Life Sciences and HealthCare Analytics Program

Get about, CRISP - ML(Q) the perfect Project Management Methodology used for handling Data Mining projects. Understand the entire process flow including Business Problem definition, Data Collection, Data Cleansing, Feature Engineering, Feature Selection, Model Building, Deployment and Maintenance. Get introduced to the principles of big data and learn about the opportunities being created. Understand about how Data is generation and explosion of data, Innovations in the space of analytics. Learn how to distinguish between data types, Exploratory data analysis, the Various moments of Business decisions and various Graphical techniques. Learn about probability and probability distribution namely Z distribution and Student's t-distribution.

Learn about Hypothesis testing, the many Hypothesis testing Statistics, work with the Null Hypothesis & Alternative hypothesis and Types of hypothesis testing. Interpret the results of Hypothesis test and probabilities of Alpha error, understand Type I and Type II errors. Get introduced to Linear regression, various components of Linear regression viz regression line, Linear regression equation, the concept of Ordinary Least Square. Get introduced to Linear regression analysis, and Linear regression examples

Understand the Linear regression in a multivariate scenario, understand collinearity and how to deal with it. Get introduced to the analysis of Attribute Data, understand the principles of Logistic regression, Binary Logistic regression analysis. Learn about the Multiple Logistic regression, Probability measures, and its interpretation. Get clarity on the confusion matrix and its elements. Get introduced to “Cut off value” estimation using AUC and ROC curve, understand False Positive Rate, False Negative Rate, Sensitivity, Specificity. Gain a birds-eye view to various advanced regression techniques and analysis of count data namely Poisson regression, Negative binomial regression. Learn when to use Poisson regression and negative binomial regression for predicting count data.

This module serves as an interesting primer for the uninitiated into the world of LSHC domain. Learn about the different sub-domains and how data science is empowering each of these domains. Get ready to deep dive into Life Science and Healthcare Industry, Sectors and Domains.

Welcome to the eclectic world of healthcare data. Understand the rich data structures contained in Electronic Health Records (EHR), work hands on MIMIC II data and other types of data like images from radiology, pathology, etc. Also Understand about healthcare IoT sensors data.

This module will give students the opportunity to understand the 4Ws of the healthcare data life cycle: What, Who, When, Why. Get a deeper understanding of the lifecycle to prepare for analyses that follow in subsequent modules.

As the name suggests, this module will help the learners to gather the requisite Python, R skills to analyze healthcare data such as EHR, MIMIC, etc.

This module will provide insights into various data models that are existing in Life Sciences and Healthcare Data. Learners can expect to gain knowledge on how to decipher and assess designs of data models using Entity-Relationship (ER) diagrams to contrast their individual merits and enunciate how to utilize them in providing Healthcare and Bio-statistics models, understanding the Genomics Exploratory Data Analysis, differential expression levels, normalization, and batch effects of the data.

• a. Genomics - DNA, RNA (Next Generation Sequencing Data Analysis), Gene MicroArray Gene Expression Data Analysis
• b. Proteomics - Deal with Proteomics Data Analysis, Drug Resistance and Drug Repurposing Data Analysis
• c. Metabolomics - Molecular-level (Abundance and Intensity values) Data Analysis, Identifying Biomarkers, etc.

This module gets into the heart and soul of the Data Science program. We start from basic statistics and progress to regression models, unsupervised and supervised learning models.

• a. Basic Statistics, Hypothesis Testing and Regression Models
• b. Unsupervised Machine Learning Techniques
• c. Supervised Machine Learning Algorithms

Utilize patient historical data, current diagnosis and treatment and other life style and behavioral indicators to predict whether a patient is likely to get re-admitted to hospital and take preventive actions.

This module gives a basic introduction to the Natural Language Processing (NLP) on clinical trials data. Learners can expect to gain pragmatic toolsets to extricate data stored in the EMRs (Electronic Medical Records).

Use MultiLayer Perceptron to build predictive analytic capabilities using the EHR and other types of healthcare data. Use CNN models to build a prediction model to predict cancer using DICOM images. Understand about limitations of Deep Learning models in Life Science & Healthcare.

This module describes how we can leverage Deep Learning networks to analyze panoramic dental x-rays and automatic segmentation of mandibles. This is a fast and accurate method and will help in diagnosis and registration of dental records.