Vibroarthrographic Research Articles

Prediction of Contralateral Lower-Limb Joint Angles Using Vibroarthrography and Surface Electromyography Signals in Time-Series Network

Multisource biosignals are being increasingly used in human–machine interaction applications. In particular, a critical problem in exoskeleton-assisted rehabilitation for patients with hemiplegia is the generation of rhythmic and symmetrical locomotion. To support lower limb rehabilitation, we propose a model for predicting contralateral joint angles using multisource biosignals. First, a vibroarthrography (VAG) sensor is attached to the affected leg, and surface electromyography sensors are attached to the sound leg. The corresponding signals are used to estimate the hip, knee, and ankle joint angles of the affected leg. Second, an algorithm based on a temporal convolution network (TCN) is introduced to predict the contralateral lower-limb joint angles during human locomotion. The TCN is compared with a long short-term memory (LSTM) network and a convolutional neural network. Experiments were conducted by 10 healthy participants. The results of the proposed model were compared with measurements from encoders in three joints mounted on an exoskeleton to verify the applicability of the proposed model. In addition, by using the outputs of a motion capture system as the ground truth, the experimental results validated the model prediction performance. The prediction root mean square error (RMSE) of the TCN was 52%–70% lower than that of the LSTM network and CNN at different paces. The prediction RMSE using VAG was 20%–24% lower than that without using VAG. Note to Practitioners—We propose a model for predicting the contralateral lower-limb joint angles of an exoskeleton. The model uses measurements from multiple biosignal sensors to support exoskeleton-assisted hemiplegia rehabilitation. A sensing system based on acoustic signals and bioelectric signals is developed to improve the joint angle prediction in the affected leg of patients with hemiplegia. A temporal convolution network is implemented to estimate joint angles, and training considering the root mean square error and Pearson correlation coefficient as evaluation indicators is performed. Unlike similar methods that include no sensory feedback from the affected leg, the proposed method incorporates VAG signals from the affected leg and sEMG signals from the sound leg to handle different degrees of hemiplegia and support the generation of rhythmic and symmetrical locomotion for gait rehabilitation using an exoskeleton.

A novel computer-assisted diagnosis method of knee osteoarthritis based on multivariate information and deep learning model

Knee Osteoarthritis (KOA) is a chronic joint disease characterized by degeneration of knee's articular cartilage. Comparing with the imaging diagnostic method, the vibroarthrographic (VAG) may be emerged as a new candidate for diagnosis of KOA in clinics. However, it is challenging for doctors to evaluate patients' condition by visually detecting VAGs due to the limited understanding about contained information. Besides, the basic physiological signals are also closely correlated with a greater risk of KOA. Based on this, we focus on studying computer-assisted diagnosis method of KOA (KOA-CAD) using multivariate information (i.e. VAGs and basic physiological signals) based on an improved deep learning model (DLM). Firstly, a new Laplace distribution-based strategy (LD-S) for classification in DLM is designed. Secondly, an aggregated multiscale dilated convolution network (AMD-CNN) is constructed to learn features from multivariate information of KOA patients. Then, a new KOA-CAD method is proposed by integrating the AMD-CNN with the LD-S to realize three CAD objectives, including the automatic KOA detection, the KOA early detection, and the KOA grading detection. Finally, the multivariate information collected clinically is applied to verify the proposed method, where the automatic detection accuracy, early detection accuracy and grading detection accuracy are 93.6%, 92.1%, and 84.2% respectively.

Using Nonlinear Vibroartrographic Parameters for Age-Related Changes Assessment in Knee Arthrokinematics.

