Kamalraj Subramaniam

Auditory evoked potential response and hearing loss: a review.

Hypoacusis is the most prevalent sensory disability in the world and consequently, it can lead to impede speech in human beings. One best approach to tackle this issue is to conduct early and effective hearing screening test using Electroencephalogram (EEG). EEG based hearing threshold level determination is most suitable for persons who lack verbal communication and behavioral response to sound stimulation. Auditory evoked potential (AEP) is a type of EEG signal emanated from the brain scalp by an acoustical stimulus. The goal of this review is to assess the current state of knowledge in estimating the hearing threshold levels based on AEP response. AEP response reflects the auditory ability level of an individual. An intelligent hearing perception level system enables to examine and determine the functional integrity of the auditory system. Systematic evaluation of EEG based hearing perception level system predicting the hearing loss in newborns, infants and multiple handicaps will be a priority of interest for future research.

EEG based hearing perception level classification using fractal features

The primary focus of this study is to develop a hearing ability level assessment system using auditory evoked potential signals (AEP). AEP signal is a non-invasive tool that provides auditory pathway information content and its stimulated interactions with neurons. To record the hearing perception levels at different sound intensity levels, namely, 20 dB, 30 dB, 40 dB, 50 dB and 60 dB in both the ears of a normal subject, a simple AEP based hearing perception level protocol has been proposed. The detrended fluctuation analysis (DFA) has been used to estimate the fractal values of different hearing perception levels of the recorded AEP signals. The extracted fractal features were then associated to different hearing perception levels of a subject and neural network models were developed. The maximum classification accuracy of the developed neural network models for the left and right ears are observed as 78.57% and 80.71% respectively. From the classification accuracy, it has been inferred that the neural network models are able to discriminate the five distinct hearing perception levels of a normal hearing person.

A Review on Upper-Limb Myoelectric Prosthetic Control

ABSTRACT One of the most interesting topics in the field of rehabilitation is that of upper-limb myoelectric prosthetic control. It is a technique by which prostheses are controlled by means of surface electromyogram (sEMG) signals collected from remnant muscle tissues at the residual limb of an amputee. Intuitive control of multifunctional upper-limb prosthesis can be accomplished using pattern recognition (PR) of sEMG signals. In spite of the tremendous progress made in the research of the so-called mind-controlled artificial arm, none of the academic achievements has yet reached the end users. This review paper portrays the current state-of-the-art approach in sEMG pattern classification-based control, identifies the factors that hinder the clinical usability of the system and focuses on the recent research directions toward translating the academic findings into a commercially acceptable robust myoelectric prosthesis. Control strategies proposed for simultaneous and proportional control (SPC) of multiple degrees of freedom (DoFs), which is identified as the most significant barrier for the transition from laboratory to clinical practice, are discussed. Directions for future research are also briefly outlined.

HEp-2 cell classification using artificial neural network approach

Human Epithelial type-2 (HEp-2) cells are used as substrates for the detection of Anti Nuclear Antibodies (ANA) in the Indirect Immunofluorescence (IIF) test to diagnose autoimmune diseases. Pathologists in the laboratory examine the IIF slides to detect and recognize theHEp-2 cell patterns to generate the report. So, the IIF test is subjective and requires objective analysis. This paper introduces a novel algorithm for HEp-2 cell pattern recognition, which can be embedded into the CAD. This is the first time HEp-2 cell Classification system separated intermediate and positive cells before preprocessing and processed further for classification to achieve better accuracy. The design used spectral, statistical and textural features along with Artificial Neural Network (ANN) for classification. The algorithm introduced in this paper was tested using ICPR 2016 IIF HEp-2 cell dataset of frequently occurring six distinct fluorescence patterns by computer simulation. Two experiments were conducted; the results concluded that features performance linearly increases with combination. The experimental results proved that Hybrid feature set approach based HEp-2 cell classification is more reliable with 93.15% accuracy on the ICPR 2016 IIF HEp-2 cell images dataset.

EEG based hearing states detection using AR modeling techniques

In this paper, a simple method to determine the hearing threshold state of a subject using the parametric model of EEG time series signal has been investigated. The proposed autoregressive (AR) pole-tracking algorithm tracks the position of the poles and extracts the upper and lower hearing threshold factors of a subject. From the results, for abnormal hearing subjects, the hearing-threshold values are about 40–50 % higher than the normal hearing subjects. The results also show that the hearing threshold factors obtained using AR modeling clearly distinguishes the normal and abnormal hearing states across 20 subjects. The results obtained are promising and it can be used to determine the hearing-threshold state for newborns, infants, and multiple handicaps, a person who lacks verbal communication and behavioral response to the sound stimulation.

