In Sweden over 10.000 patients annually, mostly children, undergo tonsillar surgery. The most common surgical method is total removal of the tonsil, tonsillectomy (TE). During the last decade partial tonsil resection, tonsillotomy (TT), has been reintroduced and is associated with lower primary morbidity and less postoperative pain. The aim of the present study was to evaluate if and to what extent 4-5 year old children’s voice function are affected by tonsillar hypertrophy compared to controls and to study the effects of the two surgical techniques on vocal function. Sixty-seven children with tonsillar hypertrophy, where randomized either to TE or TT. A voice material consisting of words with only sonorants and three sustained vowels was recorded preoperatively and six months postoperatively. Eighty-two preschool children formed two control groups. The material was evaluated perceptually by three voice specialized speech and language pathologist and analyzed acoustically. Before surgery the children with tonsillar hypertrophy are rated to have higher prevalence of hoarseness, hyponasality and ‘compressed/throaty voice’ compared with the controls. The acoustic analysis show that the children with tonsillar hypertrophy have higher prevalence of perturbation (jitter (%), shimmer (%) and NHR) compared to the controls. Postoperatively there are no significant difference between the children with tonsillar hypertrophy and the control group perceptually. The acoustic analysis of the children with tonsillar hypertrophy still show more perturbation compared to the control group. The children operated with TE are rated to have less prevalence of roughness and hyponasality after surgery. The children in the TT-group have less prevalence of roughness, hyponasality and ‘compressed/throaty voice’. The acoustic analysis show no significant difference between the two postoperative groups. / Över 10 000 patienter tonsillopereras årligen i Sverige, de flesta av dem barn. Den vanligaste kirurgiska metoden har sedan länge varit totalt avlägsnande av tonsillerna, tonsillektomi (TE). Under det senaste decenniet har partiellt borttagande av tonsillvävnad, tonsillotomi (TT) erövrat terräng då operationen är förknippad med lägre postoperativ morbiditet och lägre smärta. Syftet med föreliggande studie var att undersöka om och hur röstfunktionen är påverkad av tonsillhypertrofi hos barn i 4-5 års ålder i jämförelse med köns- och åldersmatchade kontroller samt att studera effekterna på röstfunktionen efter två olika typer av kirurgisk behandling, TE respektive TT. I föreliggande studie deltog 67 barn med tonsillhypertrofi som randomiserades till antingen TE eller TT. Ett röstmaterial bestående av ord som till största del bestod av sonoranter samt tre uthållna vokaler spelades in omedelbart före operation samt 6 månader postoperativt. Åttiotvå köns- och åldersmatchade förskolebarn utgjorde två kontrollgrupper. Materialet bedömdes perceptuellt av tre röstlogopeder samt analyserades akustiskt. Barnen med tonsillhypertrofi uppfattas före operation ha högre förekomst av de perceptuella parametrarna heshet, hyponasalitet, och klämd/halsig i jämförelse med köns- och åldersmatchade kontroller. Den akustiska analysen i sin tur visar på mer perturbation (jitter (%), shimmer (%) och NHR) hos barnen med tonsillhypertrofi. Efter kirurgisk behandling visar den perceptuella röstbedömningen att det inte föreligger någon signifikant skillnad mellan de barn som genomgått kirurgisk behandling och kontrollerna. Den akustiska analysen visar på att både barnen som genomgått TE och barnen som genomgått TT har mer perturbation i jämförelse med köns- och åldersmatchade kontroller. Vid jämförelse pre- och postoperativt visar röstbedömningen på en minskning av skrovlighet och hyponasalitet för barnen som genomgått TE samt skrovlighet, hyponasalitet och klämd/halsig för barnen som genomgått TT. Den akustiska analysen visar på att det inte föreligger någon större skillnad mellan barnen som genomgått TE och barnen som genomgått TT.

