MyMedR

Displaying publications 221 - 240 of 933 in total

Abstract:

Sort:

Fulltext Advocating collaborative teaching and learning in dentistry

Mas Suryalis Ahmad

Malaysian Dental Journal, 2016;39(1):1-8.
MyJurnal

Collaborative teaching is an educational approach that seeks to involve participation of teachers and learners in achieving learning goals and outcomes in an interactive manner (1). Such approach has been effective in equipping students with knowledge and/or skills via high levels of learning, while allowing interpersonal development such as teamwork, time management, as well as communication and written competencies (2, 3). (Copied from article)

Matched MeSH terms: Learning
Towards developing high-fidelity simulated learning environment training modules in audiology

Dzulkarnain AA, Rahmat S, Mohd Puzi NA, Badzis M

Med J Malaysia, 2017 02;72(1):37-45.
PMID: 28255138 MyJurnal

INTRODUCTION: This discussion paper reviews and synthesises the literature on simulated learning environment (SLE) from allied health sciences, medical and nursing in general and audiology specifically. The focus of the paper is on discussing the use of high-fidelity (HF) SLE and describing the challenges for developing a HF SLE for clinical audiology training.
METHODS: Through the review of the literature, this paper discusses seven questions, (i) What is SLE? (ii) What are the types of SLEs? (iii) How is SLE classified? (iv) What is HF SLE? (v) What types of SLEs are available in audiology and their level of fidelity? (vi) What are the components needed for developing HF SLE? (vii) What are the possible types of HF SLEs that are suitable for audiology training? Publications were identified by structured searches from three major databases PubMed, Web of Knowledge and PsychInfo and from the reference lists of relevant articles. The authors discussed and mapped the levels of fidelity of SLE audiology training modules from the literature and the learning domains involved in the clinical audiology courses.
RESULTS: The discussion paper has highlighted that most of the existing SLE audiology training modules consist of either low- or medium-fidelity types of simulators. Those components needed to achieve a HF SLE for audiology training are also highlighted.
CONCLUSION: Overall, this review recommends that the combined approach of different levels and types of SLE could be used to obtain a HF SLE training module in audiology training.

Matched MeSH terms: Learning
Arrhythmia detection using deep convolutional neural network with long duration ECG signals

Yıldırım Ö, Pławiak P, Tan RS, Acharya UR

Comput Biol Med, 2018 11 01;102:411-420.
PMID: 30245122 DOI: 10.1016/j.compbiomed.2018.09.009

This article presents a new deep learning approach for cardiac arrhythmia (17 classes) detection based on long-duration electrocardiography (ECG) signal analysis. Cardiovascular disease prevention is one of the most important tasks of any health care system as about 50 million people are at risk of heart disease in the world. Although automatic analysis of ECG signal is very popular, current methods are not satisfactory. The goal of our research was to design a new method based on deep learning to efficiently and quickly classify cardiac arrhythmias. Described research are based on 1000 ECG signal fragments from the MIT - BIH Arrhythmia database for one lead (MLII) from 45 persons. Approach based on the analysis of 10-s ECG signal fragments (not a single QRS complex) is applied (on average, 13 times less classifications/analysis). A complete end-to-end structure was designed instead of the hand-crafted feature extraction and selection used in traditional methods. Our main contribution is to design a new 1D-Convolutional Neural Network model (1D-CNN). The proposed method is 1) efficient, 2) fast (real-time classification) 3) non-complex and 4) simple to use (combined feature extraction and selection, and classification in one stage). Deep 1D-CNN achieved a recognition overall accuracy of 17 cardiac arrhythmia disorders (classes) at a level of 91.33% and classification time per single sample of 0.015 s. Compared to the current research, our results are one of the best results to date, and our solution can be implemented in mobile devices and cloud computing.

