M.Sc. Engg.
http://hdl.handle.net/20.500.12228/54
2024-03-29T10:28:07ZPrediction on Ischemic Heart Disease using Machine Learning Approaches
http://hdl.handle.net/20.500.12228/827
Prediction on Ischemic Heart Disease using Machine Learning Approaches
Raihan, M.
Ischemic heart disease (IHD) is a terrible experience that occurs when the flow of blood
severely reduced or cut off due to plaque deposited on the inner wall of arteries that brings
oxygen to the heart muscle, leads to the ischemic heart attack (IHA). Atherosclerosis i.e.
plaque deposition on the inner wall of arteries is a silent process, has no critical symptoms
to get a warning before IHD. For this reason, early detection is very important for the proper
management of patients prone to IHD. In this thesis work, it was tried to predict IHD on the
basis of patient history, symptoms and pathological findings of patients with heart disease
using computational intelligence. Total 506 patient’s data with a maximum of 151 features
including historic, symptomatic and pathologic findings were collected from AFC Fortis
Escort Heart Institute, Khulna, Bangladesh. First, it was tried to identify the significant risk
factors of IHD i.e. the features which are significantly correlated with IHD by applying
different feature selection techniques. Then IHD was predicted using significant risk factors
by applying different classifier algorithms. The significant risk factors of IHD were
determined by using Chi-Square correlation, Ranking the features based on information gain
and Best First Search techniques. Among 151 collected features only 28 features showed
high correlations with IHD based on 0.05 significance level and information gain 1% or
above. 10-fold cross-validation technique was applied with different classification
algorithms e.g. Artificial Neural Network (ANN), Bagging, Logistic Regression, and
Random Forest to predict IHD using the most significant 28 risk factors. IHD prediction
accuracy was observed ranges from 95.85% to 97.63% with different classifier algorithm.
Random Forest showed the best prediction performance with an accuracy of 97.63%. The
same processing technique and classification algorithms were applied to the Cleveland
hospital dataset to validate our prediction approach. The observed IHD prediction accuracy
was 80.46-83.77% without applying the proposed processing techniques, but the accuracy
degraded to 79.80-81.46% applying the proposed processing techniques. The Cleveland
hospital data contains 303 patients’ data with only 13 features whereas the collected dataset
contains 506 patient’s data with 28 nicely correlated IHD risk factors. This is why the
proposed method is not suitably applicable to Cleveland dataset.
This thesis is submitted to the Department of Biomedical Engineering, Khulna University of Engineering & Technology in partial fulfillment of the requirements for the degree of Master of Science in Biomedical Engineering, October 2019.; Cataloged from PDF Version of Thesis.; Includes bibliographical references (pages 58-69).
2019-10-01T00:00:00ZDevelopment of Multiple Class Human-Computer Interaction System using Machine Learning Algorithm for Eyeball Movement
http://hdl.handle.net/20.500.12228/759
Development of Multiple Class Human-Computer Interaction System using Machine Learning Algorithm for Eyeball Movement
Chowdhury, Mubtasim Rafid
Modern technologies in the field of Biomedical Engineering are flourishing astonishingly in recent times. Human-computer interaction (HCI) is one of the newest additions in this field. It is the study of the way people interact with computers and how the computers are or are not developed for interacting with human successfully. Electrooculography (EOG) machine can be used as HCI device. It is a technique for measuring the corneo-retinal standing potential which is present between the front and the back of the human eye. Pairs of electrodes are generally attached either above and below the eye or to the left and right of the eye to detect the eye movement. A potential difference occurs between the electrodes. Considering that the resting potential is constant, the recorded potential is a measure of the eye's position. EOG device can pick up these resting potentials while moving the eyeball in different directions. Data classification is important for HCI systems. It is to identify a new observation which belongs to a set of categories. Classification is done based on a training dataset having observations whose category membership is familiar. Various algorithms can be used to classify bio-signal data like ECG, EEG and EOG. These algorithms are called machine learning algorithm. It is a set of mathematical approaches to teaching computers to train based on large amount of data without step by step human instruction. In this research, EOG data are classified using machine learning algorithm to develop a multiple class HCI system.
