MyMedR

Displaying publications 1 - 20 of 267 in total

Abstract:

Sort:

Fulltext Using SVMs for classification of cross-document relationships

Kumar, Yogan Jaya, Naomie Salim, Ahmed Hamza Osman, Abuobieda, Albaraa

Pertanika Journal of Science & Technology, 2013;21(1):239-246.
MyJurnal

Cross-document Structure Theory (CST) has recently been proposed to facilitate tasks related to multidocument analysis. Classifying and identifying the CST relationships between sentences across topically related documents have since been proven as necessary. However, there have not been sufficient studies presented in literature to automatically identify these CST relationships. In this study, a supervised machine learning technique, i.e. Support Vector Machines (SVMs), was applied to identify four types of CST relationships, namely “Identity”, “Overlap”, “Subsumption”, and “Description” on the datasets obtained from CSTBank corpus. The performance of the SVMs classification was measured using Precision, Recall and F-measure. In addition, the results obtained using SVMs were also compared with those from the previous literature using boosting classification algorithm. It was found that SVMs yielded better results in classifying the four CST relationships.

Matched MeSH terms: Supervised Machine Learning
Fulltext Chemical named entities recognition: a review on approaches and applications

Eltyeb S, Salim N

J Cheminform, 2014;6:17.
PMID: 24834132 DOI: 10.1186/1758-2946-6-17

The rapid increase in the flow rate of published digital information in all disciplines has resulted in a pressing need for techniques that can simplify the use of this information. The chemistry literature is very rich with information about chemical entities. Extracting molecules and their related properties and activities from the scientific literature to "text mine" these extracted data and determine contextual relationships helps research scientists, particularly those in drug development. One of the most important challenges in chemical text mining is the recognition of chemical entities mentioned in the texts. In this review, the authors briefly introduce the fundamental concepts of chemical literature mining, the textual contents of chemical documents, and the methods of naming chemicals in documents. We sketch out dictionary-based, rule-based and machine learning, as well as hybrid chemical named entity recognition approaches with their applied solutions. We end with an outlook on the pros and cons of these approaches and the types of chemical entities extracted.

Matched MeSH terms: Machine Learning
Multivariate relationship modeling using nested fuzzy cognitive map

Motlagh O, Papageorgiou E, Tang S, Zamberi Jamaludin

Sains Malaysiana, 2014;43:1781-1790.

Soft computing is an alternative to hard and classic math models especially when it comes to uncertain and incomplete data. This includes regression and relationship modeling of highly interrelated variables with applications in curve fitting, interpolation, classification, supervised learning, generalization, unsupervised learning and forecast. Fuzzy cognitive map (FCM) is a recurrent neural structure that encompasses all possible connections including relationships among inputs, inputs to outputs and feedbacks. This article examines a new methods for nonlinear multivariate regression using fuzzy cognitive map. The main contribution is the application of nested FCM structure to define edge weights in form of meaningful functions rather than crisp values. There are example cases in this article which serve as a platform to modelling even more complex engineering systems. The obtained results, analysis and comparison with similar techniques are included to show the robustness and accuracy of the developed method in multivariate regression, along with future lines of research.

Matched MeSH terms: Supervised Machine Learning; Unsupervised Machine Learning
Feature extraction and classification for EEG signals using wavelet transform and machine learning techniques

Amin HU, Malik AS, Ahmad RF, Badruddin N, Kamel N, Hussain M, et al.

