family curves corrected to account for outliers occurrence and family variability. To this effect, the correction is performed by simultaneous minimization of a set of objective functions, one for each family curve consisting of two terms: a data term of conformity of the corrected curve to the given family curve, and a regularization term of proximity of the corrected curve to the mean of the corrected curves to inhibit the influence of outliers in the family. Minimization is carried out efficiently by particle swarm optimization, a method which, in contrast to gradient descent, is robust to the presence of outliers. Experimental results using real-world data in knee osteoarthritis pattern classification demonstrate the validity and efficiency of the method. Comparisons to conventional methods used to determine the most representative shape are given.

An analysis of 3D knee kinematic data complexity in knee osteoarthritis and asymptomatic controls

Three-dimensional (3D) knee kinematic data, measuring flexion/extension, abduction/adduction, and internal/external rotation angle variations during locomotion, provide essential information to diagnose, classify, and treat musculoskeletal knee pathologies. However, and so across genders, the curse of dimensionality, intra-class high variability, and inter-class proximity make this data usually difficult to interpret, particularly in tasks such as knee pathology classification. The purpose of this study is to use data complexity analysis to get some insight into this difficulty. Using 3D knee kinematic measurements recorded from osteoarthritis and asymptomatic subjects, we evaluated both single feature complexity, where each feature is taken individually, and global feature complexity, where features are considered simultaneously. These evaluations afford a characterization of data complexity independent of the used classifier and, therefore, provide information as to the level of classification performance one can expect. Comparative results, using reference databases, reveal that knee kinematic data are highly complex, and thus foretell the difficulty of knee pathology classification.

Biomechanical signal classification of surgical and non-surgical candidates for knee arthroplasty

The purpose of this article is two-fold : (1) to select a set of bio-mechanical features to characterize arthroplasty candidates and, (2) design a surgical and non-surgical candidate classifier via decision trees. The biomechanical features are generated from 3D knee kinematic patterns, namely, flexion-extension, abduction-adduction, and tibial internal-external rotation measurements taken during gait recordings. The selection of features is done by incremental selection of biomechanical parametes in a classification tree of cross-sectional data. These features are then used to generate decision rules for classification. The effectiveness of the classifier is evaluated by receiver operating characteristic curve analysis, namely, the area under the curve (AUC), sensitivity, and specificity. The classification accuracy is 85% for AUC, 80% for sensitivity, and 90% for specificity. These results demonstrate the effectiveness of the selected biomechanical features and decision tree classifier to perform automatic and objective classification of surgical and non-surgical candidates for arthroplasty.

Kinematic gait analysis of workers exposed to knee straining postures by Bayes decision rule

Deep knee flexion postures such as kneeling and squatting have been demonstrated, in recent review of occupational kneedisorders, as a risk factor of developing knee osteoarthritis (OA). This study investigates a probabilistic method to analyze kneegait kinematics measurements of workers exposed to knee straining postures to determine if they are in any way similar tothose of knee OA patients. The measurements we use are clinically relevant kinematic signals, namely the variation duringa locomotion gait cycle of the angles the knee makes with respect to the three-dimensional (3D) planes of flexion/extension,internal/external rotation, and abduction/adduction. Three groups of participants were used: a set of 24 workers exposed to kneestraining postures (KS workers) acting as a test group, a control group of 25 non-KS posture workers, and a reference knee OAgroup of 29 subjects. We compared the kinematic data of KS workers to those of knee OA patients and non-KS subjects using theBayes decision theory. The results shows that, using the 3D data taken together or the abduction/adduction data, the KS workersresembles often to the OA patients. The analysis on the transverse plane and on sagittal plane, i.e., the flexion/extension and theinternal/external rotation, are not conclusive as the similarities are not significant. The kinematic gait analysis by Bayes decisionrule shows the similarity of workers exposed to knee straining postures to OA gait pattern and justifies further prospective studiesof KS workers in order to assess if gait pattern could be modified even before the onset of the disease.

Knee Frontal Plane Dynamic Alignment in Knee Osteoarthritis Patients: Impact of a 12-week Physical Therapy Program and Relationship with Functional Scores

tendonosis or a partial tear of a gluteus medius and/or minimus tendon. The patients were then contacted and asked to complete questionnaires to assess their pain, function, and satisfaction with treatment. Main Outcome Measures: Improvement in pain, patient satisfaction and function following PRP injection. These were measured by follow-up questionnaires that included the Visual Numerical Scale, Functional Rating Index, and North American Spine Society patient satisfaction index. Results or Clinical Course: 10 patients met criteria (9 female, 1 male). Average age was 64.7 years. Mean duration of pain prior to injection was 46 months (range 8 to 120 mo). Mean follow up was 10.2 months (range 6 to 26 mo). The average pain score (as measured by VAS) was 8.10 (SD 1.7) pre-injection and 3.8 (SD 2.7) post-injection (p 1⁄4 .002). Overall patient satisfaction was 80% (as measured by NASS). Two patients reported no improvement. Of the 8 patients who reported improvement, their mean Functional Rating Index (FRI) score was 62.4 (out of 100) before their PRP injection and 21.3 six months post-injection (p 1⁄4 .001). The average pain score (VAS) among these 8 patients was 8.7 (SD 1.1) pre-injection and 2.7 (SD 2.1) post-injection. Conclusions: The results from this small retrospective case series suggest that PRP may be an efficacious conservative treatment option for patients with symptomatic gluteus medius tears, degeneration, and tendinosis. More robust prospective studies are needed to better evaluate this treatment.