Matched MeSH terms: Machine Learning
Fulltext Integration of entrepreneurship theories towards entrepreneurship development among students in Malaysia

Mohamad Izzuan Mohd Ishar, Mohd Khata Jabor

Malaysian Journal of Social Sciences and Humanities, 2017;2(1):11-20.
MyJurnal

Entrepreneurship showed an increase in the popularity of business education, engineering education,
universities and educational institutions. All students who engage in entrepreneurial education has the
potential to develop their entrepreneurial skills and knowledge. However, the majority of
entrepreneurial education program focused on the exploitation of existing opportunities and assume
that these opportunities have been identified. Research on entrepreneurship also shows that efficiency
is often ignored or receive little attention while teaching entrepreneurship. This article was developed
to assist in improving the understanding of the concept of learning which supports entrepreneurship
and the development of entrepreneurial competence.

Matched MeSH terms: Learning
Fulltext Root-oriented words generation: an easier way towards dictionary making for the Dusunic family of languages

Cosmas Julius Abah, Wong, Jane Kong Ling, Anantha Raman Govindasamy

Malaysian Journal of Social Sciences and Humanities, 2018;3(2):95-112.
MyJurnal

Dictionary production is one of the most effective methods of preserving languages and cultures. The
Dusunic Family of Languages (DFL) in Sabah, Malaysia would have welcomed the efforts to
document their languages through dictionary production as there are still lacking of dictionary,
vocabulary and phrase books. Furthermore, more than half of the languages in DFL are unwritten.
However, making dictionary conventionally is tedious and time consuming. The Dusunic Family of
Languages which are facing extinction threats do not have the luxury of time to wait for dictionary
production via the conventional method. Hence, this study explores the use of a method called Root-
Oriented Words Generation (ROWG) which is formulated based on spelling orthography of DFL to
generate one and two-syllable words list. From the words list, root words registers were compiled
which can then be used as database for dictionary production. Findings of this study showed that
ROWG was able to generate an exhaustive word lists of DFL and compile a large volume of root
words register in DFL. Hence, this study was able to highlight the feasibility and viability of using
ROWG to produce root words register of DFL which could possibly reduce the time for dictionary
production significantly. In future studies, it is recommended that the ROWG is extended to include
more than two syllable words. This study showed the potentiality of ROWG to address the looming
demise of DFL by providing a more efficient way of compiling root words for the purpose of making a
dictionary.

Matched MeSH terms: Learning
Fulltext The advantages of using concept mapping notes in classroom teaching

Rusnani Ab Latif, Akehsan Hj. Dahlan, Zamzaliza Ab Mulud, Mohd Zarawi Mat Nor

Malaysian Journal of Medical Research, 2018;2(1):37-48.
MyJurnal

Introduction: The teacher centered approach is a teaching and learning strategy has been practiced
traditionally for long in the classroom. Through this strategy the teacher plays an important role, while the
students only act as spectators, the interaction between students and teachers is only one way. In this study, the
concept mapping notes was added in teaching and learning methods during classroom teaching. Concept
mapping is one method of teaching that encourages students to becomes independent learner, critical thinking
and competent in their work.

Methodology: This study was carried out in Kolej Kejururawatan Kubang Kerian (Kelantan), and Kolej
Kejururawatan Pulau Pinang. The respondents were selected using simple random sampling. There were 109
respondents. The respondents were given 40 minutes to develop the concept mapping notes.

Results: The activity of the students in the teaching is certainly a positive sign towards achieving their
learning. The example of variety concept maps construct done by the students are highly creative and
innovative.

Conclusion: Students should take responsibility for their own learning. However, that is a role of a educator
to choose the best teaching method to makes the learning become meaningful and effective toward the student
cognitive structure that will help them to understand the topic those were taught.

