NEUROIMAGING

The course aims to provide knowledge and skills in big data analysis, both from a theoretical/methodological perspective and a practical/applicative one, using the open-source software R. The data analyzed will be from mass spectrometry imaging (MSI). Modern mass spectrometry (MS) techniques analyze the abundance of a wide range of molecules in specific biological samples, while MSI extends this by generating spatial imaging data, where each pixel contains a mass spectrum.
The continuous innovation in measurement technologies has made available vast amounts of multi-omics information, offering a comprehensive view of biological systems. Therefore, it is crucial to develop scalable algorithms and statistical models that can provide accurate biological insights. Furthermore, proper data cleaning is essential for high-quality scientific research.
These methodologies will be discussed using real data on brain tissue samples and/or other types of tissue biopsies, with particular emphasis on choosing the correct methodology for the analysis goal and interpreting the results. The lessons will be divided into theoretical and practical sessions.
Program
November 14-15: Introduction to MSI data types, explanation of MALDI-MSI, and data processing. From raw instrument data to usable data matrices. Review of dimensionality reduction methods such as Principal Component Analysis (PCA), t-distributed stochastic neighbor embedding (t-SNE), and Uniform Manifold Approximation and Projection (UMAP), as well as unsupervised clustering methods like k-means, hierarchical clustering, and community detection. Initial data visualization for data cleaning and pre-processing. Practical exercises in R using sample datasets.
November 21-22: Practical lab sessions to understand the functioning of imaging instruments, including MALDI-MSI, LC-MS, and the preparation and scanning of slides. Meetings with lab technicians, biologists, and doctors for detailed explanations.
December 5-6: Statistical methods for big data analysis, from penalized regression to advanced models that incorporate spatial information. Practical exercises in R using MSI data, along with the integration of multiple datasets. Alignment of matrices and data analysis, with the development of useful tools such as Shiny apps in R to make data accessible to doctors.
December 12-13: Overview of other imaging technologies, such as magnetic resonance and automatic cell recognition. Students will present and discuss their projects based on an imaging dataset provided at the beginning of the course, addressing real-world issues and discussing the results.

Knowledge and Understanding
• Understanding the main methods for acquiring and analyzing mass spectrometry imaging data
• Recognizing the importance of implementing robust and reproducible methodologies
Judgment Autonomy
• Evaluating the validity of results obtained using standard statistical methods or new methods developed to address specific scientific questions
• Recognizing potential errors through critical analysis of developed/applied methods

Communicative Skills
• Communicating learned knowledge and the results of their application using appropriate terminology, both orally and in writing
• Interacting with the instructor and course colleagues in a respectful and constructive manner, particularly during group work

Learning Skills
• Taking notes, selecting, and gathering information according to their importance and priority
• Being sufficiently autonomous in collecting relevant data and information related to the investigated problem

The course is open to students with basic knowledge of programming by in general and by the use of software R.

- Description of the physical principles of magnetic resonance imaging, upon which modern neuroimaging techniques are based.
- Definition of methods for image registration and segmentation, depending on the technique used and the analysis objective.
- Presentation of the main analysis techniques for studying brain activity and connectivity using functional magnetic resonance imaging.
- Evaluation of the obtained results, with particular emphasis on their interpretation and presentation methods in a scientific work.

Faulkner W.H., Rad Tech's Guide to MRI. Wiley Blackwell (2020).
Poldrack R.A., Handbook of Functional MRI Data Analysis. Cambridge University Press (2011).

The achievement of the teaching objectives is evaluated through participation in activities and exercises assigned during the course, as well as the completion of an individual project related to the acquisition and/or analysis of magnetic resonance imaging data.

At the end of the course, the student must produce a report on the individual project, in the form of a scientific article.

Students attending classes can earn additional points by participating in exercises proposed in class. The bonus will be added to the grade of the individual project.

Seminars: training in blended learning, combining activities in both synchronous and asynchronous modes.
Exercises: integrated tutorials with group work (peer-teaching, problem solving).

English

IMPORTANT: lectures will begin during October; further information will be provided by the Scientific Campus

Language of the course: English

Examination method: Development of an individual project during the course, and writing of a final report on the activities carried out

written