Australas Phys Eng Sci Med, 2015 Mar;38(1):139-49.
PMID: 25649845 DOI: 10.1007/s13246-015-0333-x

This paper describes a discrete wavelet transform-based feature extraction scheme for the classification of EEG signals. In this scheme, the discrete wavelet transform is applied on EEG signals and the relative wavelet energy is calculated in terms of detailed coefficients and the approximation coefficients of the last decomposition level. The extracted relative wavelet energy features are passed to classifiers for the classification purpose. The EEG dataset employed for the validation of the proposed method consisted of two classes: (1) the EEG signals recorded during the complex cognitive task--Raven's advance progressive metric test and (2) the EEG signals recorded in rest condition--eyes open. The performance of four different classifiers was evaluated with four performance measures, i.e., accuracy, sensitivity, specificity and precision values. The accuracy was achieved above 98 % by the support vector machine, multi-layer perceptron and the K-nearest neighbor classifiers with approximation (A4) and detailed coefficients (D4), which represent the frequency range of 0.53-3.06 and 3.06-6.12 Hz, respectively. The findings of this study demonstrated that the proposed feature extraction approach has the potential to classify the EEG signals recorded during a complex cognitive task by achieving a high accuracy rate.

Matched MeSH terms: Machine Learning*
Prediction of recombinant protein overexpression in Escherichia coli using a machine learning based model (RPOLP)

Habibi N, Norouzi A, Mohd Hashim SZ, Shamsir MS, Samian R

Comput Biol Med, 2015 Nov 1;66:330-6.
PMID: 26476414 DOI: 10.1016/j.compbiomed.2015.09.015

Recombinant protein overexpression, an important biotechnological process, is ruled by complex biological rules which are mostly unknown, is in need of an intelligent algorithm so as to avoid resource-intensive lab-based trial and error experiments in order to determine the expression level of the recombinant protein. The purpose of this study is to propose a predictive model to estimate the level of recombinant protein overexpression for the first time in the literature using a machine learning approach based on the sequence, expression vector, and expression host. The expression host was confined to Escherichia coli which is the most popular bacterial host to overexpress recombinant proteins. To provide a handle to the problem, the overexpression level was categorized as low, medium and high. A set of features which were likely to affect the overexpression level was generated based on the known facts (e.g. gene length) and knowledge gathered from related literature. Then, a representative sub-set of features generated in the previous objective was determined using feature selection techniques. Finally a predictive model was developed using random forest classifier which was able to adequately classify the multi-class imbalanced small dataset constructed. The result showed that the predictive model provided a promising accuracy of 80% on average, in estimating the overexpression level of a recombinant protein.

Matched MeSH terms: Machine Learning
Fulltext A novel clinical decision support system using improved adaptive genetic algorithm for the assessment of fetal well-being

Ravindran S, Jambek AB, Muthusamy H, Neoh SC

Comput Math Methods Med, 2015;2015:283532.
PMID: 25793009 DOI: 10.1155/2015/283532

A novel clinical decision support system is proposed in this paper for evaluating the fetal well-being from the cardiotocogram (CTG) dataset through an Improved Adaptive Genetic Algorithm (IAGA) and Extreme Learning Machine (ELM). IAGA employs a new scaling technique (called sigma scaling) to avoid premature convergence and applies adaptive crossover and mutation techniques with masking concepts to enhance population diversity. Also, this search algorithm utilizes three different fitness functions (two single objective fitness functions and multi-objective fitness function) to assess its performance. The classification results unfold that promising classification accuracy of 94% is obtained with an optimal feature subset using IAGA. Also, the classification results are compared with those of other Feature Reduction techniques to substantiate its exhaustive search towards the global optimum. Besides, five other benchmark datasets are used to gauge the strength of the proposed IAGA algorithm.

Matched MeSH terms: Machine Learning
Study on the impact of open and closed book formative examinations on pharmacy students' performance, perception, and learning approach

Ramamurthy S, Meng Er H, Nadarajah VD, Pook PCK

Curr Pharm Teach Learn, 2016 03 21;8(3):364-374.
PMID: 30070247 DOI: 10.1016/j.cptl.2016.02.017