Matched MeSH terms: Learning
Fulltext "Free Hand" or Wire Guide: To the Editor

Sinha NK, Bhardwaj A

Clin Orthop Surg, 2019 12;11(4):495.
PMID: 31788175 DOI: 10.4055/cios.2019.11.4.495
Matched MeSH terms: Learning Curve
Fulltext Integration of Islamic perspectives into medical imaging curriculum: an outcome based education approach

Zainul Ibrahim Zainuddin

International Medical Journal Malaysia, 2018;17(101):23-30.
MyJurnal

This paper presents a conceptual approach to the integration of Islamic perspectives into a Medical Imaging Curriculum to the concept of Outcome-Based Education (OBE). This work is seen within the context of harmonising Islamic principles to a currently accepted concept in education. Although there have been discussions that question the concept of OBE, this paper contends that the integration can benefit from the practicality aspect of OBE. This can reduce the complexities and fatigue in addressing the integration using an educational approach that is different to that being applied to the human sciences. This paper features the main elements in OBE in the form of Islamic programme educational objectives, Islamic programme outcomes, and Islamic domain learning outcomes. The justification to use domain learning outcomes instead of course learning outcome is given. The teaching and learning strategies, as well as the assessment, are examined through a lens that serves to provide a desirable, practical and holistic model of Islamic integration. It is felt that the currently accepted teaching and assessment methodologies can be adapted for the integration exercise. This work also highlights two often overlooked elements of OBE; teacher and student characteristics. The various terminologies that describe the Islamic teacher characteristics and the differences in student learning styles and preferences are presented. Furthermore, suggestions are made to align the assessment of the integration to various taxonomies of learning, with the aim in evaluating the internalisation of the Islamic essences. This work contents that a holistic approach towards integration of Islamic perspectives into Medical Imaging curriculum can be realised.

Matched MeSH terms: Learning
Fulltext Online kiosk for improving the number of students’ entrepreneurship projects (ent300) based on science and technology

Noor Azrin Zainuddin, Shamsatuan Nahar, Norzarina Johari, Farah Suraya Md Nasrudin, Noraisyah Abdul Aziz, Nur Diana Zamani, et al.

Jurnal Inovasi Malaysia, 2018;1(2):23-36.
MyJurnal

The use of technology in teaching and learning is increasingly synonymous with the existence of multiple online platforms. Online teaching and learning guides help lecturers and students to obtain a variety of information related to their specialization in the field of study. As all UiTM students at the Diploma level are required to take and pass the Entrepreneurial Basic course (ENT300), they need to produce an entrepreneurial project as one of the course evaluation components. However, the number of science and technology based entrepreneurship projects and products are still too few based on project titles every semester. The ENT300 Kiosk Science and Technology (KENTS) was developed specifically as a guide for the students of the Faculty of Computer Science and Mathematics and has been improved by expanding its scope to the students of the Faculty of Engineering and Faculty of Science at UiTM Johor. A more global scope in KENTS provides specialized online guides for lecturers and students in the science and technology clusters. KENTS is a platform that can be used to realize the direction of higher education in Malaysia and to assist UiTM in producing holistic graduates with entrepreneurship. This online guide platform provides teaching and learning assistance through the custom business template which is categorized into two, system development and machine design. KENTS provides a search function from a list of compilations of science-based and entrepreneurial projects that help lecturers and students find entrepreneurial ideas. KENTS database is used to store student entrepreneurial project information as an e-learning platform that can be shared by lecturers and students globally.

Matched MeSH terms: Learning
Fulltext Performance of SVM with multiple kernel learning for classification tasks of imbalanced datasets

Saeed, Sana, Ong, Hong Choon

Pertanika Journal of Science & Technology, 2019;27(1):527-545.
MyJurnal

Support vector machine (SVM) is one of the most popular algorithms in machine learning
and data mining. However, its reduced efficiency is usually observed for imbalanced
datasets. To improve the performance of SVM for binary imbalanced datasets, a new scheme
based on oversampling and the hybrid algorithm were introduced. Besides the use of a
single kernel function, SVM was applied with multiple kernel learning (MKL). A weighted
linear combination was defined based on the linear kernel function, radial basis function
(RBF kernel), and sigmoid kernel function for MKL. By generating the synthetic samples
in the minority class, searching the best choices of the SVM parameters and identifying
the weights of MKL by minimizing the objective function, the improved performance of
SVM was observed. To prove the strength of the proposed scheme, an experimental study,
including noisy borderline and real imbalanced datasets was conducted. SVM was applied
with linear kernel function, RBF kernel, sigmoid kernel function and MKL on all datasets.
The performance of SVM with all kernel functions was evaluated by using sensitivity,
G Mean, and F measure. A significantly improved performance of SVM with MKL was
observed by applying the proposed scheme.

