RESUMEN
BACKGROUND AND OBJECTIVES: The survival rate of laryngeal cancer has improved in recent day. But in case of advanced cancer, radical destructive surgery is required for survival. So early detection of laryngeal cancer prevent from this disastrous condition. In this point of view, mass screening test for early detection of laryngeal cancer is necessary. Screening tests using voice has many advantages of being simple, non invasive, and requiring less space. In this system, the most important factor is the selection of acoustic parameters to be used in voice analysis. Thus authors studied the acoustic parameters that can differentiate normal, benign, malignant laryngeal diseases by an acoustic analyzing system and we also checked the availability of parameters. MATERIALS AND METHODS: We evaluated the voice analyzed data from 25 laryngeal malignancy patients, 33 benign laryngeal disease patients, 35 normal control who visited PNUH otolaryngology department from October 1996 to May 1998. A computerized speech lab. 4300B (CSL) was used to carry out the analysis of each voice sample and statistical analysis, ANOVA. Canonical analysis and cumulative frequency curve were used. RESULTS: The statistically significant parameters that can differentiate normal and malignant laryngeal disease groups were 15 parameters and can differentiate normal and benign laryngeal disease group were 9 parameters and that can differentiate benign and malignant laryngeal disease group were 7 parameters. CONCLUSION: We consider that these parameters and detection programs may be effective in development of a screening system using voice only. Developing diagnostic tools and programs would need further study of phonetics and voice engineering.