Universitat Internacional de Catalunya

Artificial Intelligence II

Artificial Intelligence II
4
14874
4
Second semester
op
Main language of instruction: English

Other languages of instruction: Catalan, Spanish

Teaching staff


Dra. ÁLVAREZ DE LA CAMPA CRESPO, Elena - ealvarezdelacampa@uic.es

Introduction

This second part of the course offers a deep dive into the application of Artificial Intelligence (AI) techniques in the field of biomedicine. 

Throughout the programme, students will learn to implement and evaluate AI models in a real biomedical context, addressing everything from predicting drug sensitivities to interpreting genetic variants and recognising biomedical images using convolutional neural networks.

Pre-course requirements

It is recommended to have completed and passed:

  • Introduction to Bioinformatics
  • Biological Databases

It is recommended to take in parallel:

  • Artificial Intelligence I
  • Programming Knowledge

Objectives

  • Understand and apply AI techniques in biomedicine.
  • Develop practical skills in cleaning and preparing biomedical data.
  • Learn to build and evaluate classification and regression models to solve real problems in biomedicine.
  • Explore the application of advanced algorithms, such as convolutional neural networks, in the analysis of biomedical images.

Competences/Learning outcomes of the degree programme

BASIC:

  • CB1: Students have demonstrated that they possess and understand knowledge in an area of study that builds on general secondary education, and is typically at a level that, while based on advanced textbooks, also includes some aspects that involve knowledge from the forefront of their field of study.
  • CB3: Students have the ability to gather and interpret relevant data (usually within their area of study) to make judgments that include reflection on relevant social, scientific, or ethical issues.
  • CB4: Students can convey information, ideas, problems, and solutions to both specialized and non-specialized audiences.
  • CB5: Students have developed those learning skills necessary to undertake further studies with a high degree of autonomy.

GENERAL:

  • CG7: Integrate basic concepts related to the field of biomedicine both theoretically and experimentally.
  • CG10: Design, draft, and execute projects related to the area of Biomedical Sciences.
  • CG11: Recognize basic concepts from different areas linked to biomedical sciences.

SPECIFIC:

  • CE7: Apply statistical tools to studies in Health Sciences.
  • CE19: Recognize the principles of biomedical sciences related to health and learn to work in any area of Biomedical Sciences (biomedical company, bioinformatics laboratories, research laboratories, clinical analysis company, etc.).

CROSS-CURRICULAR:

  • CT1: Develop the capacity for organization and planning appropriate to the moment.
  • CT2: Develop the capacity for problem-solving.
  • CT3: Develop the capacity for analysis and synthesis.
  • CT4: Interpret experimental results and identify consistent and inconsistent elements.
  • CT5: Use the internet as a means of communication and as a source of information.
  • CT6: Know how to communicate, make presentations, and write scientific papers.
  • CT7: Be able to work in a team.
  • CT8: Reason and evaluate situations and results from a critical and constructive point of view.
  • CT9: Have the ability to develop skills in interpersonal relationships.
  • CT10: Be capable of carrying out autonomous learning.
  • CT11: Apply theoretical knowledge to practice.
  • CT12: Apply the scientific method.

Learning outcomes of the subject

  • Theoretical Understanding and Practical Application of AI in Biomedicine:
    • Acquire a deep understanding of the role of AI in biomedicine.
    • Apply AI in biomedical contexts, such as in drug discovery and genetic analysis.
  • Skills in Handling and Analyzing Biomedical Data:
    • Develop practical skills in cleaning, preparing, and analyzing biomedical data.
    • Interpret genetic and molecular data in clinical and research contexts.
  • Development and Evaluation of Classification and Regression Models:
    • Build and evaluate classification and regression models in biomedicine.
    • Understand and apply advanced techniques like convolutional neural networks in the analysis of biomedical images.
  • Critical Evaluation and Research Strategies:
    • Analyse and critically evaluate the results of AI models.
    • Propose research strategies and clinical applications based on AI.
  • Professional Development and Career Guidance:
    • Gain insight into the opportunities and challenges in careers related to bioinformatics and biomedicine.
    • Develop communication and presentation skills for AI topics in biomedicine.

Syllabus

  1. Introduction (2h): The Growing Role of Artificial Intelligence in Healthcare. An overview of how AI is transforming the field of health and biomedicine.
  2. Classification Models in Biomedicine (20h):
    • AI in Drug Discovery.
    • Implementation of a classification pipeline from scratch: data cleaning, preprocessing, and result evaluation. Focus on the application of AI in drug discovery from a biomedical perspective.
      • Theory and practice on data preparation and data splitting, using gene expression data and drug sensitivity.
      • Training and evaluation of a Random Forest classifier, addressing issues such as class imbalance and feature selection.
      • Focus on predicting the sensitivity of chemical compounds to specific cell lines, including representation of chemical fingerprints and adapting the pipeline to predict off-target toxicity.
      • Overview of Deep Learning in drug discovery and discussion about applications and career paths in bioinformatics and biomedicine.
  3. Regression Models in Biomedicine (8h):
    • Focus on predicting the molecular impact of genetic variants using regression models.
      • Theory and exercises on molecular impact of variants and building regression models.
      • Binarization of regression models, comparison with classifiers, and analysis of in silico tools.
  4. Convolutional Neural Networks in Biomedicine (10h):
    • Application of CNNs in the analysis of biomedical images.
      • Introduction to CNNs and preparation of biomedical image data.
      • CNN architectures and model optimization in biomedicine.
      • Model evaluation and analysis of results in biomedical AI projects.

Teaching and learning activities

In person



  • Lectures in blocks of between 15 and 50 minutes on a theoretical topic to be developed by the professor.
  • Clinical cases or case methods (CM): Presentation of a real or imaginary situation. Students work on the questions formulated in small groups or in active interaction with the teacher and the answers are discussed. The teacher intervenes actively and, if necessary, contributes new knowledge.
  • Virtual education (VE): Online material that students can consult from any computer at any time and that will contribute to self-learning of concepts related to the course.

Evaluation systems and criteria

In person



  • Students in the first call:
    • Continuous Assessment (35%): Includes practical exercises and short tests. 
    • Final Theoretical Exam (65%): Assessment of theoretical knowledge and understanding of the AI II concepts applied to biomedicine covered during the course. 
    • Subjective Component (up to 10%): Up to 10% of the final grade may be allocated based on subjective criteria such as engagement, participation, and adherence to rules, to encourage an active and committed attitude in the classroom. 
  • Students in the second or subsequent call: The Continuous Assessment grade is retained, and the final exam will account for 75% of the final grade. 

General points to consider about the evaluation system:

  • To calculate an average grade, a minimum score of 5 is required in the final exam. 
  • In addition to the above, to pass the course, the average of all grades must be 5 or higher. 
  • The ongoing nature of this assessment means that it is not possible to evaluate the subject if participation in 75% of the hours has not been achieved. 
  • Misuse of electronic devices (such as recording and distributing content of students or teachers during sessions, as well as using these devices for non-educational purposes) can lead to expulsion from the class.
Print version