| | Acoustic quality of cry in very-low-birth-weight infants at the age of 1 1/2 yearsAccepted 16 March 2006. Abstract BackgroundInfant cry characteristics reflect the integrity of the central nervous system. Previous studies have shown that preterm infants and infants with neurological conditions have different cry characteristics such as fundamental frequency compared to healthy full-term infants. Cry characteristics of preterm infants after the first year of life have not been studied. AimsThe aim of this study was to assess the quality of cry in 1 1/2-year-old very-low-birth-weight infants (VLBWI, ≤ 1500 g at birth). Study subjects and designStudy groups included 21 VLBWI and 25 healthy full-term controls. Thirty seconds of pain cry after vaccination was recorded at well-baby clinics. The first cry utterance was acoustically analyzed using Praat software. The quality of cry was compared between the groups. In addition, the association of cry quality to patient characteristics, to developmental outcome, and to findings in brain imaging studies of the VLBWI was studied. ResultsThe cry response was elicited in 20 of the 21 VLBWI and in 20 out of 25 full-term infants. VLBWI had higher minimum fundamental frequency and fourth formant values. Patient characteristics that were associated with cry quality were 5-min Apgar scores, the occurrence of bronchopulmonary dysplasia, Bayley Psychomotor Index scores at 12 months, and current weight and head circumference. ConclusionsDifferences found between the study groups were not explained primarily by brain pathology or by patient characteristics, so it seems that prematurity has an impact on cry quality still at the age of 1 1/2 years. 1. Introduction  The central nervous system (CNS) and the vagal tone regulate the function of anatomical structures of the larynx and the vocal tract, producing acoustic properties of cry sound [1], [2]. Consequently, the acoustic characteristics of infant cry can be affected by CNS pathology. Infants with diffuse brain damage have been found to have a higher threshold and a longer latency between a painful stimulation and a cry response compared to healthy infants [3], [4]. It has been shown that neurological and metabolic disturbances [5], [6], high bilirubin level [7], [8], and prenatal lead [9], cocaine [10], [11] and selective serotonin release inhibitor [12] exposure affect cry characteristics such as fundamental frequency. Preterm infants have been reported to have a higher fundamental frequency and shorter cries [5], [13] during the first few weeks of life compared to full-term infants. In addition, preterm infants have been found to have several other differences in cry acoustics such as more glide and biphonation, less glottal plosives [5], [13], more variability in the amplitude [8], and more harmonic doubling and noise concentration [14]. Differences have also been shown in long time average spectrum analysis of cry of preterm infants [8], [15]. The acoustic features in the cry of preterm infants have also been found to correlate to the developmental outcome of the child at 18 months and 5 years [16]. Most studies on acoustic quality of crying of preterm infants have focused on the time period of few days or weeks after birth. With increasing age, crying is thought to change into more intentional vocalisation and the fundamental frequency has been shown to rise up to 12 months of age [17]. Our hypothesis was that there are differences at the age of 1 1/2 years in the quality of cry in preterm infants compared to healthy full-term infants. 2. Methods 2.1. Subjects The study group consisted of 21 very-low-birth-weight infants (VLBWI, birth weight ≤1500 g and gestational age <37 weeks at birth), born in Turku University Hospital in 2002 and 2003. The study is a part of a larger multidisciplinary PIPARI study. The exclusion criteria were (1) neither parent spoke Finnish or Swedish, and (2) a long distance from home, i.e. the family living outside the hospital catchment area. 5 VLBWI had normal brain anatomy and 16 did not, as described in the Brain imaging section of this article. A control group of 25 healthy full-term infants was recruited at well-baby clinics at their appointment for vaccination (with combined vaccine against mumps, measles, and rubella, i.e. MMR) at 1 1/2 years. Control group consisted of children who had developed normally, had no major medical conditions, and had their appointments the same day and at the same well-baby clinic with the VLBWI participating in the study. The motor and cognitive development of the controls was evaluated by the physician or the nurse at the well-baby clinic. Exclusion criteria for the controls were the following: (1) use of illicit drugs by the mother during pregnancy, (2) gestational age <37 weeks at birth, (3) birth weight below −2.0 SD from the mean of Finnish growth charts, (4) any diagnosis of a condition potentially affecting normal development, (5) height or head circumference below −2.0 S.D. from the mean of Finnish growth charts at 1 1/2 years of age, or (6) abnormal motor or cognitive development evaluated by the physician or the nurse at the well-baby clinic. The study protocol was approved by the Ethics Review Committee of the Hospital District of the South-Western Finland. The parents gave written informed consent to participate in the study. 