Objectives

The Course is divided into   two parts: the first one [1] is dedicated to Biomedical Signal Analysis. The objective is to introduce the   most diffused methods of information processing from biomedical signals and   to describe the most significant applications in Medicine and Biology, in   both physiological studies and in clinical applications for diagnosis,   therapy and rehabilitation.

The second part of the   Course [2] deals with Medical Images.   The objective is to provide basic concepts for their characterization, to   illustrate clinical and diagnostic problems and to introduce the principles   on which some techniques for their processing and reconstruction are based.

The Course is constituted by   frontal lessons, exercise sessions, seminars and laboratory exercise aiming   at more deeply analyzing the topics of signal and image processing making use   of MatLab. The   attendance of the Course is strongly adviced.

Course   Content

Part [1]   Biomedical Signal Processing. Prof. Sergio Cerutti

Programme of lessons and exercises

Introduction   to biomedical signal processing. General block diagram of   biomedical signal processing operations. Recalls from analogue and digital filterings.

Cardiovascular   System. ECG signal: main superimposed noises and most significant   configurations from clinical standpoint. High resolution ECG. Ventricular   late potentials: pathophysiological aspects and processing methods: Simson   parameters. Study of Autonomic Nervous System (ANS) by means of processing of   heart rate variability signals on short- and long-term bases.   Pathophysiological aspects: studying of the neural control mechanisms (in   particular of heart rate and arterial blood pressure). Fetal ECG: processing   methods and enhancement of parameters useful from clinical standpoint. Fetal   states monitoring. Arterial blood pressure signal: detection systems and main   clinical parameters. Interaction models among signals related to autonomic   nervous system: ECG, arterial blood pressure and respiration; open- and   closed-loop models. Pathophysiological interpretations.

Diagnostic Classification. Classification methods. Principal Component Analysis (PCA): various applications.

Neurosensorial   System. EEG (electroencephalographic) signal analysis, evoked potentials (EPs) and event-related potentials (ERPs). Review   on traditional processing methods with main applications in clinical and in   research environments: cerebral activity detected on the scalp, at cortical   level and in deep brain stimulation (DBS).

Signal   processing and parametric identification. Time   series analysis detected from biological signals. Examples relative to ECG,   EEG, EP in the various phases of pre-processing, filtering, prediction,   extraction and estimation of parameters, diagnostic classification. Deterministic and stochastic identification.   Review on stochastic identification approaches. Model   families, in particular AR/MA/ARMA (autoregressive and moving average) and   models with exogenous input X. Parametric spectral analysis, including   backward and forward methods, maximum entropy methods, Pisarenko, Prony, etc.   Comparison with the traditional non-parametric techniques: various examples   on heart rate variability signals, EEGs and EPs.

Prerequisites: Mandatory   prerequisites are not requested; it is advisable to be aware of the fundamentals   of digital signal processing.

Part [1] Biomedical Signal   processing (short summary)

 
  The course is intended to recall general signal processing aspects, to   introduce new methods beyond those considered in the bachelor degree and   examine various clinical and research application fields. Methods.   From deterministic filtering to stochastic parametric analysis: mono and   multi-variate AR/MA/ARMA models and non parametric as well as parametric   spectral analysis. Principal component analysis. Entropy in signal   processing. Applications. Automatic analysis and classification of the   electrocardiographic signal (ECG). The autonomic nervous system,   cardiovascular variability signals, and cardiorespiratory interactions.   Foetal ECG signal. High resolution ECG and late ventricular potentials. The   central nervous system: processing of the electroencephalographic signal   (EEG) and of evoked (EP) and event-related potentials (ERP).

Bibliography

                                                                      Course Notes (on Beep)

     Fisiopatologia e Bioingegneria   Cardiovascolare (on   Beep)

     J.G. Webster, Medical Instrumentation   : Application and Design, Houghton Mifflin Co, 2010

     A.V. Oppenheim, R.W. Schafer, Digital Signal Processing, Prentice Hall, 1975

     G. De Nicolao, R. Scattolini, Identificazione parametrica,   Editore: CUSL ed, Anno edizione: 1997

Hours

lessons

30.0

Exercise

20.0

informatic   lab

0.0

experimental   lab

0.0

Project

0.0

project   lab

0.0

Notes on Evaluation Modalities

The evaluation will be given on the basis of a written exam, dealing with   the topics dealt with in the Lessons and Exercise hours of the two Parts   [1][2] (Biomedical Signals and Images) which constitute integrant parts of   the Course. The Student may choose to take the exam relative to only one Part   of the Integrated Course, provide that he/she obligates to make the other   Part of the Course within the Sessions of the Academic Year of attendance.   The exam is passed if the Student gets a positive evaluation in both the   Parts. The final grade is the average of the single grades obtained in the   two parts.