HEAD OF Biomedical Signal Processing
null

Francesca D. Faraci

PhD
RESEARCH AREA AND MAIN RESEARCH INTERESTS/TOPICS

The Biomedical Signal Processing Group is one of the research area of the newly formed MeDiTech Institute. BSP is led by Francesca D. Herrchen-Faraci and is formed by 2 PostDocs and 5 PhDs students.

Our mission is on extracting clinically relevant information from biological, subjective and contextual data. Through the adaptation and optimization of Artificial Intelligence, advanced statistical modeling, machine learning, and deep learning tools, we target the development of systems to support personalized diagnosis, therapy and bio-behavioral monitoring.

The group presently has three main research sub-area :

Sleep and EEG analysis

We focus on the development of algorithms that can evaluate different aspects of sleep based on EEG and other biosignals. We have extensive expertise in developing state-of-the-art deep learning models for automatic scoring of sleep stages. We also focus on the mining of sleep disease biomarkers and approaches for their automatic identification from individuals’ data.

Heart and ECG analysis

Our main focus in the field of cardiology concerns the application of AI techniques to assist specialists in the diagnosis of rare cardiological diseases. In particular, our research focuses on the recognition of hereditary arrhythmogenic diseases such as Long QT Syndrome, Brugada Syndrome, Early Repolarisation Syndrome. To this end, we have built a solid knowledge concerning pre-processing, rhythm analysis and ECG signal morphology.

Lifestyle management (Bio-Behaviour & Wearables)

Using advanced modelling approaches and fusion of data, monitoring both the physiological and psycho-behavioural state of individuals, we focus on research for lifestyle change. This area uses longitudinal biosignal data of individuals monitoring sleep, physical activity, and stress, collected primarily by wearables and supplemented by the context of the participants’ everyday lives. Our research seeks to develop algorithms that help to improve the quality of life, to increase sleep quality, boost stress-coping capability, or motivate for better post-operative rehabilitation.

PROJECTS

Current projects:

  • E!CMIPA – Development of a personalized digital cardiac monitoring and alert system.
    Our Input: Development of advanced techniques for Signal Processing, Machine Learning for early
    identification of adverse events, GANs for synthetic data generation.
  • E! CUOREMA Immersive Bio & Behavioural Feedback for a better heart health
    Development of a new mobile application to increase cardio rehabilitation adherence.
    Our Input: Bio-Behavioural Change Personalised Algorithms
  • AP-DD AutoPlay application for differential diagnosis
    Develop a monitoring sistem, with toys equipped with inertial sensors, for the study of ludic child
    manipulation.
    Our Input : Machine Learning for ludic behaviour classification
  • E! DESyMED Ophthalmologic medical system for Dry Eye Syndrome treatment.
    Our input : Reinforcement Learning for the Dry Eye Syndrome (DES) treatment personalization.
  • WRSD Risk Identification and Prevention of Work-Related Stress Disorders
    Our input : Bio-Behavioural Change Personalised Algorithms
  • BCPerTT Breast Cancer Personalised Treatment Telemonitoring
    Develop a personalised support for telemonitoring Breast Cancer Treatment, ( thermal cameras + a
    mobile application that integrates advanced AI algorithms and educational and social support).
    Our Input: Optimise AI algorithms for thermal Images features
  • GESSE – Guardian Earbuds Sleep Scorer Expert
    Test the feasibility of applying automated sleep scoring in in-ear technologies
    Our Input: Deep Learning algorithms for Sleep Scoring
  • RENEWAL
    Tackling the eneRgy and wEllbeiNg impact of tEleWorking prActices through muLtisource data
    Rigorously evaluation of teleworking impacts on employees’ and companies’ energy consumption, carbon
    emissions, and subjective well-being.
    Our Input: analyse teleworkers’ biometric parameters
  • BE4AIBest4EthicalAI Strategies and best practices for the development and adoption of ethical and
    reliable AI in medicine
    Our purpose is of getting a better understanding of XAI and the adoption of an AI tool in the
    medical routine.