2.2. Recording and acoustic analysis We recorded crying after the MMR vaccine was given at the age of 1 1/2 years by a nurse at a well-baby clinic. The MMR vaccination coverage in Finland is approximately 97% [18]. The infants' cry response was recorded for 30 s using SHARP MD-SR50H(S) portable digital minidisk recorder and Hama LM-09 microphone for the first 10 samples and stereo SONY ECM-MS907 microphone for the rest with microphone being held approximately 10 cm from the infant's mouth for the first 10 samples and 30 cm for the rest of the recordings. Stopwatch with a beep sound was used to indicate the skin perforation by the needle and to measure the total duration of crying. The caretaker of the child was asked not to comfort the child during the 30 s of the recording procedure. One researcher (L.R.) was present at all the well-baby clinic appointments, recorded the cries, and collected the information about the procedure including the baseline arousal state of the child before vaccination, and the time and site (buttocks or thigh) of vaccination. The baseline arousal state was divided into three mutually exclusive categories (content, not content but not crying, crying). Some of the children received another vaccination after the 30-s recording. As it might have influenced the total duration of crying, the number of inoculations was recorded. The soothing techniques used after the recording by the parent were written down. The background information of medical history, drug abuse during pregnancy, birth weight, gestational age, 5 min Apgar score, current weight, height, head circumference, and an assessment of the current developmental status of the child were obtained from the medical records of the well-baby clinic. Cry analysis was performed blinded to the infants' medical history. The acoustic analysis of the cry samples was performed by A.L. using Praat software version 4.2.05 [19]. The sampling rate was 44,100 Hz, and the signal was low pass filtered at 10,000 Hz. The first cry utterance after the vaccination was analyzed. A cry utterance was defined as a cry during expiratory phase lasting for at least 0.5 s. The choice of the utterance analyzed was made based on the continuity of the melody curve and overall structure of the spectrogram in the vicinity of the pain stimulus. The melody-type of the curve was defined as flat, rising, falling, rising–falling, falling–rising, rising–falling–rising, falling–rising–falling, repeated rising–falling, or repeated falling–rising. Values of fundamental frequency (mean, minimum and maximum) and formants F1–4 were measured from the pitch curve. The occurrence of shift, diplophonia, biphonation, hyperphonation, shatter, glide, vibrato, glottal roll, furcation, noise, and breaks was observed from the narrow band spectrogram. Long term average spectrum (LTA) was used to calculate four spectral moments; center of gravity, standard deviation, skewness, and kurtosis. First four formants were measured from a single time point of the sample. Latency was measured from the spectrograms/oscillograms by identifying the beep of the stopwatch both by listening and from the graph and measuring time from the beep to start of the first audible and visible cry utterance. The operational definitions of the acoustic variables are given in Table 1, Table 2.  | Cry utterance | First perceivable phonation segment after the vaccination. Duration at least 500 ms. | | | Fundamental frequency (F0) mean | Represents the rate at which vocal cords vibrate and corresponds to the perceivable pitch of the sound. The mean value was calculated over a distance of the analyzed utterance. | | | F0 minimum and maximum | The lowest and the highest measurable point of the fundamental frequency seen on the spectrogram. | | | Formants F1, F2, F3, and F4 | Airway above the glottis acts as a resonator accentuating certain harmonics in the phonation. Lowest accentuated frequency is termed first formant=F1, the next F2, etc. Formants were measured from a selected point of the spectrogram where they were most clear. | | | Spectral moments | Described as the distribution of harmonics/energy of the cry seen in the long-term average spectrum. | | | (1) Center of gravity (COG) | The point of balance | | | (2) Standard deviation (S.D.) | The spreading of energy | | | (3) Skewness | The asymmetricality of the spectrum looking at the point of gravity center i.e. tilt | | | (4) Kurtosis | Peakedness of energy distribution | | | | |