Rörfonation och traditionell svensk röstbehandling är två av de vanligaste behandlingsalternativen inom svensk röstlogopedi idag. Som forskare är det viktigt att kontinuerligt utvärdera de behandlingar som erbjuds för att göra vården mer kvalitativ och kostnadseffektiv. Finns det skillnader i behandlingseffekt mellan dessa två röstbehandlingar? Är skillnaderna så tydliga att det går att fastslå att den ena behandlingen är att föredra framför den andra? Med analyser av ett patientmaterial, före och efter behandling, i form av akustisk analys, perceptuella bedömningar av en grupp erfarna logopeder samt patienternas självskattningar syftar föreliggande uppsats till att besvara vad behandlingsformerna har för effekt inom dessa områden. Patientmaterialet bestod av totalt 28 patienter varav 23 fullföljde behandling. Tolv av dessa erhöll behandling med rörfonation medan resterande elva patienter erhöll traditionell svensk röstbehandling.. Utöver röstinspelningar bestod materialet av patienternas självskattningar, i form av RHI- och Swe-VAPP-formulär. Gällande behandlingseffekt på akustiska parametrar kunde inga slutsatser dras. De perceptuella parametrarna afoni, hyperfunktion och knarr förbättrades sett till hela urvalet, oavsett typ av röstbehandling. Tydliga förbättringar sågs i patienternas självskattningar efter båda behandlingsmetoderna. Röstbehandling ger således positiva resultat, men några skillnader i behandlingseffekt mellan de två olika metoderna för röstbehandling kunde inte påvisas. / Tube phonation and traditional Swedish voice therapy are two of the most common voice therapy options in Sweden today. As a researcher, it is important to continually evaluate available therapy options to make health care more qualitative and cost effective. Are there differences in treatment effect between these two voice treatments? If so, are the differences so obvious that it is possible to conclude that one treatment should be preferred over the other? Analyses of a patient material, before and after treatment, which included acoustic analysis, perceptual analysis by a group of experienced speech-language therapists and patients' selfassessments were made aiming to answer what effect the two treatment options have in these areas. The patient material consisted of 28 patients, of whom 23 completed the therapy. Twelve of them received treatment with tube phonation while the remaining eleven patients received traditional Swedish voice therapy. Besides audio recordings, the material consisted of patients' self-assessments pre- and post-therapy, including VHI and Swe-VAPP forms. The treatment effect on acoustic parameters was inconclusive. The perceptual parameters aphonia, hyper function, and creaking improved over the entire sample, regardless of the type of voice therapy. Clear improvements were seen in the patients' self-assessments after both therapy methods. Voice therapy thus gives positive results, but no differences in treatment effect between the two different methods of voice therapy were seen.

O objetivo é caracterizar a voz e vibração glótica de pregas vocais com paralisia unilateral. Método: trata-se de um estudo retrospectivo que utilizou o banco de dados do grupo de pesquisa em engenharia médica da Universidade Federal de São Carlos. Foram utilizados dados de 35 pacientes adultos (11 homens e 24 mulheres) e cinco instrumentos de análise multidimensional. O primeiro deles foi o protocolo de auto-avaliação (Índice de Desvantagem Vocal - IDV). Foram extraídos os escores de domínio: físico, orgânico, emocional e total. Foram utilizados sinais de voz para realizar as análises: perceptivo-auditiva (escala analógica visual - EVA e parâmetros: grau global - G, rugosidade - R, soprosidade - S, tensão - T e Instabilidade na emissão vocal - I); acústica tradicional (parâmetros de perturbação - Jitter e Shimmer), experimental (parâmetro Coaptação Glótica - CG) e a análise não linear (NL) da dinâmica vocal (parâmetros qualitativos: número de loops, graus de convergência e regularidade; e quantitativos: desvio padrão do eixo e da dispersão - DPE, DPD). A quinta análise utilizou quimografias geradas de videolaringoscopias em alta velocidade para extração de parâmetros quantitativos (Razão de Abertura - RA, de Fechamento - RF e Assimetria). Análise estatística: descritiva e correlação de Spearman. Resultados: A análise perceptiva caracterizou intensidade moderada do desvio vocal. A rugosidade foi o tipo de voz predominante em amostras masculinas (55,84%) e a soprosidade em amostras femininas (63,59%). Os domínios de IDV apresentaram baixo impacto da disfonia na qualidade de vida dos pacientes. Houve correlação do domínio físico com S em amostras masculinas e com todos os parâmetros perceptivos em amostras femininas. Ambas as amostras apresentaram parâmetros de perturbação aumentados (Jitter: 0,90% homens e 1,20% mulheres; Shimmer: 3,30% homens e 2,76% mulheres), e de CG diminuídos (0,28 homens e 0,25 mulheres). Amostras femininas apresentaram correlação de Jitter e Shimmer com os parâmetros perceptivos. Amostras masculinas apresentaram correlação de Jitter com R e de Shimmer com R e T. A CG correlacionou com a S e l em amostras masculinas. Na análise NL da dinâmica vocal de ambas as amostras apresentaram reduzidos: número de loops, grau de convergência e de regularidade, enquanto que os valores de DPE e DPD apresentaram-se aumentados. Houve correlação de parâmetros perceptuais com os da análise NL em ambas as amostras. Os resultados de parâmetros quimográficos são: RA 1,55 para homens e 1,50 para mulheres; RF 0,37 para homens e 0,18 para mulheres; e Assimetria: 23,68º para homens e 9, 85º para mulheres. Todos os parâmetros perceptuais femininos correlacionaram com os quimográficos, exceto a I. A I de amostras masculinas correlacionou com RF. Conclusão: a caracterização