Learning multiview representations for image analysis with applications in biosensors

• Grant number: 24/01744-8
• Support Opportunities: Scholarships in Brazil - Master
• Start date: October 01, 2024
• End date: February 28, 2026
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques
• Principal Investigator: Lucas Correia Ribas
• Grantee: Ricardo Trivizan Fares
• Host Institution: Instituto de Biociências, Letras e Ciências Exatas (IBILCE). Universidade Estadual Paulista (UNESP). Campus de São José do Rio Preto. São José do Rio Preto , SP, Brazil
• Associated research grant: 18/22214-6 - Towards a convergence of technologies: from sensing and biosensing to information visualization and machine learning for data analysis in clinical diagnosis, AP.TEM

Abstract

With the increasing complexity and diversity of images captured by different devices, there is a need for more sophisticated image analysis methods. The images captured by biosensors are examples of contemporary data that are still little studied. They are composed of complex patterns that require highly discriminative methods, capable of describing their complexity to perform recognition tasks. Currently, many image analysis techniques focus on a single perspective of the image, limiting the ability to extract complex and robust information. In this sense, the main objective of this project is to develop methods based on learning multiview representations for image analysis. This approach allows the integration of multiple perspectives of the same image, thus increasing the complementarity of information and the robustness of the representation. Thus, this project focuses on three key research points for method development: (i) study, acquisition and generation of image views. (ii) learning multiview representations; (iii) multi-view feature aggregation. In addition to the theoretical front, this project aims to apply the methods developed in the characterization and classification of biosensor images aiming at new strategies for applications such as early diagnosis of cancer and detection of viruses or contaminations. These images are provided by collaborators of the thematic project (process, 2018/22214-6) to which this master's project is linked. In this way, it is expected that the methods developed will contribute to advances in the area of image analysis with more robust methods, in addition to new detection and diagnosis strategies in the areas of physical chemistry and medicine.

Scientific publications

(References retrieved automatically from Web of Science and SciELO through information on FAPESP grants and their corresponding numbers as mentioned in the publications by the authors)

RIBAS, LUCAS C.; CASACA, WALLACE; FARES, RICARDO T.. Conditional Generative Adversarial Networks and Deep Learning Data Augmentation: A Multi-Perspective Data-Driven Survey Across Multiple Application Fields and Classification Architectures. AI, v. 6, n. 2, p. 24-pg., 2025-02-01. (24/01744-8, 13/07375-0, 23/04583-2, 23/14427-8)