Biostatistics
[FARM2177]
[22.5h] 3 credits
This course is taught in the 2d semester.
Teacher(s): Jean Cumps
Language: French
Level: Second
Aims
Main themes
Content and teaching methods
Summarizing and presenting data in tables and graphs
- Extract and organize electronically stored data
- Produce useful graphical and numerical summaries
Univariate statistics
- Descriptive aspect (median, standard deviation, variance, interval of confidence)
- Validation aspect (test on normality of distribution, discordance tests on outliers, precision, accuracy)
- Significance tests: type 1 and type 2 errors
- Capability analysis
Bivariate analysis: one-way and two-way ANOVA
- Descriptive aspect: multiple box-plot, means or medians
- Validation aspects: normal distribution of residuals, detection of outliers.
- Significance tests: type 1 (t test, Tukey test or Dunnett test) and type 2 (power test).
Linear regression model
- Parameter determination.
- Validation aspect: limit of detection and quantification.
- Inverse prediction
Non-linear regression
- Kinetic models
- Michaelis-Menten and Hill models
- Pharmacokinetic models
- Dissolution models
MANOVA and repeated measures analysis of variance
Multivariate statistical methods:
Logistic regression and ROC curves
Survival analysis
Exercises with statistical software (JMP)
- Use of an intranet site to illustrate the course (slides, JavaScript illustrations, summary) and the exercises (exercises, solutions to exercises, tables of statistics).
- Connections with clinical and biomedical applications.
Other information (prerequisite, evaluation (assessment methods), course materials recommended readings, ...)
Evaluation based on the treatment or the discussion of examples issued from the scientific literature in the medical or pharmaceutical field.
Staff: 1 professor /20 students for the practical exercises.
Teaching aided with computer, practical exercises with statistical software JMP