vocal de pacientes com paralisia unilateral de pregas vocais foi possível usando a avaliação multidimensional da voz. / The goal is to characterize the voice and glottal vibration of patients with unilateral vocal fold paralysis. Method: this is a retrospective study that used database samples of Universidade de São Carlos\'s research group in medical engineering. Data from 35 adults patients (11 men and 24 women) and five were used. They were used five tools for multidimensional analysis. The first one was the self-assessment protocol (Voice Handicap Index - VHI). Physical, organic, emotional and total scores were extracted. Voice signals were used in analyzes: perceptual (visual analogue scale - VAS and parameters: overall degrees - G, roughness - R, Breathiness - B, tension - T and instability in phonation - I); traditional acoustic (perturbation parameters - Jitter and Shimmer), experimental (glottal cloure parameter - GC) and nonlinear (NL) dynamic vocal analysis (Qualitative: number of loops, regularity and convergence degree. Quantitative: standard deviation of dispersion - SDD and standard deviation of the axis - SDA). The fifth analysis it was performed through extraction of kymographic parameters generated of the High speed videolaryngoscopys (Opening reason - Or; Closed reason - Cr and Asymmetry). Statistical analysis: descriptive and Spearman correlation. Results: perceptual analysis characterized moderate intensity of vocal deviation. Roughness was the predominant kind of voice in males samples (55,84%) and Breathiness in females samples (63,59%). VHI\'s scores showed low impact of dysphonia in the patients\' quality of life. There was correlation between score physical and B in the male samples and between score physical and all perceptual parameters in the female samples. Both samples showed increased of perturbation parameters (Jitter: 0.90% men and 1.20% women and Shimmer: 3.30% men and 2.76% women) and GC decreased (0.28 men and women 0.25). Female samples showed correlation of Jitter and Shimmer with all perceptual parameters. Male samples showed correlation of Jitter with the R and Shimmer with R and T. GC correlation with B and I only male samples. Qualitative NL dynamic vocal analysis showed number of loops, regularity and convergence degree reduced, whereas quantitative analysis of SDD and SDA increased. There were correlations between perceptual parameters and of nonlinear dynamic vocal analysis for both the samples. Results of Kymographic parameters are Or: 1.55 for men and 1.50 for women; Cr: 0.37 for men and 0.18 for women; and Asymmetry: 23.68º for men and 9.85º for women. All perceptual parameters correlate with Or and Cr at female samples, except I. Male samples showed correlation between I and Cr. Conclusion: the vocal characterization of patients with unilateral vocal fold paralysis was possible using the multidimensional voice assessment.

Ce travail de recherche s’inscrit dans le domaine de la phonétique clinique et relève plus particulièrement de l’intelligibilité de la parole d’enfants sourds implantés cochléaires et d’enfants normo-entendants. L’intérêt de cette thèse réside dans la comparaison des productions d’enfants sourds implantés cochléaires et d’enfants normo-entendants, appariés en âge chronologique. Nous analysons des paramètres segmentaux et suprasegmentaux en production, tels que les voyelles du français standard et les consonnes fricatives /f, s, ʃ/, l’étude du débit de parole, l’intelligibilité de mots monosyllabiques auprès d’auditeurs naïfs, ou encore l’intelligibilité de la parole semi-spontanée. Nous nous intéressons également à la façon dont les productions des enfants implantés cochléaires sont perçues par un jury d’auditeurs naïfs (n=10 ; moyenne d’âge 37 ans). Le but premier de ce travail est de mettre en avant les différences ou similitudes majeures entre nos deux groupes en fonction de l’âge chronologique des enfants mais également en fonction du recul à l’implant, du suivi rééducatif et du mode de communication. Les enfants sourds porteurs d’un implant cochléaire présentent une qualité de la voix et de la parole comparable à celle d’enfants normo-entendants de même âge chronologique, mais non similaire puisque des différences acoustiques, segmentales et supra-segmentales ont été mises en avant. Il pourrait être intéressant d’étendre cette observation aux enfants implantés précocement. / This research task falls under the field of clinical phonetics and more particularly raises of the intelligibility of the word of deaf children with cochlear implant and normal-hearing children. The interest of this thesis lies in the comparison of the productions of deaf children with cochlear implant and normal-hearing children, paired in chronological age. We analyze segmental and suprasegmental parameters in production, such as the vowels of standard French and the fricative consonants /f, S, ʃ/, the study of the speech rate of word, the comprehensibility of monosyllabic words with naive listeners, or still the comprehensibility of the semi-spontaneous word. We are also interested in the way in which the productions of the established children cochléaires are perceived by a jury of naive listeners (n=10; average age 37 years). The primary purpose of this work is to highlight the major differences or similarities between the two groups based on the chronological age of the children but also on the age back to the cochlear implant, the rehabilitative monitoring and communication mode. Deaf children with cochlear implants present a quality of the voice and word comparable with that of normal-hearing children of the same chronological age, but non similar age since differences acoustic, segmental and suprasegmental were put ahead. It might be interesting to extend this observation to early implanted children.