HEp-2 cell classification using binary decision tree approach

Human Epithelial Type-2 (HEp-2) cells have been accepted as substrates for antinuclear antibodies (ANA) detection by Indirect ImmunoFluorescence (IIF) method in the blood serum of a patient. In order to diagnose the autoimmune diseases, specialists observe the fluorescencepattern of HEp-2 cells by examining the IIF slides using the fluorescence microscope and then generate the report. The reliability of the ANA screening can be enhanced by a computer aided analysis (CAD), as it is of intuitive nature. In preprocessing green channel image was extracted as it contains more information, noise was removed using Weiner filter and contrast stretching was performed for better analysis. This paper proposed a binary decision tree approach for the HEp-2 cell classification, which can be embedded into CAD system used for ANA detection. This approach relied on texture descriptors, which were proved to be the effective differentiators for HEp-2 cell patterns. From the experimental result, it was observed that binary decision tree approach yields an accuracy of 97.83% on the available IIF HEp-2 cell dataset.

A machine learning approach for distinguishing hearing perception level using auditory evoked potentials

An auditory loss is one of the most common disabilities present in newborns and infants in the world. A conventional hearing screening tests applicability is limited as it requires a feedback response from the subject under test. To overcome such problems, the primary focus of this study is to develop an intelligent hearing ability level assessment system using auditory evoked potential signals (AEP). AEP signal is a non-invasive tool that can reflect the stimulated interactions with neurons along the stations of the auditory pathway. The AEP responses of fourteen normal hearing subjects to auditory stimuli (20 dB, 30 dB, 40 dB, 50 dB and 60 dB) were derived from electroencephalogram (EEG) recordings. Higuchis fractal method is applied to extract the fractal features from the recorded AEP signals. The extracted fractal features were then associated to different hearing perception levels of the subjects. Feed-forward and feedback neural networks are employed to distinguish the different hearing perception levels. The performance of the proposed intelligent hearing ability level assessment found to exceed 85% accuracy. This study indicates that AEP responses to the auditory stimuli to the normal hearing persons can predict the higher order auditory stimuli followed by the lower order auditory stimuli and consequently the state of auditory development of subjects.

EEG based estimation of hearing frequency perception by artificial neural networks

Auditory evoked potentials are a type of EEG signal emanated from the scalp of the brain by an acoustical stimulus. In this paper, auditory evoked potential (AEP) signals emanated while hearing the click-sound stimuli excited at three different frequencies were recorded. Spatio-temporal features of four distinct bands were extracted from the recorded AEP signal. The extracted features were then associated to the hearing frequency perception response of an individual and neural network models for left and right ears were developed. The maximum classification accuracy of the developed neural network model in discriminating the hearing frequency perception response of a person has been observed as 94.5 per cent.

EEG based hearing perception level estimation for normal hearing persons

Auditory evoked potential (AEP) is a type of EEG signal emanated from the scalp of the brain by an acoustical stimulus. AEP response reflects the auditory ability level of an individual. In this paper, AEP signals were recorded by stimulating repetive click-sound of 1000 Hz at different stimulus intensity levels of 25 dB, 40 dB, 50 dB and 70 dB. Spectral entropy features of four distinct bands were extracted from the recorded AEP signal. The extracted features were associated to the hearing perception level of an individual and a neural network models was developed. The maximum classification accuracy of the developed neural network model was observed as 91.4 per cent in discriminating the specified stimulus intensity levels. From the result, it is clear that a different auditory stimuli level reflects corresponding hearing perception level of a person. This study might lead to a real-time practical system for non-invasively estimating the hearing perceptional level of a person.

Human Epithelial Type-2 Cell Image Classification Using an Artificial Neural Network with Hybrid Descriptors

ABSTRACT Antinuclear antibody (ANA) testing is best performed using the indirect immunofluorescence (IIF) method with human epithelial type-2 (HEp-2) cells as the substrate. IIF is a subjective procedure in which HEp-2 patterns are analyzed manually from the microscope. Therefore, ANA test results greatly rely on the experience and expertise of pathologists. Hence, complete automation of the ANA test is required to avoid incorrect diagnoses. This paper represents an algorithm for the complex HEp-2 cell classification problem. The proposed algorithm used a small hybrid feature set that characterizes the texture and morphology of the HEp-2 cells along with artificial neural network (ANN). The hybrid features were extracted by breaking up the image into eight binary images. The proposed hybrid descriptors were more efficient than the popular co-occurrence matrix descriptor and local binary pattern descriptors for texture analysis. The proposed algorithm was evaluated on the ICPR 2016 IIF HEp-2 cell image dataset. The results indicated that the hybrid descriptor with an ANN approach achieved improved performance, with “96.8%” mean class accuracy.