The objective of this PhD project is, as its research core, the application of Machine Learning techniques and Big Data analytics to monitor, in a non-invasive way, vital parameters of individuals engaged in tasks that require a high psychophysical effort. The industrial partners of this project are Formula Medicine (as Italian industrial partner with advisor Dr. Riccardo Ceccarelli) and AOTech (foreign industrial partner with advisor mr. Sebastien Philippe). Formula Medicine is a sports medicine center able to offer medical assistance and training programs both physical and mental. Its strength is represented by the Mental Economy Gym, a gym dedicated to the optimization of mental resources. AOTech, on the other hand, is a partner company of Formula Medicine and a leader in the definition of high-tech products and services for the automotive industry and motorsport. AOTech has designed a sport driving simulator, able to reproduce all the main world circuits. The simulator, thanks to the equipment of a hydraulic system, allows to relive physical and mental sensations very similar to those perceived during real driving. The software system also allows vehicle’s data extraction. Within the present project, also taking into account the research domain of the industrial partners, the focus has been addressed to the monitoring of athletes belonging to motorsport with two linked but distinct objectives: the first, strictly related to the analysis of the drivers’ body performances and the second dedicated to the automatic identification of cardiac pathologies starting from electrocardiographic data. Finally, the know-how on the monitoring of biomedical parameters, acquired during the first years of this PhD project in the field of motorsport, was exported to the field of software engineering with the aim of verifying the possibility of predicting the correctness of a programming task that a software developer performs, based on the continuous monitoring of his body parameters. As a first result of the PhD, novel metrics have been defined to objectify effort, physical consumption, stress, and other factors. These metrics have been included in the software in use in Formula Medicine to have a measure of performance. In addition, part of them were correlated with the race performance of the drivers, through the integration of the body data with the data derived from the driving simulator used in AOTech. With regards to the second research focus, a decision support system was defined in the context of early diagnosis of cardiac diseases. The recommendation system consists of several algorithms that accept as input a digital electrocardiographic lead and identify the presence of a possible cardiac pathology. Finally, in the software engineering research field, the production of a developer was measured by evaluating the absence of defects in the source code. Preliminary results show that the proposed approach—that takes into account biomedical and code-based features—allows to discriminate with fair accuracy the outcome of a programming task, reaching an accuracy higher than 80%. This result was compared to state of the art metrics based on measures on the source code. It was higher than source code metrics, thus demonstrating the importance of biometric measurements in the identification of correctness of a coding task.

Innovative information systems to monitor biomedical parameters during high demanding tasks

LAUDATO, Gennaro
2021-04-14

Abstract

The objective of this PhD project is, as its research core, the application of Machine Learning techniques and Big Data analytics to monitor, in a non-invasive way, vital parameters of individuals engaged in tasks that require a high psychophysical effort. The industrial partners of this project are Formula Medicine (as Italian industrial partner with advisor Dr. Riccardo Ceccarelli) and AOTech (foreign industrial partner with advisor mr. Sebastien Philippe). Formula Medicine is a sports medicine center able to offer medical assistance and training programs both physical and mental. Its strength is represented by the Mental Economy Gym, a gym dedicated to the optimization of mental resources. AOTech, on the other hand, is a partner company of Formula Medicine and a leader in the definition of high-tech products and services for the automotive industry and motorsport. AOTech has designed a sport driving simulator, able to reproduce all the main world circuits. The simulator, thanks to the equipment of a hydraulic system, allows to relive physical and mental sensations very similar to those perceived during real driving. The software system also allows vehicle’s data extraction. Within the present project, also taking into account the research domain of the industrial partners, the focus has been addressed to the monitoring of athletes belonging to motorsport with two linked but distinct objectives: the first, strictly related to the analysis of the drivers’ body performances and the second dedicated to the automatic identification of cardiac pathologies starting from electrocardiographic data. Finally, the know-how on the monitoring of biomedical parameters, acquired during the first years of this PhD project in the field of motorsport, was exported to the field of software engineering with the aim of verifying the possibility of predicting the correctness of a programming task that a software developer performs, based on the continuous monitoring of his body parameters. As a first result of the PhD, novel metrics have been defined to objectify effort, physical consumption, stress, and other factors. These metrics have been included in the software in use in Formula Medicine to have a measure of performance. In addition, part of them were correlated with the race performance of the drivers, through the integration of the body data with the data derived from the driving simulator used in AOTech. With regards to the second research focus, a decision support system was defined in the context of early diagnosis of cardiac diseases. The recommendation system consists of several algorithms that accept as input a digital electrocardiographic lead and identify the presence of a possible cardiac pathology. Finally, in the software engineering research field, the production of a developer was measured by evaluating the absence of defects in the source code. Preliminary results show that the proposed approach—that takes into account biomedical and code-based features—allows to discriminate with fair accuracy the outcome of a programming task, reaching an accuracy higher than 80%. This result was compared to state of the art metrics based on measures on the source code. It was higher than source code metrics, thus demonstrating the importance of biometric measurements in the identification of correctness of a coding task.
14-apr-2021
Machine learning; Big data; Biometric; Decision Support System; ECG
File in questo prodotto:
File Dimensione Formato  
Tesi_G_Laudato.pdf

Open Access dal 15/10/2022

Descrizione: Tesi di Dottorato
Dimensione 14.47 MB
Formato Adobe PDF
14.47 MB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11695/100496
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact