Signal Laboratory
- Presentation
- Services
- Resources
- Outstanding projects
El Biomedical Signals Laboratory promotes biomedical research and innovation through the development of advanced engineering solutions in biomedical signal processing and artificial intelligence applied to the analysis of physiological data. It operates as a transversal scientific and technological platform at the Parc Taulí Research and Innovation Institute (I3PT-CERCA), offering specialized support to I3PT research groups and national and international institutions in the fields of health and biomedical engineering.
The Laboratory's mission is to generate clinically relevant engineering solutions through advanced signal processing, data modeling and artificial intelligence, with the aim of improving patient monitoring, supporting clinical decision-making and advancing precision medicine in critical care. Its activity covers the entire translational cycle: from the acquisition of physiological signals and the development of algorithms to clinical validation and real-time implementation in hospital environments.
The laboratory maintains a strong commitment to open science and FAIR principles, managing anonymized physiological databases designed to support the development of robust, reproducible and clinically transferable algorithms. It also actively contributes to the training of the next generation of biomedical engineers and data scientists by supervising undergraduate, master's and doctoral research.
- Acquisition, integration and processing of physiological signals from monitored patients.
- Creation and management of anonymized physiological databases and generation of synthetic data.
- Advanced processing and analysis of continuous biomedical signals.
- Extraction of physiological characteristics, biomarkers and derived variables.
- Exploratory analysis, visualization and interpretation of physiological data.
- Development, validation and evaluation of biomedical signal processing algorithms and artificial intelligence.
- Adaptation and real-time implementation of physiological processing algorithms previously developed in offline environments.
- Integration of physiological monitoring and analysis algorithms in streaming architectures and real-time clinical systems.
- Methodological support and scientific and technical advice in research projects related to biomedical signals, clinical monitoring and artificial intelligence.
Infrastructure and equipment
- Specialized training in patient monitoring and connectivity systems (BC-Link).
- Physiological database RAMIC-I — Real-time Agnostic Monitoring for Intensive Care (DOI: https://doi.org/10.34810/DATA2509): high-resolution, FAIR-compliant dataset of mechanically ventilated patients in the ICU, supporting research on patient-ventilator interactions, predictive modeling, and optimization of invasive mechanical ventilation.
Computational infrastructure
- Windows Server 2022 Datacenter; 2 × Intel® Xeon® Gold 6348 @ 2,60 GHz; 32 GB of RAM; 500 GB of local storage; 5 TB NAS.
Development and analysis environments
- MATLAB, Python, PostgreSQL, pgAdmin, BetterCare, Spyder, Visual Studio Code, Jupyter Notebook.
IntelliLung
Multicenter prospective observational study with centers in Germany, Spain and Poland, which develops an AI-based digital tool for the automated generation of personalized information for patients and families in the ICU, and validates IntelliLung, an AI-based clinical decision support system for the optimization of mechanical ventilation and other life support treatments in critically ill patients.
Funding: Horizon Europe (101057434)
Entropy Care
Development of an offline clinical workstation for the optimization of the mechanical ventilation withdrawal process and the prevention of self-extubations in critical patients, using advanced nonlinear analysis techniques and artificial intelligence to detect complex patient-ventilator interactions, in collaboration with Althaia Hospital (Manresa).
Funding: Department of Health (SLT017/20/000153)
Characterization of Flow Starvation
Multicenter study that evaluates, using artificial intelligence, the incidence of high inspiratory effort in critically ill patients. The project combines retrospective data from the ICU of the Parc Taulí University Hospital with prospective data from the ICU of the Hospital Clínic de Barcelona, and includes the real-time implementation of algorithms for automatic detection of high inspiratory effort during mechanical ventilation.
Funding: AES ISCIII (PI24/01268)