OBJECTIVES: To study the impact of open and closed book formative examinations on pharmacy students' learning approach and also to assess their performance and perception about open book (OB) and closed book (CB) systems of examination.
METHODS: A crossover study was conducted among Year 1 and Year 2 pharmacy students. Students were invited to participate voluntarily for one OB and one CB online formative test in a chemistry module in each year. Evaluation of their learning approach and perception of the OB and CB systems of examination was conducted using Deep Information Processing (DIP) questionnaire and Student Perception questionnaire respectively. The mean performance scores of OB and CB examinations were compared.
RESULTS: Analysis of DIP scores showed that there was no significant difference (p > 0.05) in the learning approach adopted for the two different examination systems. However, the mean score obtained in the OB examination was significantly higher (p < 0.01) than those obtained in the CB examination. Preference was given by a majority of students for the OB examination, possibly because it was associated with lower anxiety levels, less requirement of memorization, and more problem solving.
CONCLUSION: There is no difference in deep learning approach of students, whether the format is of the OB or CB type examinations. However, the performance of students was significantly better in OB examination than CB. Hence, using OB examination along with CB examination will be useful for student learning and help them adapt to growing and changing knowledge in pharmacy education and practice.

Matched MeSH terms: Machine Learning
Statistical-learning strategies generate only modestly performing predictive models for urinary symptoms following external beam radiotherapy of the prostate: A comparison of conventional and machine-learning methods

Yahya N, Ebert MA, Bulsara M, House MJ, Kennedy A, Joseph DJ, et al.

Med Phys, 2016 May;43(5):2040.
PMID: 27147316 DOI: 10.1118/1.4944738

Given the paucity of available data concerning radiotherapy-induced urinary toxicity, it is important to ensure derivation of the most robust models with superior predictive performance. This work explores multiple statistical-learning strategies for prediction of urinary symptoms following external beam radiotherapy of the prostate.

Matched MeSH terms: Machine Learning*
Fulltext Knowledge-Based Methods To Train and Optimize Virtual Screening Ensembles

Swift RV, Jusoh SA, Offutt TL, Li ES, Amaro RE

J Chem Inf Model, 2016 05 23;56(5):830-42.
PMID: 27097522 DOI: 10.1021/acs.jcim.5b00684

Ensemble docking can be a successful virtual screening technique that addresses the innate conformational heterogeneity of macromolecular drug targets. Yet, lacking a method to identify a subset of conformational states that effectively segregates active and inactive small molecules, ensemble docking may result in the recommendation of a large number of false positives. Here, three knowledge-based methods that construct structural ensembles for virtual screening are presented. Each method selects ensembles by optimizing an objective function calculated using the receiver operating characteristic (ROC) curve: either the area under the ROC curve (AUC) or a ROC enrichment factor (EF). As the number of receptor conformations, N, becomes large, the methods differ in their asymptotic scaling. Given a set of small molecules with known activities and a collection of target conformations, the most resource intense method is guaranteed to find the optimal ensemble but scales as O(2(N)). A recursive approximation to the optimal solution scales as O(N(2)), and a more severe approximation leads to a faster method that scales linearly, O(N). The techniques are generally applicable to any system, and we demonstrate their effectiveness on the androgen nuclear hormone receptor (AR), cyclin-dependent kinase 2 (CDK2), and the peroxisome proliferator-activated receptor δ (PPAR-δ) drug targets. Conformations that consisted of a crystal structure and molecular dynamics simulation cluster centroids were used to form AR and CDK2 ensembles. Multiple available crystal structures were used to form PPAR-δ ensembles. For each target, we show that the three methods perform similarly to one another on both the training and test sets.

Matched MeSH terms: Machine Learning*
Fulltext Bioactive Molecule Prediction Using Extreme Gradient Boosting

Babajide Mustapha I, Saeed F

Molecules, 2016 Jul 28;21(8).
PMID: 27483216 DOI: 10.3390/molecules21080983

Following the explosive growth in chemical and biological data, the shift from traditional methods of drug discovery to computer-aided means has made data mining and machine learning methods integral parts of today's drug discovery process. In this paper, extreme gradient boosting (Xgboost), which is an ensemble of Classification and Regression Tree (CART) and a variant of the Gradient Boosting Machine, was investigated for the prediction of biological activity based on quantitative description of the compound's molecular structure. Seven datasets, well known in the literature were used in this paper and experimental results show that Xgboost can outperform machine learning algorithms like Random Forest (RF), Support Vector Machines (LSVM), Radial Basis Function Neural Network (RBFN) and Naïve Bayes (NB) for the prediction of biological activities. In addition to its ability to detect minority activity classes in highly imbalanced datasets, it showed remarkable performance on both high and low diversity datasets.

