作者: Mathieu Balaguer , Timothy Pommée , Jérôme Farinas , Julien Pinquier , Virginie Woisard
DOI: 10.1002/HED.25949
关键词:
摘要: Background: The development of automatic tools based on acoustic analysis allows to overcome the limitations perceptual assessment for patients with head and neck cancer. aim this study is provide a systematic review literature describing effects oral oropharyngeal cancer speech intelligibility using analysis. Methods: Two databases (PubMed Embase) were surveyed. selection process, according preferred reporting items reviews meta-analyses (PRISMA) statement, led final set 22 articles. Results: Nasalance studied mainly in patients. vowels are mostly formant vowel space area, consonants by means spectral moments specific parameters their phonetic characteristic. Machine learning methods allow classifying “intelligible” or “unintelligible” T3 T4 tumors. Conclusions: comprehensive models combining different measures would better consideration functional impact disorder.