Changes in articular surfaces can be associated with the aging process and as such may lead to quantitative and qualitative impairment of joint motion. This study is aiming to evaluate the age-related quality of the knee joint arthrokinematic motion using nonlinear parameters of the vibroarthrographic (VAG) signal. To analyse the age-related quality of the patellofemoral joint (PFJ), motion vibroarthrography was used. The data that were subject to analysis represent 220 participants divided into five age groups. The VAG signals were acquired during flexion/extension knee motion and described with the following nonlinear parameters: recurrence rate () and multi-scale entropy (). and decrease almost in a linear way with age (main effects of group ; means (): ; and . The post-hoc analysis showed that there were statistically significant differences () in all comparisons with the exception of the 5th–6th life decade. For , statistically significant differences () occurred for: 3rd–7th, 4th–7th, 5th–7th and 6th life decades. Our results imply that degenerative age-related changes are associated with lower repeatability, greater heterogeneity in state space dynamics, and greater regularity in the time domain of VAG signal. In comparison with linear VAG measures, our results provide additional information about the nature of changes of the vibration dynamics of PFJ motion with age.

A method of feature fusion and dimension reduction for knee joint pathology screening and separability evaluation criteria

Background and objectiveKnee-joint vibroarthrographic (VAG) signal is an effective method for performing a non-invasive knee osteoarthritis (KOA) diagnosis, VAG signal analysis plays a crucial role in achieving the early pathological screening of the knee joint. In order to improve the accuracy of knee pathology screening and to investigate the method suitable for embedded in wearable diagnostic device for knee joint, this paper proposes a knee pathology screening method. Aiming to fill the gap of lacking suitable and unified evaluation indexes for single feature and fusion feature, this paper proposes feature separability evaluation criteria. MethodsIn this paper, we propose a knee joint pathology screening method based on feature fusion and dimension reduction combined with random forest classifier, as well as, the evaluation criteria of feature separability. As for pathological screening method, this paper proposes the idea of multi-dimensional feature fusion, using principal component analysis (PCA) to reduce the redundant part of fusion feature (F-F) to obtain deep fusion feature (D-F-F) with more separability. Meanwhile, this paper proposes the maximal information coefficient (MIC) and correlation matrix collinearity (CMC) feature evaluation criteria, these not only can be used as new feature quantitative metrics, but also illustrate that the divisibility of the deep fusion feature is more potent than that before feature dimension reduction. ResultsThe experimental results show that the method in this paper has good performance in pathology classification on random forest classifier with 96% accuracy, especially the accuracy of SVM and K-NN are also improved after feature dimension reduction. ConclusionThe results indicate that this classification research has high screening efficiency for KOA diagnosis and could provide a feasible method for computer-assisted non-invasive diagnosis of KOA. And we provide a novel way for separability evaluation of VAG signal features.

Diagnostics of Articular Cartilage Damage Based on Generated Acoustic Signals Using ANN-Part II: Patellofemoral Joint.

Cartilage loss due to osteoarthritis (OA) in the patellofemoral joint provokes pain, stiffness, and restriction of joint motion, which strongly reduces quality of life. Early diagnosis is essential for prolonging painless joint function. Vibroarthrography (VAG) has been proposed in the literature as a safe, noninvasive, and reproducible tool for cartilage evaluation. Until now, however, there have been no strict protocols for VAG acquisition especially in regard to differences between the patellofemoral and tibiofemoral joints. The purpose of this study was to evaluate the proposed examination and acquisition protocol for the patellofemoral joint, as well as to determine the optimal examination protocol to obtain the best diagnostic results. Thirty-four patients scheduled for knee surgery due to cartilage lesions were enrolled in the study and compared with 33 healthy individuals in the control group. VAG acquisition was performed prior to surgery, and cartilage status was evaluated during the surgery as a reference point. Both closed (CKC) and open (OKC) kinetic chains were assessed during VAG. The selection of the optimal signal measures was performed using a neighborhood component analysis (NCA) algorithm. The classification was performed using multilayer perceptron (MLP) and radial basis function (RBF) neural networks. The classification using artificial neural networks was performed for three variants: I. open kinetic chain, II. closed kinetic chain, and III. open and closed kinetic chain. The highest diagnostic accuracy was obtained for variants I and II for the RBF 9-35-2 and MLP 10-16-2 networks, respectively, achieving a classification accuracy of 98.53, a sensitivity of 0.958, and a specificity of 1. For variant III, a diagnostic accuracy of 97.79 was obtained with a sensitivity and specificity of 0.978 for MLP 8-3-2. This indicates a possible simplification of the examination protocol to single kinetic chain analyses.