| Shift | Sudden shifts of fundamental frequency cutting the melody curve (the shift part duration at least 0.2 s). | | | Diplophonia | Parallel lines between F0 and its harmonics. | | | Biphonation | Double phonation caused by false vocal folds activation producing their own harmonic structure occurring together but not parallel with F0 and its harmonics. | | | Hyperphonation | Cry segments of very high F0 (F0>1000 Hz). | | | Shatter/distortion | Breaking of phonation with high intensity and unclear harmonic structure. | | | Glide | Rapid change of F0 (600 Hz or more per second). | | | Vibrato | At least four rapid up and down movements of F0. | | | Glottal roll | Small vibrations of low intensity (duration at least 0.2 s). | | | Furcation | A split in F0 seen in the spectrogram where harmonic breaks into a series of separate lines. | | | Noise | Noise segments without periodicity, i.e. no visible F0 and its harmonics. | | | Breaks | Interruptions in the phonation caused for example by coughing. | | | | |
In addition, the cry samples were divided into three groups–markedly abnormal cry, moderately abnormal cry, and normal cry–according to cry characteristics presented in Table 3. This grouping was done blinded to the infants' medical history to find out whether cry characteristics considered abnormal in previous studies can be used to distinguish VLBWI with or without brain findings from controls. | | | | Grouping | Criteria used | |
|---|
| Markedly abnormal cry | Unstable cry with high frequency | | | Mean fundamental frequency (F0) over 600 Hz and hyperphonation | | | Gliding | | | Frequent noise segments | | | Melody curve rise with marked unstability in signal | | | Furcation | | | High first formant (>1500 Hz) | | | Moderately abnormal cry | High frequency cry with some instability or monotonic cry | | | Mean F0 over 600 Hz | | | Max F0 over 980 Hz | | | Melody curve flat with no changes | | | Melody curve rise or fall–rise | | | Duration of cry utterance <1.3 s or duration >5 s | | | Shatter | | | Hyperphonation | | | Normal cry | Stable and clear signal with no sharp or multiple changes in the pitch or the type of phonation | | | Mean F0 under 500 Hz (excl. cries with min pitch <100 Hz) | | | Melody curve (fall)–rise–fall and glottal roll/ vibrato in the end of the cry utterance | | | Diplophonia/glide but otherwise stable signal | | | No noise | | | The uncompatibles | | | | |
2.3. Brain imaging 2.3.1. Ultrasound examinations and magnetic resonance imaging of the brain Out of the 21 VLBWI, 16 had one or more intracranial findings either in the cranial ultrasound (US) or in the magnetic resonance imaging (MRI). US examinations were performed by a neonatologist in neonatal intensive care unit for all study VLBWI at 3 to 5 days, at 7 to 10 days, and at 1 month of age and, thereafter, monthly until discharge from the hospital. The detailed US methods used are reported by Maunu et al. [20]. The findings in US were intraventricular hemorrhages of grades I (n=4), II (n=2), and IV (n=1), multiple periventricular cysts (n=1), caudothalamic cysts (n=3) and ventriculomegaly with 1, 2, 3, or 4 dilated horns (n=6, 2, 1, and 2, respectively). Ultrasound imaging was normal for two VLBWI with findings in MRI. Magnetic Resonance Imaging (MRI) of the brain was performed at term at the same day with the US examination for all study VLBWI. One pediatric neuroradiologist (R.P.) analyzed the MRI findings blinded to the clinical information and to the result of the US examinations of the infant. For detailed MRI methods see Maunu et al. [20]. The findings seen in MRI were caudothalamic cysts (n=4), capsula interna injury (n=3), other white matter injury (n=2), corpus callosum hypoplasia (n=1), cerebellum hypoplasia (n=1), and ventriculitis (n=1). The width of extracerebral space was 5 mm in two patients and above 5 mm in one patient. MRI was normal for 5 VLBWI with abnormal findings in US. 2.4. Neurological and cognitive testing Hammersmith infant neurological examination was performed to VLBWI at 12 months corrected age by a physician. This test includes sections assessing cranial nerve function, posture, movements, tone, reflexes and reactions, motor milestones, and behaviour. Section scores were then added together with the minimum global score being 0 and maximum 78. Hammersmith Infant Scale global score equal or above 73 was regarded as optimal and the scores below 73 as suboptimal based on term normative data [21], [22], [23]. Bayley Scales of Infant Development, 2nd edition [24], was performed to all preterm participants at 12 and 24 months of corrected age by a psychologist to calculate the Mental Developmental Index (MDI) and the Psychomotor Index (PDI). The reaction to auditory stimulus (turning head towards a sound) was tested for all infants at the well-baby clinics. In addition to this all preterm infants had auditory stimulus tested at 36 weeks of gestational age, at term, and at 1 and 2 months of corrected age. Thirteen of the VLBWI were also tested by brainstem auditory evoked potential (BAEP) measurement. 