FRAMEWORK
The increasing prevalence of chronic diseases, particularly cardiorenal and gastrointestinal conditions, represents one of the greatest challenges facing healthcare systems. This calls for approaches that enable more continuous and personalised patient monitoring, with a focus on the early detection of clinical changes.
Most assessment methods still rely on intermittent clinical examinations and in-person consultations with healthcare professionals, making it difficult to collect data on a regular basis and to identify early signs of clinical risk.
In this context, advances in sensing technologies, Artificial Intelligence and digital health are creating new opportunities to integrate clinical monitoring into everyday environments, making it more accessible, discreet and non-invasive.
PROPOSED SOLUTION
T-PRISM (Toilet Platform for Real-time Integrated Sensing and Monitoring for Analysis) It aims to develop and preclinically validate a smart multisensor toilet capable of providing continuous, non-invasive health monitoring.
The solution will integrate multiple data acquisition and analysis technologies to assess indicators associated with cardiovascular, renal and gastrointestinal health.
The platform will combine data from urinalysis covering 14 biomarkers with electrocardiography (ECG), photoplethysmography (PPG), bioimpedance, temperature, weight and stool analysis, providing a comprehensive and integrated view of the user's health status.
By leveraging Artificial Intelligence and multimodal data analysis, the system aims to support the early detection of physiological changes and provide clinically relevant information to assist healthcare professionals in decision-making.
CCG/ZGDV CONTRIBUTION
CCG/ZGDV is responsible for developing the Computer Vision and Artificial Intelligence components for stool analysis. Its work focuses on the development of advanced image analysis algorithms capable of automatically identifying and characterising clinically relevant indicators for gastrointestinal health assessment.
The planned functionalities include stool classification according to the Bristol Stool Scale, analysis of colour and fragmentation, detection of mucus, and identification of visual indicators potentially associated with the presence of occult blood.
The results generated will be integrated with the other data collected by the platform, contributing to a multimodal assessment of the user's health status.
Through this contribution, CCG/ZGDV further strengthens its expertise in Artificial Intelligence for Healthcare and Computer Vision by developing technologies capable of transforming complex visual information into clinically relevant digital biomarkers for continuous, non-invasive health monitoring.