Diagnostics of Articular Cartilage Damage Based on Generated Acoustic Signals Using ANN-Part I: Femoral-Tibial Joint.

Osteoarthritis (OA) is a chronic, progressive disease which has over 300 million cases each year. Some of the main symptoms of OA are pain, restriction of joint motion and stiffness of the joint. Early diagnosis and treatment can prolong painless joint function. Vibroarthrography (VAG) is a cheap, reproducible, non-invasive and easy-to-use tool which can be implemented in the diagnostic route. The aim of this study was to establish diagnostic accuracy and to identify the most accurate signal processing method for the detection of OA in knee joints. In this study, we have enrolled a total of 67 patients, 34 in a study group and 33 in a control group. All patients in the study group were referred for surgical treatment due to intraarticular lesions, and the control group consisted of healthy individuals without knee symptoms. Cartilage status was assessed during surgery according to the International Cartilage Repair Society (ICRS) and vibroarthrography was performed one day prior to surgery in the study group. Vibroarthrography was performed in an open and closed kinematic chain for the involved knees in the study and control group. Signals were acquired by two sensors placed on the medial and lateral joint line. Using the neighbourhood component analysis (NCA) algorithm, the selection of optimal signal measures was performed. Classification using artificial neural networks was performed for three variants: I—open kinetic chain, II—closed kinetic chain, and III—open and closed kinetic chain. Vibroarthrography showed high diagnostic accuracy in determining healthy cartilage from cartilage lesions, and the number of repetitions during examination can be reduced only to closed kinematic chain.

Estimation of differences in selected indices of vibroacoustic signals between healthy and osteoarthritic patellofemoral joints as a potential non-invasive diagnostic tool

Osteoarthritis (OA) is currently the most generic form of joint disease. It is a complex process in which degenerative changes occur in the articular cartilage [AC], subchondral bone, and synovial membrane and can lead to permanent joint failure. The primary and most commonly used method of diagnosing degenerative changes is classic radiography. Magnetic resonance imaging (MRI) may be used to assess the extent of damage to joint surfaces, but this method is limited by the availability of specialised equipment and the excessive cost of the examination. Arthroscopy, an invasive procedure, is considered the “gold standard” in joint diagnosis. The occurrence of degenerative changes is closely related to the friction and lubrication processes within the joint. The main causes of osteoarthritis are a change or lack of synovial fluid, deformation of the joint bones, local damage to the articular cartilage, and a change in the mechanical properties of the articular cartilage due to water loss from the damaged superficial layer. An alternative, non-invasive method that allows for a delicate assessment of the condition of moving joints is vibroarthrography (VAG). The analysis of vibroacoustic signals generated by moving joint surfaces has an immense potential in the non-invasive assessment of the degree of damage to articular cartilage, meniscus and ligaments and the general diagnosis of degenerative diseases. The purpose of this study is to analyse and statistically compare the basic characteristics of vibroacoustic signals recorded with a CM-01B contact microphone placed on the patella for motion in the 90°–0°–90° range in a closed kinetic chain (CKC) in a control group (HC) and a group of patients diagnosed with osteoarthritis (OA), qualified for the knee alloplasty.