Matched MeSH terms: Machine Learning
A fuzzy integral method based on the ensemble of neural networks to analyze fMRI data for cognitive state classification across multiple subjects

Cacha LA, Parida S, Dehuri S, Cho SB, Poznanski RR

J Integr Neurosci, 2016 Dec;15(4):593-606.
PMID: 28093025 DOI: 10.1142/S0219635216500345

The huge number of voxels in fMRI over time poses a major challenge to for effective analysis. Fast, accurate, and reliable classifiers are required for estimating the decoding accuracy of brain activities. Although machine-learning classifiers seem promising, individual classifiers have their own limitations. To address this limitation, the present paper proposes a method based on the ensemble of neural networks to analyze fMRI data for cognitive state classification for application across multiple subjects. Similarly, the fuzzy integral (FI) approach has been employed as an efficient tool for combining different classifiers. The FI approach led to the development of a classifiers ensemble technique that performs better than any of the single classifier by reducing the misclassification, the bias, and the variance. The proposed method successfully classified the different cognitive states for multiple subjects with high accuracy of classification. Comparison of the performance improvement, while applying ensemble neural networks method, vs. that of the individual neural network strongly points toward the usefulness of the proposed method.

Matched MeSH terms: Machine Learning*
Fulltext Classification of Suncus murinus species complex (Soricidae: Crocidurinae) in Peninsular Malaysia using image analysis and machine learning approaches

Abu A, Leow LK, Ramli R, Omar H

BMC Bioinformatics, 2016 Dec 22;17(Suppl 19):505.
PMID: 28155645 DOI: 10.1186/s12859-016-1362-5

BACKGROUND: Taxonomists frequently identify specimen from various populations based on the morphological characteristics and molecular data. This study looks into another invasive process in identification of house shrew (Suncus murinus) using image analysis and machine learning approaches. Thus, an automated identification system is developed to assist and simplify this task. In this study, seven descriptors namely area, convex area, major axis length, minor axis length, perimeter, equivalent diameter and extent which are based on the shape are used as features to represent digital image of skull that consists of dorsal, lateral and jaw views for each specimen. An Artificial Neural Network (ANN) is used as classifier to classify the skulls of S. murinus based on region (northern and southern populations of Peninsular Malaysia) and sex (adult male and female). Thus, specimen classification using Training data set and identification using Testing data set were performed through two stages of ANNs.
RESULTS: At present, the classifier used has achieved an accuracy of 100% based on skulls' views. Classification and identification to regions and sexes have also attained 72.5%, 87.5% and 80.0% of accuracy for dorsal, lateral, and jaw views, respectively. This results show that the shape characteristic features used are substantial because they can differentiate the specimens based on regions and sexes up to the accuracy of 80% and above. Finally, an application was developed and can be used for the scientific community.
CONCLUSIONS: This automated system demonstrates the practicability of using computer-assisted systems in providing interesting alternative approach for quick and easy identification of unknown species.

Matched MeSH terms: Machine Learning*
Fulltext SMARTbot: A Behavioral Analysis Framework Augmented with Machine Learning to Identify Mobile Botnet Applications

