MLGroup @ AIC These are some of the research lines the group is working in:
Artificial Intelligence in Medicine
This research line arises from the limitations medical professionals face in clinical practice. It focuses on the development and application of Computer Vision and Deep Learning techniques in the medical field. Primarily, it involves the analysis of various imaging modalities (e.g., MRI, ultrasound, mammography) to support the diagnosis and treatment of different diseases (e.g., breast cancer, cardiovascular diseases, dry eye). Other types of medical data (e.g., medical history) can be integrated, combining various sources of information to improve clinical decision-making and personalize patient care. The ultimate goal is to drive technological advances that contribute to more efficient and effective healthcare systems while improving patients’ quality of life.
Links to some representative publications:
Contact person: Beatriz Remeseiro
Machine Learning in Marine Ecology
Marine ecosystems hold unparalleled significance in sustaining life on our planet. The utilization of machine learning techniques in this domain aims to automate the monitoring of these complex ecosystems, facilitating the implementation of informed conservation strategies. The use of technologies such as deep learning, computer vision, and quantification, alongside the utilization of diverse data sources such as satellite imagery, drone imagery, or in-situ automated water sampling, machine learning models aid in critical tasks such as species identification, habitat mapping, and population monitoring.
Links to some representative publications:
- Automatic plankton quantification using deep features
- Validation methods for plankton image classification systems
Contact person: Pablo González
Image by LuqueStock on Freepik
Recommender Systems
Recommender systems research focuses on developing algorithms that provide personalized suggestions to users. These systems analyze vast amounts of data to identify patterns and preferences, delivering tailored recommendations in areas like e-commerce, streaming services, social media, and online advertising. Key techniques include collaborative filtering, which uses the preferences of similar users, and content-based filtering, which recommends items with similar attributes. Hybrid approaches combine these methods for improved results. The goal is to improve user experience by delivering relevant and engaging content, increasing user satisfaction and loyalty. This field continually evolves with the growing complexity of data and the rising demand for personalization.
Links to some representative publications:
Contact person: Jorge Díez Peláez
In progress
…and more to be added soon!