Feature Extraction and Classification for the Detection of Knee Joint Disorders using Random Forest Classifier

A non-invasive technique using knee joint vibroarthrographic (VAG) signals can be used for the early diagnosis of knee joint disorders. Among the algorithms devised for the detection of knee joint disorders using VAG signals, algorithms based on entropy measures can provide better performance. In this work, the VAG signal is preprocessed using wavelet decomposition into sub band signals. Features of the decomposed sub bands such as approximate entropy, sample entropy and wavelet energy are extracted as a quantified measure of complexity of the signal. A feature selection based on Principal Component Analysis (PCA) is performed in order to select the significant features. The extracted features are then used for classification of VAG signal into normal and abnormal VAG using random forest classifier. It is observed that the classifier provides a better accuracy with feature selection using principal component analysis. And the result shows that the classifier is able to classify the signal with an accuracy of 87%, error rate of 0.13, sensitivity of 0.874 and specificity of 0.777.

A Patellofemoral Chondropathy Quantification from Vibroarthrography: A Preliminary Study

Objective: To preliminarily investigate if the proposed parameter derived from the knee vibroarthrographic (VAG) signals, namely the VAG score, could potentially be used to quantify the patellofemoral chondropathy. Materials and Methods: Five subjects with meniscus injury as an indication for arthroscopy were recruited for the present preliminary crosssectional study. Prior to the arthroscopy, the Kujala scores were assessed and the knee VAG signals were recorded from all the subjects. Subjects were asked to actively perform the knee flexion-extension for three cycles in the supine position while recording the VAG signal. The proposed VAG score was defined as the power spectral density of the signal in the frequency of 450 to 1,000 Hz. Patellofemoral chondropathy was arthroscopically graded using French Society of Arthroscopy system (SFA) score and SFA category. Results: The SFA score was significantly strongly correlated with the VAG score (r=0.87, p=0.02) but not with the Kujala score (r=–0.79, p=0.11). Likewise, the SFA category was significantly correlated with the VAG score (r=0.98, p<0.01) but not with the Kujala score (r=–0.67, p=0.22). Conclusion: The proposed signal-based VAG score was demonstrated to be preliminarily able to better assess the patellofemoral chondropathy non-invasively, compared to the Kujala score, which is questionnaire-based. Keywords: Chondropathy; Vibroarthrographic signal; SFA score; SFA category; Kujala score

Vibroarthrographic signals for the low-cost and computationally efficient classification of aging and healthy knees

Knee disorders are a common but easily overlooked disease and are often caused by natural or early aging. However, the aging process is difficult for the patient to self-diagnose until it deteriorates to osteoarthritis (OA). Vibroarthrographic (VAG) signals make the aging process visible and may contribute to reducing the incidence of disability caused by delayed medical diagnosis. Recently, although many studies have used VAG signals to diagnose knee joint disease, they have not yet been used in the diagnosis of knee aging. In this paper, a method is proposed to observe the aging process of the knee through VAG signal analysis. The accuracy of classifying naturally aging and healthy knees with this method is 0.9275, and the area under the receiver operating characteristic curve is 0.8928. The paper has filled the gap of using VAG signal to diagnose the natural knee aging process and laid the foundation for the non-invasive diagnosis of early knee aging.

Screening of knee-joint vibroarthrographic signals using time and spectral domain features

It has been determined that variations in vibroarthrographic (VAG) signal characteristics have a direct association with various diseases of the knee joints including degenerative and traumatic etiology. Several studies have proposed temporal and time-frequency parameters for the analysis and classification of VAG signals; however, very few statistical modelling methods have been explored to analyze distinctions in the spectral domain characteristics of VAG signals. The aim of our work is to classify VAG signals into normal and abnormal thus portraying knee health condition by proposing a novel set of features. Spectral domain transformations of VAG signals were derived using the short-time Fourier transform (STFT) approach to extract statistical characteristic features from VAG signals. Several time-domain statistical features; the mean and variance values of spectral domain features namely – spectral mean, peak, skewness, kurtosis, flux, and slope were utilized to express VAG signal fluctuations. Further, mutual information test, ANOVA F-test, and Chi-square test were used to select significant features and assess their effectiveness. Random forest, Gaussian Naive Bayes, and support vector machine (SVM) algorithms were used to perform signal pattern classification. The results showed that subjects with knee joint abnormalities possess higher values of spectral flux, spectral slope, and shape factor than healthy subjects. An overall classification accuracy of 89.23%, sensitivity of 87.09% and specificity of 91.18% were obtained using the random forest classifier trained with features selected by the mutual information test. The proposed system could effectively screen almost 94% of subjects accurately based on area under the ROC curve.