Matched MeSH terms: Machine Learning
Discrete hopfield neural network in restricted maximum k-satisfiability logic programming

Mohd Shareduwan Mohd Kasihmuddin, Saratha Sathasivam, Mohd Asyraf Mansor

Sains Malaysiana, 2018;47:1327-1335.

Maximum k-Satisfiability (MAX-kSAT) consists of the most consistent interpretation that generate the maximum number
of satisfied clauses. MAX-kSAT is an important logic representation in logic programming since not all combinatorial
problem is satisfiable in nature. This paper presents Hopfield Neural Network based on MAX-kSAT logical rule. Learning
of Hopfield Neural Network will be integrated with Wan Abdullah method and Sathasivam relaxation method to obtain
the correct final state of the neurons. The computer simulation shows that MAX-kSAT can be embedded optimally in
Hopfield Neural Network.

Matched MeSH terms: Learning
Evaluation of Three-Dimensional Heads up Ophthalmic Surgery Demonstration From the Perspective of Surgeons and Postgraduate Trainees

Cheng TC, Yahya MFN, Mohd Naffi AA, Othman O, Seng Fai T, Yong MH, et al.

J Craniofac Surg, 2021 Oct 01;32(7):2285-2291.
PMID: 33770023 DOI: 10.1097/SCS.0000000000007645

BACKGROUND: To evaluate the satisfaction of surgeons and trainees with three-dimensional (3D) ophthalmic surgery during a demonstration compared to traditional surgery.
METHODS: This validated questionnaire-based study was conducted over 1-month during which Ngenuity 3D surgery was demonstrated. All surgeons and trainees exposed were recruited to complete a questionnaire comprising visualization, physical, ease of use, teaching and learning, and overall satisfaction.
RESULTS: All 7 surgeons and 33 postgraduate students responded. Surgeons reported no significant difference except overall (P = 0.047, paired t-test). Postgraduate trainees reported significantly better experience with 3D for illumination (P = 0.008), manoeuvrability (P = 0.01), glare (P = 0.037), eye strain (P = 0.008), neck and upper back strain (P = 0.000), lower back pain (P = 0.019), communication (P = 0.002), comfortable environment (P = 0.001), sharing of knowledge (P = 0.000), and overall (P = 0.009).
CONCLUSIONS: During early experience, surgeons and trainees reported better satisfaction with 3D overall. Trainees had better satisfaction with 3D in various subcomponents of visualization, physical, ease of use, and education.

Matched MeSH terms: Learning
Fulltext Conditional Learning Deficits in Children with ADHD can be Reduced Through Reward Optimization and Response-Specific Reinforcement

De Meyer H, Tripp G, Beckers T, van der Oord S

Res Child Adolesc Psychopathol, 2021 09;49(9):1165-1178.
PMID: 33792820 DOI: 10.1007/s10802-021-00781-5

When children with ADHD are presented with behavioral choices, they struggle more than Typically Developing [TD] children to take into account contextual information necessary for making adaptive choices. The challenge presented by this type of behavioral decision making can be operationalized as a Conditional Discrimination Learning [CDL] task. We previously showed that CDL is impaired in children with ADHD. The present study explores whether this impairment can be remediated by increasing reward for correct responding or by reinforcing correct conditional choice behavior with situationally specific outcomes (Differential Outcomes). An arbitrary Delayed Matching-To-Sample [aDMTS] procedure was used, in which children had to learn to select the correct response given the sample stimulus presented (CDL). We compared children with ADHD (N = 45) and TD children (N = 49) on a baseline aDMTS task and sequentially adapted the aDMTS task so that correct choice behavior was rewarded with a more potent reinforcer (reward manipulation) or with sample-specific (and hence response-specific) reinforcers (Differential Outcomes manipulation). At baseline, children with ADHD performed significantly worse than TD children. Both manipulations (reward optimization and Differential Outcomes) improved performance in the ADHD group, resulting in a similar level of performance to the TD group. Increasing the reward value or the response-specificity of reinforcement enhances Conditional Discrimination Learning in children with ADHD. These behavioral techniques may be effective in promoting the learning of adaptive behavioral choices in children with ADHD.