Projects Just Finished :

  • SPAS: Sleep Physician Assistant System
    Development of a Platform to assist Sleep Physician in all his routine.
    Our Input : Deep Learning Automated Sleep Scoring
  • MyDoctor LifeStyle
    To test the feasibility of providing a personalised support, accordingly to Lifestyle Medicine
    Pillars, to Family Doctors.
    Our Input: Bio-Behavioural Feedback for Lifestyle Change
PUBLICATIONS
  • Luigi Fiorillo, Giuliana Monachino, Julia van der Meer, Marco Pesce, Jan Warncke, Markus Schmidt, Claudio Bassetti, Athina Tzovara, Paolo Favaro,
    Francesca Faraci, U-Sleep: resilient to AASM guidelines, arXiv preprint arXiv:2209.11173, 2022. Accepted by NPJ Ditigal Medicine
  • Luigi Fiorillo, Davide Pedroncelli, Valentina Agostini, Paolo Favaro, Francesca Dalia Faraci, Multi-Scored Sleep Databases: How to Exploit the Multiple-Labels in Automated Sleep Scoring, Sleep, 2023;, zsad028, https://doi.org/10.1093/sleep/zsad028
  • Stefania Ancona, Francesca D. Faraci, Elina Khatab, Luigi Fiorillo, Oriella Gnarra, Tobias Nef, Claudio L. A. Bassetti & Panagiotis Bargiotas (2022). Wearables in the home-based assessment of abnormal movements in Parkinson’s disease: a systematic review of the literature. Journal of neurology, 1-11
  • Natalia Norori, Qiyang Hu, Florence Marcelle Aellen, Francesca Dalia Faraci, Athina Tzovara (2021). Addressing bias in big data and AI for health care: A call for open science. Patterns, 2(10), 100347.
  • Alessandro Mascheroni, Eun Kyoung Choe , Yuhan Luo , Michele Marazza , Clara Ferlito , Serena Caverzasio, Francesco Mezzanotte, Alain Kaelin-Lang, Francesca Faraci, Alessandro Puiatti , Pietro Luca Ratti The SleepFit Tablet Application for Home-Based Clinical Data Collection in Parkinson Disease: User-Centric Development and Usability Study JMIR mHealth and uHealth, 9(6), e16304
  • Ratti, Pietro-Luca; Faraci, Francesca; Hackethal, Sandraa ; Mascheroni, Alessandro ; Ferlito, Clara ; Caverzasio, Serena ; Amato, Ninfa ; Choe, Eun Kyoung ; Luo, Yuhand ; Nunes-Ferreira, Paulo-Edsona ; Galati, Salvatore ; Puiatti, Alessandro ; Kaelin-Lang, Alain A New Prospective, Home-Based Monitoring of Motor Symptoms in Parkinson’s Disease Journal of Parkinson’s disease, 9(4), 803-809.
  • Towards personalized burnout prevention system: a probabilistic approach for analysis of data from wearable devices with subjective feedback. A preliminary study M Bechny, R Svihrova, LG Arango, A Baldassari, M Grossenbacher, Y Ilchenko, FD Faraci JOURNAL OF SLEEP RESEARCH 31
  • Simone Sguazza, Alessandro Puiatti, Sandra Bernaschina, Francesca Faraci, Gianpaolo Ramelli, Vincenzo D’Apuzzo, Emmanuelle Rossini, Michela Papandrea Sensor data synchronization in a IoT environment for infants motricity measurement IoT Technologies for HealthCare: 6th EAI International Conference, HealthyIoT 2019, Braga, Portugal, December 4–6, 2019, Proceedings 6
  • Francesca D Faraci, Michela Papandrea, Alessandro Puiatti, Stefania Agustoni, Sara Giulivi, Vincenzo D’Apuzzo, Silvia Giordano, Flavio Righi, Olmo Barberis, Evelyne Thommen, Emmanuelle Rossini Autoplay: a smart toys-kit for an objective analysis of children ludic behavior and development
    2018 IEEE International Symposium on Medical Measurements and Applications (MeMeA)
FINANCIAL SUPPORT
  • Eurostars
  • Innosuisse
  • ABREOC
  • SNF