A novel automatic Knee Osteoarthritis detection method based on vibroarthrographic signals

Knee Osteoarthritis (KOA) is a common and chronic degenerative joint disease. Comparing with the traditional examinations (e.g., X-ray, MRI), the vibroarthrographic (VAG) examination, which is a low-costly, atraumatic, and at-home way, may open up new alternatives to KOA detection in clinic. However, it is a hard problem for doctors to evaluate the condition of KOA patients by visually detecting VAG signal due to the very limited understanding of pathological information included in VAG signals. Originated from this, we focus on exploring a reliable KOA auxiliary diagnostic method using VAG signals. In this paper, a new feature extraction method is first proposed, where the kernel-radius-based feature (KR-F) and statistic-based feature (S-F) are extracted respectively in the transient phase space of VAG signal. Furthermore, two features are integrated in the feature-fusion level (KR-S-FF), and then fed into the back propagation neural network (BPNN) to complete the KOA detection automatically. Finally, the proposed automatic KOA detection method is verified in a clinical VAG dataset, which is collected from one hospital in Xi’an, China. Simulation results convey that the proposed method gives the high detection accuracy of 98.2 % with sensitivity of 97.9 % and specificity of 98.5 %, showing that it may provide an effective non-invasive tool for KOA disorders.

Vibroarthrographic Signal Spectral Features in 5-Class Knee Joint Classification.

Vibroarthrography (VAG) is a non-invasive and potentially widely available method supporting the joint diagnosis process. This research was conducted using VAG signals classified to five different condition classes: three stages of chondromalacia patellae, osteoarthritis, and control group (healthy knee joint). Ten new spectral features were proposed, distinguishing not only neighboring classes, but every class combination. Additionally, Frequency Range Maps were proposed as the frequency feature extraction visualization method. The results were compared to state-of-the-art frequency features using the Bhattacharyya coefficient and the set of ten different classification algorithms. All methods evaluating proposed features indicated the superiority of the new features compared to the state-of-the-art. In terms of Bhattacharyya coefficient, newly proposed features proved to be over 25% better, and the classification accuracy was on average 9% better.

Post-processing algorithm for removing soft-tissue movement artifacts from vibroarthrographic knee-joint signal.

Vibroarthrographic (VAG) signals are sounds or vibrations caused when a knee joint is flexed or stretched. VAG signal collection is noninvasive and can be performed using an accelerometer or microphone attached to the skin. However, the sensor attached to the skin will move with the soft tissue caused by flexion and extension, causing the baseline of the VAG signal to drift. We call these interferences soft tissue movement artifacts (STMAs). In this study, an algorithm is proposed to filter out STMAs. We compare the proposed method's results with noises collected by an accelerometer. The noise reduction effect is evaluated, revealing an 11.85% increase in the peak signal-to-noise ratio and a 28.18% increase in signal-to-noise ratio compared with the case in which STMA noise was not removed.Clinical Relevance-This study focuses on a proposed post-processing method that can remove soft tissue movement artifacts that cause baseline wander and could thus improve the accuracy of clinical applications of VAG signals.

