Published Research Papers

With the increase in volume of data and presence of structured and unstructured data in the biomedical field, there is a need for building models which can handle complex & non-linear relations in the data and also predict and classify outcomes with higher accuracy. Deep learning models are one of such models which can handle complex and nonlinear data and are being increasingly used in the biomedical field in recent years. Deep learning methodology evolved from artificial neural networks which process the input data through multiple hidden layers with higher level of abstraction. Deep Learning networks are used in various fields such as image processing, speech recognition, fraud deduction, classification and prediction. Objectives of this paper is to provide an overview of Deep Learning Models and its application in the biomedical domain using R Statistical software. Deep Learning concepts are illustrated by using the R statistical software package. X-ray Images from NIH datasets used to explain the prediction accuracy of the deep learning models. Deep Learning models helped to classify the outcomes under study with 91% accuracy. The paper provided an overview of Deep Learning Models, its types, and its application in the biomedical domain. This paper has shown the effect of deep learning networks in classifying images into normal and disease with 91% accuracy with the help of the R statistical package.

AIM: The aim of this paper is to provide an overview of artificial neural network (ANN) in biomedical domain and compare it with the logistic regression model. METHODS: Artificial neural network models and logistic regression models were created and compared using a sample of a modified dataset adapted to the dataset from Framingham Heart Study. R statistical software package is used to create and compare the models. RESULTS: The results indicated that the ANN model is more accurate in classifying the dependent variable than the logistic regression model (84.4 % vs 82.9 %). CONCLUSION: This paper has shown the effect of artificial neural network models in classifying the survival status (event or non-event).

Survival analysis is one of the main areas of focus in medical research in recent years. Survival analysis involves the concept of 'Time to event'. The event may be mortality, onset of disease, response to treatment etc. Purpose of this paper is to provide an overview of frequentist and Bayesian Approaches to Survival Analysis. The paper starts with the overview of the basic concepts of survival analysis and then discusses the frequentist and Bayesian approaches to survival analysis in the biomedical domain with help of a hypothetical survival dataset. The survival analysis of the hypothetical data sets showed that for the specific dataset and specific hypothesis, the Bayesian approach provided direct probability that the null hypothesis is true or not and the probability that the unknown parameter (mean survival time) lies in a given credible interval wherein the frequentist approach provided p-values and confidence intervals for interpreting whether the null hypothesis is true or not and the percentage of intervals which will contain the parameter when the experiment is repeated under the same conditions. The use of Bayesian survival analysis in the biomedical domain has increased due to the availability of advanced commercial and free software, its ability to handle design and analysis issues in survival models and the ease of interpretation of the research findings.

Objectives: With the exponential increase in the number of articles published every year in the biomedical domain, there is a need to build automated systems to extract unknown information from the articles published. Text mining techniques enable the extraction of unknown knowledge from unstructured documents.
Methods: This paper reviews text mining processes in detail and the software tools available to carry out text mining. It also reviews the roles and applications of text mining in the biomedical domain.
Results: Text mining processes, such as search and retrieval of documents, pre-processing of documents, natural language processing, methods for text clustering, and methods for text classification are described in detail.
Conclusions: Text mining techniques can facilitate the mining of vast amounts of knowledge on a given topic from published biomedical research articles and draw meaningful conclusions that are not possible otherwise.

This paper provides a step by step overview of the processes involved in mining of biomedical literature using the R-Statistical Package. Abstracts from the PubMed database on a given topic are retrieved, stored, and pre-processed using R programming codes. The resultant term document matrix is used to find associations between terms and the frequency of the terms in each document. Finally, clouds of words and clustering of documents are created using the R software to discover the associations between the documents. The results from the process provided a step by step understanding of the retrieval of abstracts, pre-processing of abstracts and clustering of abstracts using a user-based query term.

Maximum likelihood Estimation is an important aspect of the frequentist approach which was introduced by RA Fisher [1]. The Maximum Likelihood estimation method helps us to find the estimator for an unknown population parameter. There are other methods of estimation also available such as Least Square Estimation and Bayesian Estimation methods but Maximum Likelihood Estimation is the widely used method to estimate parameters. This paper provides an overview of the Maximum Likelihood Method with an example to calculate a Maximum Likelihood Estimator from a sample data set.