Matched MeSH terms: Learning
The Effect of Teaching Formulaic Expressions Through Contrastive Lexical Approach on Iranian Pre-intermediate EFL Learners' Writing Skill

Ebrahimi F, Namaziandost E, Ziafar M, Ibna Seraj PM

J Psycholinguist Res, 2021 Oct;50(5):1087-1105.
PMID: 33830415 DOI: 10.1007/s10936-021-09778-z

This study aimed to investigate the effect of the contrastive lexical approach on Iranian EFL learners' writing skills. For this study, forty pre-intermediate students from a private English language institutes in Ahvaz, Iran were selected. Then, they were randomly divided into two equal groups of 20; one experimental and one control group. To have two groups of equal numbers, we used a block randomization sampling method. All of these students were female, ranging in age from 18 to 30. Their level of English language proficiency had already been determined by the Institute to be pre-intermediate. First, they were given a pre-test to determine their writing ability. Afterward, the experimental group received writing practices through the Contrastive Lexical Approach (CLA), during 14 sessions. Each session lasted for an hour and a half. The teacher sensitized learners in the experimental group towards the presence of L2 equivalents for L1 formulaic expressions, while the control group received an ordinary, traditional instruction, during which learners read texts containing the same formulaic expressions as for the experimental group without receiving any translation and were then asked to write about the same topics. At the end of the course, a post-test was administered to the two groups. Data were analyzed through independent and paired samples t tests after ensuring the normality of the data. Finally, to discover the power of the statistical tests, the effect size was also calculated. The study showed that using a contrastive lexical approach has a significant positive effect on Iranian EFL learners' writing skills. As the findings in this study propose, the writing skill can be improved through the use of a contrastive lexical approach. Teaching through a contrastive lexical approach, hopefully, gives the learners the chance to fathom their skillful writing competence, which requires the proper use of varied forms of structures and expressions and this, in turn, may sensitize them to know more about what language features to work on to increase their writing proficiency.

Matched MeSH terms: Learning
Computer Vision and Machine Learning Analysis of Commercial Rice Grains: A Potential Digital Approach for Consumer Perception Studies

Aznan A, Gonzalez Viejo C, Pang A, Fuentes S

Sensors (Basel), 2021 Sep 23;21(19).
PMID: 34640673 DOI: 10.3390/s21196354

Rice quality assessment is essential for meeting high-quality standards and consumer demands. However, challenges remain in developing cost-effective and rapid techniques to assess commercial rice grain quality traits. This paper presents the application of computer vision (CV) and machine learning (ML) to classify commercial rice samples based on dimensionless morphometric parameters and color parameters extracted using CV algorithms from digital images obtained from a smartphone camera. The artificial neural network (ANN) model was developed using nine morpho-colorimetric parameters to classify rice samples into 15 commercial rice types. Furthermore, the ANN models were deployed and evaluated on a different imaging system to simulate their practical applications under different conditions. Results showed that the best classification accuracy was obtained using the Bayesian Regularization (BR) algorithm of the ANN with ten hidden neurons at 91.6% (MSE = <0.01) and 88.5% (MSE = 0.01) for the training and testing stages, respectively, with an overall accuracy of 90.7% (Model 2). Deployment also showed high accuracy (93.9%) in the classification of the rice samples. The adoption by the industry of rapid, reliable, and accurate methods, such as those presented here, may allow the incorporation of different morpho-colorimetric traits in rice with consumer perception studies.