Discrimination of knee osteoarthritis patients from asymptomatic individuals based on pain sensitivity and knee vibroarthrographic recordings

Objective: Recording of knee vibroarthrographic (VAG) activity during activities of daily living (ADL) can contribute to diagnose knee osteoarthritis (KOA). However, classifying KOA patients based on knee VAG during ADL has been an elusive problem not related to knee pain. Therefore, the aims of this study was to classify KOA patients based on (1) VAG during ADL and (2) knee pain sensitivity and then compare their results. Approach: The experimental procedure consisted of the recording of VAG signals during four ADLs (over-ground gait, stairs descent, stairs ascent and sit-to-stand) from eight patellar and peri-patellar locations in 20 KOA and 20 asymptomatic participants. Pressure pain thresholds (PPT) were obtained from eight locations around the knee joint to quantify pain sensitivity. A random forest classifier was utilized to identify KOA patients based on VAG signal features and PPTs. The most important features contributing to the classification accuracy were determined. The KOA patients participated in a second identical experimental session to examine the day-to-day reproducibility. Main results: The participants were classified with accuracy of 90%, 70%, 64% and 82% during over-ground gait, stairs descent, stairs ascent and sit to stand, respectively. However, the accuracy of the classifier was reduced by about 10%–25% due to a systematic bias in the extracted features across days. Features of the VAG signals in time and frequency domains as well as nonlinear features were found importantly contributing towards the classification accuracy. The VAG features extracted from the lateral side of the knee was found to be more informative than other locations. The classification based on PPT reached 77%. Medial and proximal knee PPT points contributed to the classification accuracy. Significance: This study showed that using multichannel VAG signals to identify KOA patients allows better accuracy than the use of PPTs. However, VAG setup must be standardized to avoid day-to-day bias.

Effects of Immobilization and Re-Mobilization on Knee Joint Arthrokinematic Motion Quality.

Background: Knee immobilization is a common intervention for patients with traumatic injuries. However, it usually leads to biomechanical/morphological disturbances of articular tissues. These changes may contribute to declining kinetic friction-related quality of arthrokinematics; however, this phenomenon has not been analyzed in vivo and remains unrecognized. Thus, the aim of the present study is to investigate the effect of immobilization and subsequent re-mobilization on the quality of arthrokinematics within the patellofemoral joint, analyzed by vibroarthrography (VAG). Methods: Thirty-four patients after 6-weeks of knee immobilization and 37 controls were analyzed. The (VAG) signals were collected during knee flexion/extension using an accelerometer. Patients were tested on the first and last day of the 2-week rehabilitation program. Results: Immobilized knees were characterized by significantly higher values of all VAG parameters when compared to controls (p < 0.001) on the first day. After 2 weeks, the participants in the rehabilitation program that had immobilized knees showed significant improvement in all measurements compared to the baseline condition, p < 0.05. However, patients did not return to normal VAG parameters compared to controls. Conclusion: Immobilization-related changes within the knee cause impairments of arthrokinematic function reflected in VAG signal patterns. The alterations in joint motion after 6 weeks of immobilization may be partially reversible; however, the 2-week physiotherapy program is not sufficient for full recovery.

A telehealth system framework for assessing knee-joint conditions using vibroarthrographic signals

Hectic lifestyle coupled sometimes with deficient choices in food and exercise, has led to an increase in the number of individuals being diagnosed with joint generation disorders. In addition, a substantial number of individuals in specialized occupations (such as army) or sports suffer from joint related injuries, thereby leading to surgical replacements and in some severe cases as disability. Standard methods of detecting joint disorders or disabilities include arthroscopy and arthrogram. Vibroarthrography (VAG) is a non-invasive technique wherein the specialist listens to sounds generated from joint movements recorded using actigraphs. In this study, we propose a system which encodes, analyses and segments actigraphy-based VAG data for assessing the severity of cartilage degeneration, and highlighting instances of activity which causes crackling sounds during limb movements. The proposed system encapsulates IoMT (Internet of Medical Things) requirements by providing efficient data compression and analysis at the source. Using an actigraphy dataset from 89 participants, our experiments yielded that the system is able to compress the actigraphy data into 3-bits per sample, thereby reducing the signal size by about 88%, without losing any vital limb movement information. This has been further validated using a simple pattern classification of 3-bit quantized VAG data from participants with healthy and unhealthy knee joints. The method, which yielded a recognition accuracy of 84.6%, as compared to raw VAG data which yielded 80% accurate results. In addition, the proposed adaptive segmentation scheme in the system, leverages the 3-bit encoding by correctly identifying 90% of the segments of interest from each VAG signal.