Karim A, Salleh R, Khan MK

PLoS One, 2016;11(3):e0150077.
PMID: 26978523 DOI: 10.1371/journal.pone.0150077

Botnet phenomenon in smartphones is evolving with the proliferation in mobile phone technologies after leaving imperative impact on personal computers. It refers to the network of computers, laptops, mobile devices or tablets which is remotely controlled by the cybercriminals to initiate various distributed coordinated attacks including spam emails, ad-click fraud, Bitcoin mining, Distributed Denial of Service (DDoS), disseminating other malwares and much more. Likewise traditional PC based botnet, Mobile botnets have the same operational impact except the target audience is particular to smartphone users. Therefore, it is import to uncover this security issue prior to its widespread adaptation. We propose SMARTbot, a novel dynamic analysis framework augmented with machine learning techniques to automatically detect botnet binaries from malicious corpus. SMARTbot is a component based off-device behavioral analysis framework which can generate mobile botnet learning model by inducing Artificial Neural Networks' back-propagation method. Moreover, this framework can detect mobile botnet binaries with remarkable accuracy even in case of obfuscated program code. The results conclude that, a classifier model based on simple logistic regression outperform other machine learning classifier for botnet apps' detection, i.e 99.49% accuracy is achieved. Further, from manual inspection of botnet dataset we have extracted interesting trends in those applications. As an outcome of this research, a mobile botnet dataset is devised which will become the benchmark for future studies.

Matched MeSH terms: Machine Learning
Fulltext Round Randomized Learning Vector Quantization for Brain Tumor Imaging

Sheikh Abdullah SN, Bohani FA, Nayef BH, Sahran S, Al Akash O, Iqbal Hussain R, et al.

Comput Math Methods Med, 2016;2016:8603609.
PMID: 27516807 DOI: 10.1155/2016/8603609

Brain magnetic resonance imaging (MRI) classification into normal and abnormal is a critical and challenging task. Owing to that, several medical imaging classification techniques have been devised in which Learning Vector Quantization (LVQ) is amongst the potential. The main goal of this paper is to enhance the performance of LVQ technique in order to gain higher accuracy detection for brain tumor in MRIs. The classical way of selecting the winner code vector in LVQ is to measure the distance between the input vector and the codebook vectors using Euclidean distance function. In order to improve the winner selection technique, round off function is employed along with the Euclidean distance function. Moreover, in competitive learning classifiers, the fitting model is highly dependent on the class distribution. Therefore this paper proposed a multiresampling technique for which better class distribution can be achieved. This multiresampling is executed by using random selection via preclassification. The test data sample used are the brain tumor magnetic resonance images collected from Universiti Kebangsaan Malaysia Medical Center and UCI benchmark data sets. Comparative studies showed that the proposed methods with promising results are LVQ1, Multipass LVQ, Hierarchical LVQ, Multilayer Perceptron, and Radial Basis Function.

Matched MeSH terms: Machine Learning*
Fulltext Tracking-learning-detection using extreme learning machine with Haar-like features

Melisa Anak Adeh, Mohd Ibrahim Shapiai, Ayman Maliha, Muhammad Hafiz Md Zaini

Journal of Advanced Research in Applied Sciences and Engineering Technology, 2016;5(2):1-11.
MyJurnal

Nowadays, the applications of tracking moving object are commonly used in various
areas especially in computer vision applications. There are many tracking algorithms
have been introduced and they are divided into three groups which are generative
trackers, discriminative trackers and hybrid trackers. One of the methods is TrackingLearning-Detection
(TLD) framework which is an example of the hybrid trackers where
combination between the generative trackers and the discriminative trackers occur. In
TLD, the detector consists of three stages which are patch variance, ensemble classifier
and KNearest Neighbor classifier. In the second stage, the ensemble classifier depends
on simple pixel comparison hence, it is likely fail to offer a better generalization of the
appearances of the target object in the detection process. In this paper, OnlineSequential
Extreme Learning Machine (OS-ELM) was used to replace the ensemble
classifier in the TLD framework. Besides that, different types of Haar-like features were
used for the feature extraction process instead of using raw pixel value as the features.
The objectives of this study are to improve the classifier in the second stage of detector
in TLD framework by using Haar-like features as an input to the classifier and to get a
more generalized detector in TLD framework by using OS-ELM based detector. The
results showed that the proposed method performs better in Pedestrian 1 in terms of
F-measure and also offers good performance in terms of Precision in four out of six
videos.