Matched MeSH terms: Machine Learning
Fulltext Rapid Detection of Fraudulent Rice Using Low-Cost Digital Sensing Devices and Machine Learning

Aznan A, Gonzalez Viejo C, Pang A, Fuentes S

Sensors (Basel), 2022 Nov 09;22(22).
PMID: 36433249 DOI: 10.3390/s22228655

Rice fraud is one of the common threats to the rice industry. Conventional methods to detect rice adulteration are costly, time-consuming, and tedious. This study proposes the quantitative prediction of rice adulteration levels measured through the packaging using a handheld near-infrared (NIR) spectrometer and electronic nose (e-nose) sensors measuring directly on samples and paired with machine learning (ML) algorithms. For these purposes, the samples were prepared by mixing rice at different ratios from 0% to 100% with a 10% increment based on the rice's weight, consisting of (i) rice from different origins, (ii) premium with regular rice, (iii) aromatic with non-aromatic, and (iv) organic with non-organic rice. Multivariate data analysis was used to explore the sample distribution and its relationship with the e-nose sensors for parameter engineering before ML modeling. Artificial neural network (ANN) algorithms were used to predict the adulteration levels of the rice samples using the e-nose sensors and NIR absorbances readings as inputs. Results showed that both sensing devices could detect rice adulteration at different mixing ratios with high correlation coefficients through direct (e-nose; R = 0.94-0.98) and non-invasive measurement through the packaging (NIR; R = 0.95-0.98). The proposed method uses low-cost, rapid, and portable sensing devices coupled with ML that have shown to be reliable and accurate to increase the efficiency of rice fraud detection through the rice production chain.

Matched MeSH terms: Machine Learning
Fulltext Machine Learning-Based Epileptic Seizure Detection Methods Using Wavelet and EMD-Based Decomposition Techniques: A Review

Thangarajoo RG, Reaz MBI, Srivastava G, Haque F, Ali SHM, Bakar AAA, et al.

Sensors (Basel), 2021 Dec 20;21(24).
PMID: 34960577 DOI: 10.3390/s21248485

Epileptic seizures are temporary episodes of convulsions, where approximately 70 percent of the diagnosed population can successfully manage their condition with proper medication and lead a normal life. Over 50 million people worldwide are affected by some form of epileptic seizures, and their accurate detection can help millions in the proper management of this condition. Increasing research in machine learning has made a great impact on biomedical signal processing and especially in electroencephalogram (EEG) data analysis. The availability of various feature extraction techniques and classification methods makes it difficult to choose the most suitable combination for resource-efficient and correct detection. This paper intends to review the relevant studies of wavelet and empirical mode decomposition-based feature extraction techniques used for seizure detection in epileptic EEG data. The articles were chosen for review based on their Journal Citation Report, feature selection methods, and classifiers used. The high-dimensional EEG data falls under the category of '3N' biosignals-nonstationary, nonlinear, and noisy; hence, two popular classifiers, namely random forest and support vector machine, were taken for review, as they are capable of handling high-dimensional data and have a low risk of over-fitting. The main metrics used are sensitivity, specificity, and accuracy; hence, some papers reviewed were excluded due to insufficient metrics. To evaluate the overall performances of the reviewed papers, a simple mean value of all metrics was used. This review indicates that the system that used a Stockwell transform wavelet variant as a feature extractor and SVM classifiers led to a potentially better result.

Matched MeSH terms: Machine Learning
Fulltext Weak and strong ties and its connection to experts' problem-solving styles in scaffolding students' PBL activities on social media

