
Speaker: Teka Kimbi Ntimanputu
20 Marzo 2026 | 11:30
DEIB - NECSTLab Meeting Room (Ed. 20)
Online by Zoom
Contatti: Prof. Marco Santambrogio
Sommario
On Friday, March 20th, 2026, we will have a new talk for the series #NECSTFridayTalk.During this talk, we will have, as speaker, Teka Kimbi Ntimanputu, MSc student at Dipartimento di Elettronica, Informazione e Bioingegneria.
Cybersickness remains a significant issue for users of immersive Virtual Reality (VR), resulting from the interaction between visual stimuli, individual susceptibility, and system design. While physiological monitoring is often used to assess this condition objectively, many existing models treat cybersickness as a single task, merging the onset of symptoms with their intensity. This simplification makes it difficult to interpret the model’s decisions and limits its use in systems that need to adapt to the user in real-time.
This work introduces CASSANDRA, a framework designed to separate these two dimensions. Instead of a single prediction, the system models cybersickness through two related tasks: the binary detection of symptom onset and the estimation of symptom severity. The approach uses data from consumer-grade electrocardiography (ECG) combined with environmental descriptors that track visual motion and scene dynamics. This represents a step forward in enabling the system to associate physiological responses directly with the technical triggers occurring within the virtual environment.
By using a shared preprocessing and evaluation pipeline, this research compares how different factors influence the start of discomfort versus its progression. The results show that this two-layer approach can achieve accuracy comparable to state-of-the-art methods and outperform direct regression models that attempt to estimate all variables simultaneously, even when using consumer-grade sensors. Furthermore, explainability analyses suggest that the features driving the initial detection of symptoms are distinct from those that track their worsening. Ultimately, CASSANDRA provides a practical technical basis for monitoring users in real-time and designing VR environments capable of responding to user distress.
The NECSTLab is a DEIB laboratory, with different research lines on advanced topics in computing systems: from architectural characteristics, to hardware-software codesign methodologies, to security and dependability issues of complex system architectures.
#NECSTLab #Computerscience
Every week, the “NECSTFridayTalk” invites researchers, professionals or entrepreneurs to share their work experiences and projects they are implementing in the “Computing Systems”.