Matched MeSH terms: Machine Learning
Fulltext Predicting probable Alzheimer's disease using linguistic deficits and biomarkers

Orimaye SO, Wong JS, Golden KJ, Wong CP, Soyiri IN

BMC Bioinformatics, 2017 Jan 14;18(1):34.
PMID: 28088191 DOI: 10.1186/s12859-016-1456-0

BACKGROUND: The manual diagnosis of neurodegenerative disorders such as Alzheimer's disease (AD) and related Dementias has been a challenge. Currently, these disorders are diagnosed using specific clinical diagnostic criteria and neuropsychological examinations. The use of several Machine Learning algorithms to build automated diagnostic models using low-level linguistic features resulting from verbal utterances could aid diagnosis of patients with probable AD from a large population. For this purpose, we developed different Machine Learning models on the DementiaBank language transcript clinical dataset, consisting of 99 patients with probable AD and 99 healthy controls.
RESULTS: Our models learned several syntactic, lexical, and n-gram linguistic biomarkers to distinguish the probable AD group from the healthy group. In contrast to the healthy group, we found that the probable AD patients had significantly less usage of syntactic components and significantly higher usage of lexical components in their language. Also, we observed a significant difference in the use of n-grams as the healthy group were able to identify and make sense of more objects in their n-grams than the probable AD group. As such, our best diagnostic model significantly distinguished the probable AD group from the healthy elderly group with a better Area Under the Receiving Operating Characteristics Curve (AUC) using the Support Vector Machines (SVM).
CONCLUSIONS: Experimental and statistical evaluations suggest that using ML algorithms for learning linguistic biomarkers from the verbal utterances of elderly individuals could help the clinical diagnosis of probable AD. We emphasise that the best ML model for predicting the disease group combines significant syntactic, lexical and top n-gram features. However, there is a need to train the diagnostic models on larger datasets, which could lead to a better AUC and clinical diagnosis of probable AD.

Matched MeSH terms: Machine Learning*
Blind source computer device identification from recorded VoIP calls for forensic investigation

Jahanirad M, Anuar NB, Wahab AWA

Forensic Sci Int, 2017 Mar;272:111-126.
PMID: 28129583 DOI: 10.1016/j.forsciint.2017.01.010

The VoIP services provide fertile ground for criminal activity, thus identifying the transmitting computer devices from recorded VoIP call may help the forensic investigator to reveal useful information. It also proves the authenticity of the call recording submitted to the court as evidence. This paper extended the previous study on the use of recorded VoIP call for blind source computer device identification. Although initial results were promising but theoretical reasoning for this is yet to be found. The study suggested computing entropy of mel-frequency cepstrum coefficients (entropy-MFCC) from near-silent segments as an intrinsic feature set that captures the device response function due to the tolerances in the electronic components of individual computer devices. By applying the supervised learning techniques of naïve Bayesian, linear logistic regression, neural networks and support vector machines to the entropy-MFCC features, state-of-the-art identification accuracy of near 99.9% has been achieved on different sets of computer devices for both call recording and microphone recording scenarios. Furthermore, unsupervised learning techniques, including simple k-means, expectation-maximization and density-based spatial clustering of applications with noise (DBSCAN) provided promising results for call recording dataset by assigning the majority of instances to their correct clusters.

Matched MeSH terms: Machine Learning
Automated diagnosis of congestive heart failure using dual tree complex wavelet transform and statistical features extracted from 2s of ECG signals

Sudarshan VK, Acharya UR, Oh SL, Adam M, Tan JH, Chua CK, et al.