Abdullah AH, Neo TK, Low JH

F1000Res, 2021;10:1076.
PMID: 35035894 DOI: 10.12688/f1000research.73210.2

Background: Studies have acknowledged that social media enables students to connect with and learn from experts from different ties available in the students' personal learning environment (PLE). Incorporating experts into formal learning activities such as scaffolding problem-solving tasks through social media, allows students to understand how experts solve real-world problems. However, studies that evaluate experts' problem-solving styles on social media in relation to the tie strength of the experts with the students are scarce in the extant literature. This study aimed to explore the problem-solving styles that the experts portrayed based on their ties with the students in problem-based learning (PBL) on Facebook. Methods: This study employed a simultaneous within-subject experimental design which was conducted in three closed Facebook groups with 12 final year management students, six business experts, and one instructor as the participants. The experts were invited by the students from the weak and strong ties in their PLE. Hinging on the Strength of Weak Ties Theory (Granovetter, 1973) and problem-solving styles (Selby et al., 2004), this study employed thematic analysis using the ATLAS.ti qualitative data analysis software to map the experts' comments on Facebook. Results: The experts from strong and weak ties who had a prior relationship with the students showed people preference style by being more sensitive to the students' learning needs and demonstrating firmer scaffolding compared to the weak ties' experts who had no prior relationship with the students. Regardless of the types of ties, all experts applied all manner of processing information and orientation to change but the degree of its applications are correlated with the working experience of the experts. Conclusion: The use of weak or strong ties benefited the students as it expedited their problem-solving tasks since the experts have unique expertise to offer depending on the problem-solving styles that they exhibited.

Matched MeSH terms: Problem-Based Learning
A novel machine learning scheme for classification of medicinal herbs based on 2D-FTIR fingerprints

Yoon TL, Yeap ZQ, Tan CS, Chen Y, Chen J, Yam MF

Spectrochim Acta A Mol Biomol Spectrosc, 2022 Feb 05;266:120440.
PMID: 34627017 DOI: 10.1016/j.saa.2021.120440

A proof-of-concept medicinal herbs identification scheme using machine learning classifiers is proposed in the form of an automated computational package. The scheme makes use of two-dimensional correlation Fourier Transformed Infrared (FTIR) fingerprinting maps derived from the FTIR of raw herb spectra as digital input. The prototype package admits a collection of 11 machine learning classifiers to form a voting pool. A common set of oversampled dataset containing 5 different herbal classes is used to train the pool of classifiers on a one-verses-others manner. The collections of trained models, dubbed the voting classifiers, are deployed in a collective manner to cast their votes to support or against a given inference fingerprint whether it belongs to a particular class. By collecting the votes casted by all voting classifiers, a logically designed scoring system will select out the most probable guess of the identity of the inference fingerprint. The same scoring system is also capable of discriminating an inference fingerprint that does not belong to any of the classes the voting classifiers are trained for as the 'others' type. The proposed classification scheme is stress-tested to evaluate its performance and expected consistency. Our experimental runs show that, by and large, a satisfactory performance of the classification scheme of up to 90 % accuracy is achieved, providing a proof-of-concept viability that the proposed scheme is a feasible, practical, and convenient tool for herbal classification. The scheme is implemented in the form of a packaged Python code, dubbed the "Collective Voting" (CV) package, which is easily scalable, maintained and used in practice.

Matched MeSH terms: Machine Learning
Feature selection and risk prediction for patients with coronary artery disease using data mining

Md Idris N, Chiam YK, Varathan KD, Wan Ahmad WA, Chee KH, Liew YM

Med Biol Eng Comput, 2020 Dec;58(12):3123-3140.
PMID: 33155096 DOI: 10.1007/s11517-020-02268-9

Coronary artery disease (CAD) is an important cause of mortality across the globe. Early risk prediction of CAD would be able to reduce the death rate by allowing early and targeted treatments. In healthcare, some studies applied data mining techniques and machine learning algorithms on the risk prediction of CAD using patient data collected by hospitals and medical centers. However, most of these studies used all the attributes in the datasets which might reduce the performance of prediction models due to data redundancy. The objective of this research is to identify significant features to build models for predicting the risk level of patients with CAD. In this research, significant features were selected using three methods (i.e., Chi-squared test, recursive feature elimination, and Embedded Decision Tree). Synthetic Minority Over-sampling Technique (SMOTE) oversampling technique was implemented to address the imbalanced dataset issue. The prediction models were built based on the identified significant features and eight machine learning algorithms, utilizing Acute Coronary Syndrome (ACS) datasets provided by National Cardiovascular Disease Database (NCVD) Malaysia. The prediction models were evaluated and compared using six performance evaluation metrics, and the top-performing models have achieved AUC more than 90%. Graphical abstract.

Matched MeSH terms: Machine Learning

Filters

Please provide feedback to Administrator ([email protected])

External Links