Dottoranda in Bioingegneria E Scienze Medico-chirurgiche , 36o ciclo (2020-2023)
Dipartimento di Ingegneria Meccanica e Aerospaziale (DIMEAS)
Assegnista di Ricerca
Dipartimento di Ingegneria Meccanica e Aerospaziale (DIMEAS)
Docente esterno e/o collaboratore didattico
Dipartimento di Ingegneria Meccanica e Aerospaziale (DIMEAS)
Profilo
Dottorato di ricerca
Titolo della tesi
Predicting Personalized Pathological Risks and Dynamics in Cardiology - How to Exploit Data-Driven Approaches in Clinical Decision Support Systems
Argomento di ricerca
Predicting pathological risks and dynamics through explainable artificial intelligence techniques.
Tutori
- Umberto Morbiducci
- Marco Agostino Deriu
- Umberto Michelucci
Presentazione della ricerca
Interessi di ricerca
Biografia
Michela Sperti (SCOPUS ID: 57217331713, h-index: 5) is a third-year Ph.D. student at Politecnico di Torino, Bioengineering Department, with Prof. M. A. Deriu. She graduated in Biomedical Engineering at Politecnico di Torino in 2019 with a thesis on machine learning techniques for cardiovascular risk prediction in rheumatic patients. She worked for one year as a Research Assistant under the European-funded MSCA VIRTUOUS project (which aims to apply machine learning techniques to investigate taste and food properties). Currently, she is studying explainability techniques for machine learning and deep learning models applied in clinical decision support systems with a special focus on cardiovascular risk prediction. The final aim of her research is the understanding of complex mechanisms that underlie physiological processes. She is very passionate about teaching and is committed to communicating her results. She is the author of nine articles published in peer-review journals, three conference papers and she took part in three international workshops as both a teaching assistant and a speaker. She is Contributing Author of the book "Applied Deep Learning with TensorFlow 2" by U. Michelucci.
Settore scientifico discliplinare
(Area 0009 - Ingegneria industriale e dell'informazione)
Competenze
Settori ERC
SDG
Premi e riconoscimenti
- Best Review Article of the Year Award, Presented to the authors of the highest impact review article published in 2022 in Minerva Cardiology Angiology. (2023)
- Description of role of Michela Sperti in the writing and preparation of “Applied Deep Learning with TensorFlow 2 – Second Edition” published by APRESS/Springer in 2022. (2022)
Didattica
Insegnamenti
Corso di laurea magistrale
- Biomechanical design. A.A. 2023/24, INGEGNERIA BIOMEDICA. Collaboratore del corso
- Biomeccanica multiscala. A.A. 2023/24, INGEGNERIA BIOMEDICA. Collaboratore del corso
- Biomechanical design. A.A. 2022/23, INGEGNERIA BIOMEDICA. Collaboratore del corso
- Biomeccanica multiscala. A.A. 2022/23, INGEGNERIA BIOMEDICA. Collaboratore del corso
- Biomechanical design. A.A. 2021/22, INGEGNERIA BIOMEDICA. Collaboratore del corso
- Rational Drug Design: Principles and Applications. A.A. 2021/22, INGEGNERIA BIOMEDICA. Collaboratore del corso
- Biomeccanica multiscala. A.A. 2021/22, INGEGNERIA BIOMEDICA. Collaboratore del corso
Ricerca
Gruppi di ricerca
Pubblicazioni
Pubblicazioni degli ultimi anni
Coautori PoliTO
Pubblicazioni per tipo
Pubblicazioni più recenti Vedi tutte le pubblicazioni su Porto@Iris
- Kassem, Karim; Sperti, Michela; Cavallo, Andrea; Vergani, Andrea Mario; Fassino, Davide; ... (2024)
An innovative artificial intelligence-based method to compress complex models into explainable, model-agnostic and reduced decision support systems with application to healthcare (NEAR). In: ARTIFICIAL INTELLIGENCE IN MEDICINE, vol. 151. ISSN 0933-3657
Contributo su Rivista - Venturini, F.; Michelucci, U.; Sperti, M.; Gucciardi, A.; Deriu, M. A. (2023)
Understanding the learning mechanism of convolutional neural network applied to fluorescence spectra. In: SPIE OPTO, 2023. ISBN: 9781510659810
Contributo in Atti di Convegno (Proceeding) - Venturini, F; Michelucci, U; Sperti, M; Gucciardi, A; Deriu, Ma (2022)
One-dimensional convolutional neural networks design for fluorescence spectroscopy with prior knowledge: explainability techniques applied to olive oil fluorescence spectra. In: SPIE Photonics Europe. ISSN 0277-786X. ISBN: 9781510651548
Contributo in Atti di Convegno (Proceeding) - Sperti, M; Michelucci, U; Venturini, F; Gucciardi, A; Deriu, Ma (2022)
Chemical analysis of olive oils from fluorescence spectra thanks to one-dimensional convolutional neural networks. In: SPIE Photonics Europe. ISSN 0277-786X. ISBN: 9781510651548
Contributo in Atti di Convegno (Proceeding) - Sperti, Michela; Malavolta, Marta; Ciniero, Gloria; Borrelli, Simone; Cavaglià, Marco; ... (2020)
JAK Inhibitors in Immune-Mediated Rheumatic Diseases: from a Molecular Perspective to Clinical Studies. In: JOURNAL OF MOLECULAR GRAPHICS & MODELLING, vol. 104. ISSN 1093-3263
Contributo su Rivista