Comput Biol Med, 2017 04 01;83:48-58.
PMID: 28231511 DOI: 10.1016/j.compbiomed.2017.01.019

Identification of alarming features in the electrocardiogram (ECG) signal is extremely significant for the prediction of congestive heart failure (CHF). ECG signal analysis carried out using computer-aided techniques can speed up the diagnosis process and aid in the proper management of CHF patients. Therefore, in this work, dual tree complex wavelets transform (DTCWT)-based methodology is proposed for an automated identification of ECG signals exhibiting CHF from normal. In the experiment, we have performed a DTCWT on ECG segments of 2s duration up to six levels to obtain the coefficients. From these DTCWT coefficients, statistical features are extracted and ranked using Bhattacharyya, entropy, minimum redundancy maximum relevance (mRMR), receiver-operating characteristics (ROC), Wilcoxon, t-test and reliefF methods. Ranked features are subjected to k-nearest neighbor (KNN) and decision tree (DT) classifiers for automated differentiation of CHF and normal ECG signals. We have achieved 99.86% accuracy, 99.78% sensitivity and 99.94% specificity in the identification of CHF affected ECG signals using 45 features. The proposed method is able to detect CHF patients accurately using only 2s of ECG signal length and hence providing sufficient time for the clinicians to further investigate on the severity of CHF and treatments.

Matched MeSH terms: Machine Learning*
A Critical Review for Developing Accurate and Dynamic Predictive Models Using Machine Learning Methods in Medicine and Health Care

Alanazi HO, Abdullah AH, Qureshi KN

J Med Syst, 2017 Apr;41(4):69.
PMID: 28285459 DOI: 10.1007/s10916-017-0715-6

Recently, Artificial Intelligence (AI) has been used widely in medicine and health care sector. In machine learning, the classification or prediction is a major field of AI. Today, the study of existing predictive models based on machine learning methods is extremely active. Doctors need accurate predictions for the outcomes of their patients' diseases. In addition, for accurate predictions, timing is another significant factor that influences treatment decisions. In this paper, existing predictive models in medicine and health care have critically reviewed. Furthermore, the most famous machine learning methods have explained, and the confusion between a statistical approach and machine learning has clarified. A review of related literature reveals that the predictions of existing predictive models differ even when the same dataset is used. Therefore, existing predictive models are essential, and current methods must be improved.

Matched MeSH terms: Machine Learning
Fulltext Machine learning techniques for diabetic macular edema (DME) classification on SD-OCT images

Alsaih K, Lemaitre G, Rastgoo M, Massich J, Sidibé D, Meriaudeau F

Biomed Eng Online, 2017 Jun 07;16(1):68.
PMID: 28592309 DOI: 10.1186/s12938-017-0352-9

BACKGROUND: Spectral domain optical coherence tomography (OCT) (SD-OCT) is most widely imaging equipment used in ophthalmology to detect diabetic macular edema (DME). Indeed, it offers an accurate visualization of the morphology of the retina as well as the retina layers.
METHODS: The dataset used in this study has been acquired by the Singapore Eye Research Institute (SERI), using CIRRUS TM (Carl Zeiss Meditec, Inc., Dublin, CA, USA) SD-OCT device. The dataset consists of 32 OCT volumes (16 DME and 16 normal cases). Each volume contains 128 B-scans with resolution of 1024 px × 512 px, resulting in more than 3800 images being processed. All SD-OCT volumes are read and assessed by trained graders and identified as normal or DME cases based on evaluation of retinal thickening, hard exudates, intraretinal cystoid space formation, and subretinal fluid. Within the DME sub-set, a large number of lesions has been selected to create a rather complete and diverse DME dataset. This paper presents an automatic classification framework for SD-OCT volumes in order to identify DME versus normal volumes. In this regard, a generic pipeline including pre-processing, feature detection, feature representation, and classification was investigated. More precisely, extraction of histogram of oriented gradients and local binary pattern (LBP) features within a multiresolution approach is used as well as principal component analysis (PCA) and bag of words (BoW) representations.
RESULTS AND CONCLUSION: Besides comparing individual and combined features, different representation approaches and different classifiers are evaluated. The best results are obtained for LBP[Formula: see text] vectors while represented and classified using PCA and a linear-support vector machine (SVM), leading to a sensitivity(SE) and specificity (SP) of 87.5 and 87.5%, respectively.

Matched MeSH terms: Machine Learning*

Filters

Please provide feedback to Administrator (afdal@afpm.org.my)

External Links