Dynamic Analysis of Vibration, Muscle Firing, and Force as a Novel Model for Non-Invasive Assessment of Joint Disruption in the knee: A Multiple Case Report

We present a new method for understanding knee pathology through non-invasive techniques. The combination of electromyography (EMG), vibroarthrographic (VAG), and force analysis in proposed to examine the force transfer between unhealthy and healthy knees. A multiple case report is presented to demonstrate the technique and its potential application for future study. The comparison of four individuals’ knee characteristics will be explained using this innovative methodology.

Knee joint vibroarthrography of asymptomatic subjects during loaded flexion-extension movements.

Vibroarthrography (VAG) has been proposed as a diagnostic tool for knee joint disorders. The methodological aspects of VAG assessments, such as sensor placement and outcomes measures have not been thoroughly investigated. The aim of this study was to investigate the effects of sensor placement and the effect of the knee extension and flexion movements using different added loads on six parameters extracted from VAGs of asymptomatic subjects. We hypothesized non-uniform distribution of VAGs during knee movement resulting from the underlying biomechanical features. Linear mixed model statistical analysis was carried out. Interpolation maps of the parameters were created to illustrate their spatial distribution over the knee joint. All parameters were significantly related to load, movement type, and location. There were significant load × movement type and location × movement type interaction for ARV and MPF, respectively. Further, the topographical maps of the six VAG parameters differed underlining the complementary information of the six VAG outcomes. The non-uniformity of the spatial distribution, movement type, and the knee joint load dependency of the investigate parameters open for new uses of VAG for joint diagnostic purposes including activities of daily living. Graphical abstract Visual summary of the main findings of the study. This figure shows that knee flexion-extension movement with various added external loads (0-1.25-2.5-5kg) were investigated, vibroarthrogaphic (VAG) signals from eight locations over the knee were recorded, linear and nonlinear analyses were conducted revealing non-uniformity of the spatial distribution of the VAGs.

CLASSIFICATION OF NORMAL AND KNEE JOINT DISORDER VIBROARTHROGRAPHIC SIGNALS USING MULTIFRACTALS AND SUPPORT VECTOR MACHINES

The development of reliable Computer Aided Diagnosis (CAD) systems would help in the early detection of Knee Joint Disorder (KJD). In this work, normal and KJD vibroarthrographic (VAG) signals are classified using multifractals and Support Vector Machines (SVM). Multifractal dimension [Formula: see text] is calculated from the VAG signals for various [Formula: see text]-values ([Formula: see text]). Geometrical features are calculated from the multifractal spectrum. The dimension of the feature set is reduced using Principal Component Analysis (PCA). The significant features obtained from the multifractal spectrum are fed as the input to the SVM classifier. The accuracy of the classifier is analyzed using kernels such as linear, quadratic, polynomial and Radial Basis Functions (RBF). The results suggest that VAG signals exhibits the multifractal property. The fluctuations in the normal and abnormal signals are well predicted in small scales of segments of time series. The features such as [Formula: see text] and Mean[Formula: see text] are high in abnormal VAG signals. These features give statistically significant values in differentiating the normal and abnormal subjects ([Formula: see text]). The area under the Receiver Operating Characteristic (ROC) curve is high for polynomial function (0.98). The SVM classifier with polynomial function gives 92.13% of accuracy in differentiating the normal and abnormal subjects. The calculation of multifractal spectrum and geometrical features from VAG signals requires optimization of few parameters, easy to compute, computationally inexpensive, and less time consuming. Hence, the CAD system seems to be clinically significant for the classification of normal and KJD subjects.