Quantitative Techniques in Medicine - MM61311
Biostatistics - MM61511
Pattern Recognition& Machine Intelligence in Medicine - MM61504
Epidemiological Analysis - MM72332

SEMESTER (Autumn)

MM61311     Quantitative Techniques in Medicine   (3-1-0)      4 credit

Elementary matrix theory and inverse of a matrix, solutions of system of linear equations its application in bioscience. Bayes’ Theorem, Binomial, Poisson and Normal distributions and its applications in medicine. Retrospective (case control) and prospective (cohort) studies; randomized controlled trials; Different risk measures. Simple random and stratified sampling, standard error of a sample mean and of a proportion and their differences. Categorical and numerical data, Box-Whisker’s plot, Stem-and-leaf plot and histogram, central tendency and dispersion. Diagnostic and prognostic studies, scatter plots, correlation analysis, linear and logistic regression techniques with ANOVA and model goodness of fit. Principles of multivariate methods and their application to clinical data, Principal component analysis, Discrimination and classification, Clustering methods and its applications for clinical data. Hypothesis testing – parametric tests (Z-test, one sample t-test, paired t-test, F-test) and non-parametric tests (Chi-square test, Wilcoxon test, Kolmogorov Smirnov test) and Type I and II errors, p-value and confidence intervals with its statistical and clinical significance. SPSS tutorial for some diagnostic studies.

MM61511  Biostatistics   (3-1-0)      4 credit

Measurements and descriptive statistics in medical research and practice: Data types and scales of measurement: continuous vs. enumeration data, Sampling distributions - normal distribution (continuous data), binomial distribution (proportions, based on enumeration data), Measures of central tendency-mean, median, mode Measures of variability-standard deviation and standard error, Probability and odds Confidence limits on the mean Disease incidence and prevalence, Sensitivity, specificity, positive and negative predictive values Risk measures-relative risk, attributable risk, odds ratios, risk factors, Survival curves. Sampling : Concept of a source population, Random sampling, Estimation of population statistics, Standard error of a sample mean and of a proportion, and their differences, Confidence intervals
Regression and Correlation: Simple, Partial and Multiple Correlation, Simple Linear / Nonlinear Regression, Logistic Regression for dichotomous variable.
Statistical Inference and Hypothesis Testing: Hypothesis generation, Null hypothesis, Type I and II errors, Statistical Power, Interpretation of P-values and confidence intervals, Statistical and clinical significance. Comparing 2 or more groups: Comparing means of two populations with the t-test (continuous data), Comparing proportions of responders in two populations (enumeration data), Chi square with corrections (goodness of fit, test of independence). One - Way ANOVA: F-test, Treatments & Factors.

SEMESTER (Spring)

MM61504  Pattern Recognition and Machine Intelligence in Medicine   4 credit

Feature Extraction and classification stages. Different approaches to pattern recognition. Statistical Pattern Recognition: Hypothesis testing, Linear classifiers, Bayes decision theory, Parametric and nonparametric classification techniques, Unsupervised learning and clustering, Syntactic pattern recognition, Fuzzy set theoretic approach to PR. Introduction to AI, Problem Space Representation, Heuristic Search Techniques, Knowledge Representation, Predicate Logic, Reasoning under uncertainty, Statistical Reasoning, Planning, Learning, Expert System Design, Expert System Shell, Case Studies of Typical Expert Systems in Medicine.
Neural networks: Multi-layer perceptrons, radial basis functions, self organizing feature map, counter propagation and recurrent networks.

MM72332    Epidemiological Analysis      (3-1-0)           4 credit

Probability (conditional probability, Bayes’ theorem), discrete and continuous distributions (binomial, poisson, normal, chi-square), statistical estimation (point estimates, confidence intervals) and hypothesis testing (significance level, type I and II errors, power, p-value), and multiple linear regression (model and assumptions, categories and dummy variables, inference about parameters, interactions, polynomial, transformations, and model diagnostics). Measures of disease frequency and exposure effects, Systematic and random error and selection bias, Confounding and interaction, Classical methods of analysis for cohort studies, Classical methods of analysis for case-control studies, Unconditional logistic regression, Poisson regression for cohort studies, Survival analysis and proportional hazards regression, Practical issues in study design and analysis, Conditional logistic regression for case-control studies, Meta-analysis, Clustering of data in epidemiological studies. SPSS tutorial for epidemiological studies.