Emanuele Matassa
Audio-Based Infant Pain Detection and Pattern Recognition Using Explainable Deep Learning.
Rel. Marco Agostino Deriu. Politecnico di Torino, Corso di laurea magistrale in Ingegneria Biomedica, 2026
Abstract
Infant cry analysis represents a common technique used in efforts to decode the require-ments of babies. Babies do not have the ability to express their needs; therefore, it is essen-tial to examine the behavior represented by the cry as the only means of communication about their condition. However, without using any methods for analyzing the baby's cry, it becomes difficult to understand the needs. Thus, using artificial intelligence makes it possi-ble to measure some acoustic features of crying sound, including fundamental frequency and amplitude. Some researchers have recently suggested applying machine learning or deep learning to the problem of assessing infant crying.
Nevertheless, it should be noted that building an AI algorithm might be a challenging task considering the diversity of pain levels in a newborn baby
Relatori
Anno Accademico
Tipo di pubblicazione
Numero di pagine
Informazioni aggiuntive
Corso di laurea
Classe di laurea
Aziende collaboratrici
URI
![]() |
Modifica (riservato agli operatori) |