EOG data for different directional eyeball movement is acquired with the help of Biopac MP3X Acquisition unit. By placing the disposable surface electrodes on the right position of the skull and connecting all the leads and wires to the proper channel, the setup is ready to pick up the EOG data. Subjects are instructed to follow the LED sequence in the navigational setup. The data of 7 subjects aged between 22 to 48 years are taken for this experiment. The data is then saved using Biopac Student Lab Software and then preprocessed to prepare an EOG dataset for classification. With Weka 3.9.2, the classification procedure is done on the prepared dataset. Six classification algorithms i.e. naïve bayes, support vector machine, logistic regression, k-nearest neighbor, random forest and bagging are applied on the dataset. Comparison is shown among the algorithms based on different parameters. In the EOG dataset, features are also added which can be correlated with the classes. This correlation method is performed in IBM SPSS Statistics 25 software to find the most significant features related to the class.
From the classification result, the accuracy of the different classifiers are obtained. The accuracy of Naïve Bayes is 30.7692%, SVM is 30.7692%, Logistic regression is 53.8462%, KNN is 7.6923%, Random forest is 84.61% and Bagging is 92.31% respectively. From the comparison among the classifiers based on different parameters, bagging has the highest and KNN has the lowest accuracy among them. The proposed method is then compared with other researches where it is seen that other methods applied only two or three algorithms but in the proposed method six machine learning algorithms are used. It is observed that bagging is more suiTable algorithm for EOG data than other algorithms used in the mentioned related works. As for the correlation, only the chi-square test is performed as Fisher’s exact test can be performed for 2x2 matrix whereas there are 9 classes in the dataset. From the chi-square test result, it is seen that the mean of channel 1 (horizontal channel) and channel 2 (vertical channel) used to acquire EOG data for eyeball movement are the most significant features. These two featuress are directly related to the classes.
This thesis is submitted to the Department of Biomedical Engineering, Khulna University of Engineering & Technology in partial fulfillment of the requirements for the degree of Master of Science in Biomedical Engineering, August, 2019.; Cataloged from PDF Version of Thesis.; Includes bibliographical references (pages 97-104).
2019-08-01T00:00:00ZIn Silico Characterization and Homology Modeling of Histamine Receptors
http://hdl.handle.net/20.500.12228/509
In Silico Characterization and Homology Modeling of Histamine Receptors
Zobayer, Md. Nayem
Histamine plays vital role in molecular mechanism of allergic reactions. Therefore, characterization and homology modeling of Histamine receptor is of great importance to design effective vaccines. In this thesis, different methods are applied to analyse biomolecular features of histamine receptors and design best models of these receptors. In addition to this, the study tried to identify potential B cell and T cell epitope based vaccine of an allergen and consequently, emphasized on to develop a B cell prediction tool. Identified four histamine receptors, such as Histamine H1, Histamine H2, Histamine H3 and Histamine H4 have been analysed through ProtParam to extract physiochemical properties and ClastalW algorithm has been applied to identify conserved regions. Motif and Transmembrane regions have been identified through MEME suit and TMHMM servers, respectively. For homology modeling, I-tasser has been used and generated models have been validated through RAMPAGE, ERRAT and PROCHECK. Targeted api m3 allergen then rendered through self-optimized prediction method with alignment for physiochemical feature extraction. NetCTL 1.2 has been applied to identify preliminary T cell epitope candidates and then scrutinized by Stabilized Matrix Base Method, relative to IC50 values. Predicted T cell epitopes have been further analysed for conservancy and population coverage via IEDB tools. B cell epitopes of api m3 allergen have been predicted through, BCPREDS, ABCpred, Bepipred and Bcepred. In addition, classifier based single interface B cell epitope prediction and/or validation tool has been developed through establishing efficient MATLAB algorithms to classify beta turn regions, hydrophilic regions, surface accessible regions and antigenic regions. Lastly, with superimposing graphical representation of these four criteria in a single interface graph plotted to identify B cell epitopes via this tool. Extracted results denotes that, Histamine receptors possess molecular weight around 55.7 KDa, theoretical pI 9.33-9.62, instability index 34.93-47.00, aliphatic index (AI) was above 90 and the receptors were hydrophobic except histamine H1 receptor. Moderately conserved region was found in 75-94 amino acid position. A profound motif has been identified from 84-149 amino acid position for four histamine receptors with significantly lower E-value. It has been identified that, these receptors are seven pass transmembrane protein and a gap between transmembrane helix number five and six was found in each histamine receptor except Histamine