Skip to Main Content (Press Enter)

Logo UNISR
  • ×
  • Home
  • People
  • Outputs
  • Organizations
  • Expertise & Skills
  • Equipment

UNIFIND
Logo UNISR

|

UNIFIND

unisr.it
  • ×
  • Home
  • People
  • Outputs
  • Organizations
  • Expertise & Skills
  • Equipment
  1. Outputs

A pilot study for investigating the feasibility of supervised machine learning approaches for the classification of pedestrians struck by vehicles

Academic Article
Publication Date:
2021
Short description:
A pilot study for investigating the feasibility of supervised machine learning approaches for the classification of pedestrians struck by vehicles / Casali, M., Malchiodi, D., Spada, C., Zanaboni, A.M., Cotroneo, R., Furci, D., Sommariva, A., Genovese, U., Blandino, A.. - In: JOURNAL OF FORENSIC AND LEGAL MEDICINE. - ISSN 1752-928X. - 84:(2021), pp. 1-7. [10.1016/j.jflm.2021.102256]
abstract:
This research focuses on the application of Artificial Intelligence (AI) methodologies to the problem ofclassifying vehicles involved in lethal pedestrian collisions. Specifically, the vehicle type is predicted on thebasis of traumatic injury suffered by casualties, exploiting machine learning algorithms. In the present study,AI-assisted diagnosis was shown to have correct prediction about 70% of the time. In pedestrians struck bytrucks, more severe injuries were appreciated in the facial skeleton, lungs, major airways, liver, and spleen aswell as in the sternum/clavicle/rib complex, whereas the lower extremities were more affected by fracturesin pedestrians struck by cars. Although the distinction of the striking vehicle should develop beyond autopsyevidence alone, the presented approach which is novel in the realm of forensic science, is shown to be effectivein building automated decision support systems. Outcomes from this system can provide valuable informationafter the execution of autoptic examinations supporting the forensic investigation. Preliminary results fromthe application of machine learning algorithms with real-world datasets seem to highlight the efficacy of theproposed approach, which could be used for further studies concerning this topic.
Iris type:
1.1 Articolo in rivista
List of contributors:
Casali, Michelangelo; Malchiodi, Dario; Spada, Claudio; Zanaboni, Anna Maria; Cotroneo, Rosy; Furci, Domenico; Sommariva, Andrea; Genovese, Umberto; Blandino, Alberto
Authors of the University:
BLANDINO ALBERTO
Handle:
https://iris.unisr.it/handle/20.500.11768/152014
Published in:
JOURNAL OF FORENSIC AND LEGAL MEDICINE
Journal
  • Overview

Overview

URL

https://www.sciencedirect.com/science/article/pii/S1752928X21001414
  • Use of cookies

Powered by VIVO | Designed by Cineca | 26.11.1.0