Keynote Speakers - ColCACI 2025

Talk: Diffusion MRI tractography: principles and applications

Abstract: Diffusion Magnetic Resonance Imaging (dMRI) is a non-invasive neuroimaging technique that maps the structure of white matter (WM) by characterizing the diffusion of water molecules in tissue. Tractography, derived from dMRI data, enables the reconstruction of 3D representations of WM fiber pathways, allowing the study of both well-known bundles and previously unidentified fascicles across individuals. This lecture will provide an overview of dMRI acquisition and preprocessing pipelines, followed by a review of key methods for analyzing structural brain connectivity using tractography. We will also discuss some example applications in neuroimaging research.

Talk: Accelerating Secure Computing for IoT Sensors

Abstract:

Security in IoT sensors has focused on cryptography-based but has failed to consider System-on-Chip (SoC) interoperability during side-channel, and physical attacks, as they concentrate solely on standalone behavior. This talk will discuss SoC security solutions that shield a SoC against supply voltage and physical attacks while apprising SoC interoperation. Along with the supply monitoring shields, this talk will present an efficient AES-256b encryption module accelerator, a physical unclonable function, and a true random number generator as SoC peripheral instances.

Talk: Small Steps Toward a Human-Centered, Trustworthy, Transparent, and Sustainable AI in Latin America.

Abstract: This talk discusses foundational concepts and desirable properties of Artificial Intelligence, presenting ongoing research at the IoT & Data Atelier Research Group in Recife, Brazil, that represents small but meaningful steps toward explaining complex computational intelligence systems using network science, understanding the impact of data regularization on out-of-distribution generalization, advancing anomaly detection in industrial machines, promoting algorithmic fairness in missing data imputation for multivariate clinical time series, and detecting money laundering in bank transactions. The keynote advocates for the development of AI in Latin America that is trustworthy, transparent, sustainable, and human-centered.

Talk: From Explainability to Trustworthiness in the Application of AI to Biomedical Contexts.

Abstract:

Despite a decade of advances in AI for biomedical applications, clinical adoption remains limited. This talk examines the limitations of widely used explainability tools like saliency maps, questioning their real value in medical decision-making. We will explore persistent barriers to adoption — including data bias, lack of generalization, and shortcut learning — and highlight how these challenges are tied to overlooked domain and protected variables such as dataset provenance, device technology, and patient demographics. The keynote advocates for trustworthy AI through techniques like domain-adversarial training, Bayesian modeling, and diffusion models in latent spaces, paving the way for more robust, fair, and clinically reliable solutions.

Talk: Research into Artificial Intelligence: Real-World Applications from Agriculture to Smart Sensors

Abstract: Artificial intelligence research drives the development of real-world solutions by integrating AI, intelligent sensors, and optical technologies to tackle key challenges in diverse areas such as precision agriculture, medical imaging, environmental monitoring, seismic imaging, and information security. This presentation highlights how interdisciplinary collaboration enables the creation of sensors capable of perceiving beyond human vision, systems that learn and adapt, and platforms that transform complex data into actionable insights. The showcased applications include smart sensing technologies for agriculture, privacy-preserving computer vision, medical diagnostics, and geophysical exploration.

TUTORIALS