H2 receptor, which can be potential drug target candidate. Generated 3D models have been passed through every spheres of validation. Api m3 allergen has been found relative thermostable nature and only 10.46% of the overall secondary structure consisted of beta turn region. Five MHC class I T-cell epitopes were identified and scrutinized and YTEESVSAL found out as the best epitope. For MHC class II T-cell epitopes YPKDPYLYYDFYPLE and GGPLLRIFTKHMLDV have been found as most prominent T-cell epitopes of api m3 allergen. This study also revealed that, GDRIPDEKN and PHVPEYSSS, as the most effective B-cell epitopes of api m3. The proposed tool efficiently identified B cell epitopes and provided result in a single interface. The tool can aid in B cell research and vaccine development. Finally, the suggested potential drug targets can be applied in designing more sustainable antihistamines and relevant drugs in treating allergic diseases. Predicted T-cell and B-cell epitopes of api m3 allergen could help the researchers to test these vaccines further for immunoreactivity applying in vivo analysis. As still there is no report of T-cell and B-cell epitopes of Apis mellifera, this study can be the pioneer in finding effective vaccine against allergens of honeybee. This research also predicted potential B cell epitope regions from an antigenic protein. The most exciting feature of this part of the study is, it presents results of potential B cell epitopes on a single interface, so that, researchers don‘t need to search for every feature (e.g., hydrophilicity, antigenicity, beta turn, surface accessibility etc.) separately. Finally, the study can certainly aid in B cell epitope-based vaccine design research.
This thesis is submitted to the Department of Biomedical Engineering, Khulna University of Engineering & Technology in partial fulfillment of the requirements for the degree of Masters of Science in Biomedical Engineering, February 2019.; Cataloged from PDF Version of Thesis.; Includes bibliographical references (pages 75-86).
2019-02-01T00:00:00ZClassification Accuracy Enhancement of fNIR based Imagery Movement by Modified Common Spatial Pattern
http://hdl.handle.net/20.500.12228/500
Classification Accuracy Enhancement of fNIR based Imagery Movement by Modified Common Spatial Pattern
Kabir, Md. Faisal
Motor imagery event classification from functional near-infrared spectroscopy (fNIRS) is one of the most interesting problems of current brain-computer interfaces (BCIs) challenges because it needs no additional data guiding visual or listening protocol. A vital step of the fNIRS signal classification by machine learning approach is feature extraction. The feature extraction from multiple channel fNIRS signal is always challenging due to its high dimensionality. There exist several conventional feature extraction procedures like principal component analysis, nonlinear principal component analysis, independent component analysis, norm analysis, spectral norm analysis, etc. This research work studies such existing feature extraction method to classify the hand movement events of fNIRS signals. The accuracies of these methods have been found less than the expectation. Therefore, some more accurate method is needed. In this regard, usually common spatial pattern (CSP) is used to reduce the dimensional reduction and improving the classification accuracy. The conventional CSP method can be proven also ineffective for the motor imagery fNIRS signal due to its high level of trial to trial variations. The present research work proposes an algorithm named by standardized common spatial pattern (SCSP) based feature extraction method for fNIRS based motor imagery classification which can perform well in the context of the trial to trial significant variation. The classification results corresponding to the proposed feature extraction method reveal that the proposed SCSP algorithm outperformed the conventional CSP method and channel-wise method for classifying the two motor imagery event classifications. For classification accuracy measurement, four well-known classifiers: artificial neural network (ANN), k-nearest neighbor (kNN), support vector machine (SVM), and linear discriminant analysis (LDA) have been used. We have utilized both the fNIRS data of oxidized hemoglobin (HbO) and deoxidized hemoglobin (HbR) for classifying the motor imagery fNIRS data with the conventional and proposed methods. From the comparisons, we have found that in both cases, the proposed method outperforms the conventional methods in the context of the classification accuracies. To validate the classification accuracies, the sensitivities and specificities of the classifier are also calculated in this work. We believe that the proposed SCSP method will contribute in the field of feature extraction method for other types of fNIRS based BCI system, effectively. In addition, this method may be applied for the feature extraction to the other multidimensional signals so far.
This thesis is submitted to the Department of Biomedical Engineering, Khulna University of Engineering & Technology in partial fulfillment of the requirements for the degree of Master of Science in Engineering in Biomedical Engineering, January 2019.; Cataloged from PDF Version of Thesis.; Includes bibliographical references in each chapter.
2019-01-01T00:00:00Z