MyMedR

Displaying all 4 publications

Abstract:

Sort:

Binary particle swarm optimization for feature selection in detection of infants with hypothyroidism

Zabidi A, Khuan LY, Mansor W, Yassin IM, Sahak R

Annu Int Conf IEEE Eng Med Biol Soc, 2011;2011:2772-5.
PMID: 22254916 DOI: 10.1109/IEMBS.2011.6090759

Hypothyroidism in infants is caused by the insufficient production of hormones by the thyroid gland. Due to stress in the chest cavity as a result of the enlarged liver, their cry signals are unique and can be distinguished from the healthy infant cries. This study investigates the effect of feature selection with Binary Particle Swarm Optimization on the performance of MultiLayer Perceptron classifier in discriminating between the healthy infants and infants with hypothyroidism from their cry signals. The feature extraction process was performed on the Mel Frequency Cepstral coefficients. Performance of the MLP classifier was examined by varying the number of coefficients. It was found that the BPSO enhances the classification accuracy while reducing the computation load of the MLP classifier. The highest classification accuracy of 99.65% was achieved for the MLP classifier, with 36 filter banks, 5 hidden nodes and 11 BPS optimised MFC coefficients.
Optimization of MFCC parameters using Particle Swarm Optimization for diagnosis of infant hypothyroidism using Multi- Layer Perceptron

Zabidi A, Lee YK, Mansor W, Yassin IM, Sahak R

Annu Int Conf IEEE Eng Med Biol Soc, 2010;2010:1417-20.
PMID: 21096346 DOI: 10.1109/IEMBS.2010.5626712

This paper presents a new application of the Particle Swarm Optimization (PSO) algorithm to optimize Mel Frequency Cepstrum Coefficients (MFCC) parameters, in order to extract an optimal feature set for diagnosis of hypothyroidism in infants using Multi-Layer Perceptrons (MLP) neural network. MFCC features is influenced by the number of filter banks (f(b)) and the number of coefficients (n(c)) used. These parameters are critical in representation of the features as they affect the resolution and dimensionality of the features. In this paper, the PSO algorithm was used to optimize the values of f(b) and n(c). The MFCC features based on the PSO optimization were extracted from healthy and unhealthy infant cry signals and used to train MLP in the classification of hypothyroid infant cries. The results indicate that the PSO algorithm could determine the optimum combination of f(b) and n(c) that produce the best classification accuracy of the MLP.
Classification of intelligence quotient via brainwave sub-band power ratio features and artificial neural network

Jahidin AH, Megat Ali MS, Taib MN, Tahir NM, Yassin IM, Lias S

Comput Methods Programs Biomed, 2014 Apr;114(1):50-9.
PMID: 24560277 DOI: 10.1016/j.cmpb.2014.01.016

This paper elaborates on the novel intelligence assessment method using the brainwave sub-band power ratio features. The study focuses only on the left hemisphere brainwave in its relaxed state. Distinct intelligence quotient groups have been established earlier from the score of the Raven Progressive Matrices. Sub-band power ratios are calculated from energy spectral density of theta, alpha and beta frequency bands. Synthetic data have been generated to increase dataset from 50 to 120. The features are used as input to the artificial neural network. Subsequently, the brain behaviour model has been developed using an artificial neural network that is trained with optimized learning rate, momentum constant and hidden nodes. Findings indicate that the distinct intelligence quotient groups can be classified from the brainwave sub-band power ratios with 100% training and 88.89% testing accuracies.
Fulltext Estimation of kidneys and urinary bladder doses based on the region of interest in (18)fluorine-fluorodeoxyglucose positron emission tomography/computed tomography examination: a preliminary study

Mustapha FA, Bashah FAA, Yassin IM, Fathinul Fikri AS, Nordin AJ, Abdul Razak HR

Quant Imaging Med Surg, 2017 Jun;7(3):310-317.
PMID: 28811997 DOI: 10.21037/qims.2017.05.03

BACKGROUND: Kidneys and urinary bladder are common physiologic uptake sites of 18fluorine-fluorodeoxyglucose ((18)F-FDG) causing increased exposure of low energy ionizing radiation to these organs. Accurate measurement of organ dose is vital as (18)F-FDG is directly exposed to the organs. Organ dose from (18)F-FDG PET is calculated according to the injected (18)F-FDG activity with the application of dose coefficients established by International Commission on Radiological Protection (ICRP). But this dose calculation technique is not directly measured from these organs; rather it is calculated based on total injected activity of radiotracer prior to scanning. This study estimated the (18)F-FDG dose to the kidneys and urinary bladder in whole body positron emission tomography/computed tomography (PET/CT) examination by comparing dose from total injected activity of (18)F-FDG (calculated dose) and dose from organs activity based on the region of interest (ROI) (measured dose).
METHODS: Nine subjects were injected intravenously with the mean (18)F-FDG dose of 292.42 MBq prior to whole body PET/CT scanning. Kidneys and urinary bladder doses were estimated by using two approaches which are the total injected activity of (18)F-FDG and organs activity concentration of (18)F-FDG based on drawn ROI with the application of recommended dose coefficients for (18)F-FDG described in the ICRP 80 and ICRP 106.
RESULTS: The mean percentage difference between calculated dose and measured dose ranged from 98.95% to 99.29% for the kidneys based on ICRP 80 and 98.96% to 99.32% based on ICRP 106. Whilst, the mean percentage difference between calculated dose and measured dose was 97.08% and 97.27% for urinary bladder based on ICRP 80 while 96.99% and 97.28% based on ICRP 106. Whereas, the range of mean percentage difference between calculated and measured organ doses derived from ICRP 106 and ICRP 80 for kidney doses were from 17.00% to 40.00% and for urinary bladder dose was 18.46% to 18.75%.
CONCLUSIONS: There is a significant difference between calculated dose and measured dose. The use of organ activity estimation based on drawn ROI and the latest version of ICRP 106 dose coefficient should be explored deeper to obtain accurate radiation dose to patients.