Advanced search
Start date

Deep learning for pattern recognition on multi-sensor and multidimensional data

Grant number: 23/10442-2
Support Opportunities:Scholarships in Brazil - Post-Doctoral
Effective date (Start): September 01, 2023
Effective date (End): February 28, 2025
Field of knowledge:Physical Sciences and Mathematics - Computer Science - Computer Systems
Principal Investigator:Odemir Martinez Bruno
Grantee:Leonardo Felipe dos Santos Scabini
Host Institution: Instituto de Física de São Carlos (IFSC). Universidade de São Paulo (USP). São Carlos , 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
Associated scholarship(s):24/00530-4 - Texture Feature Aggregation and Learning with Vision Transformers and its Applications on Biological and Medical Images, BE.EP.PD


Sensors and biosensors play a crucial role in various fields, such as early cancer diagnosis, detection of viruses, contamination in food, water, etc. Intending to develop cost-effective and accurate detection and diagnosis strategies, researchers have explored the use of machine learning techniques and computer vision. While these approaches have shown promising results in biological images, sensor, and biosensor images have been less explored in the literature. In the FAPESP thematic project (process 2018/22214-6) linked to this proposal, a significant amount of image data is generated from sensors and biosensors using techniques such as scanning electron microscopy, optical microscopy, and confocal laser scanning microscopy (CLSM). These images possess unique textural properties, and previous research indicates that texture analysis methods hold great promise for characterizing them. Meanwhile, deep learning and computer vision have made remarkable strides recently, with techniques such as Vision Transformers (ViTs) quickly emerging and delivering impressive results. However, ViT models have not yet been thoroughly analyzed for texture analysis. Therefore, this postdoctoral project seeks to develop cutting-edge computational techniques to analyze sensor and biosensor images, focusing on deep learning and the latest models, such as ViTs. These models will be adapted to specific applications through transfer learning, and novel methods for feature aggregation and extraction will be developed. These proposals embody our overarching goal of handling sensor and biosensor images from multiple sources with appropriate modeling techniques. The developed models will be employed in several tasks according to the demands of the research groups in the thematic project. In this sense, in addition to the contributions of novel computer vision and deep learning techniques, we believe that the results will also aid in achieving one of the thematic project's objectives: computer-assisted diagnosis utilizing multiple data sources. (AU)

News published in Agência FAPESP Newsletter about the scholarship:
Articles published in other media outlets (0 total):
More itemsLess items

Please report errors in scientific publications list using this form.