2.5. Statistical analysis SAS (version 9.1; SAS Institute, Cary, NC) and SPSS for Windows (version 12.0; SPSS, Chicago, IL) statistical packages were used for analysis. Differences were considered statistically significant if p-value was below 0.05. The differences between VLBWI and controls were tested using two-sample t-test for normally distributed continuous variables and Mann–Whitney U-test for non-normally distributed continuous variables. Analysis of covariance (ANCOVA) was used to further study the group differences in continuous variables adjusted for brain pathology, corrected age, and current weight. ANCOVA analyses were conducted both with and without outliers. A Kruskal–Wallis test was used to compare duration of crying between different arousal states before vaccination. Categorical variables were compared using Pearson's χ2 or Fisher's exact test. The following patient characteristics were examined as possible predictors of abnormal cry characteristics in exploratory analysis: gender, low 5-min Apgar scores (<5), duration of ventilator treatment, bronchopulmonary dysplasia (BPD), Bayley scores at 12 and at 24 months of corrected age, Hammersmith Infant Scale rating at 12 months of corrected age as a dichotomised variable, current weight, height, and head circumference. The effects of weight, height, head circumference, gender, and low Apgar scores were studied using the whole sample and separately within VLBWI. Other variables were examined only within VLBWI. The correlations between continuous cry characteristics and patient characteristics were studied using Spearman's correlation coefficient. Comparisons of two cry characteristics were performed using two-sample t-test or Mann–Whitney U-test as appropriate. Dichotomic outcomes with at least four events were analyzed using χ2-test or Fisher's exact test for the categorical and by logistic regression for the continuous predictors. 3. Results 4. Discussion This study shows that there are some differences in the acoustic quality of cry between VLBWI and healthy full-term children at the age of 1 1/2 years. The minimum fundamental frequency was higher in VLBWI, and this difference remained significant after adjusting for corrected age, weight, and brain findings. The fourth formant values were also higher in the VLBWI. In addition, cry quality was affected by 5-min Apgar scores, BPD, Bayley Psychomotor Index at 12 months, and the current weight and head circumference of the child. We recorded pain cry after MMR vaccination to get as standardised a cry stimulus as possible for all infants. This way we could also avoid causing unnecessary discomfort to the child, as MMR is part of the Finnish vaccination schedule. Furthermore, pain cry has been the most commonly studied cry type used to compare group differences in cry acoustics. It has been shown that differences even in the type of vaccine injected can affect cry quality [25]. Stress level has been shown to affect cry acoustics through vagal tone [2]. In our study, the baseline arousal state of the infant could not be standardised and crying as the baseline arousal state prolonged the total duration of crying. As some infants were already crying before the vaccination and all infants did not respond with cry at all, the latency time was measurable for a part of the sample only. The MMR vaccination in Finland is given at a well-baby clinic visit when the children are close to 1 1/2 years of chronological age. That is why there were differences in the corrected age between VLBWI and control infants in our study. We took this difference into consideration by adjusting the continuous variables as mentioned above. Although acoustic theory suggests that increasing size decreases fundamental frequency [1], fundamental frequency has been shown to rise from the age of 0 to 12 months in longitudinal studies [17], [26]. This rise has been interpreted as an increasing control of cry production as the child grows older. One might assume that the increase in fundamental frequency continues after 12 months of age. In our study, however, minimum fundamental frequency value was higher in premature infants than in controls even though they were younger in their corrected age at the time of the recordings. The effect of prematurity on minimum fundamental frequency remained after adjusting for the corrected age, brain findings, and weight. In contrast to studies of newborn infants [5], [6], [13], the maximum or the mean fundamental frequency at 1 1/2 years of age was not related to prematurity or brain pathology. Patient characteristics had effects on cry quality in our study. Most effects could be seen within the VLBWI only. In previous study with younger infants, duration of respiratory assistance has been found to be related to cry duration, the occurrence of harmonic doubling and vibrato [27]. In our study, BPD had an effect on cry quality but the duration of ventilator treatment did not. We found an association between higher current weight and longer cry utterance duration in VLBWI. However, there was no significant difference between the groups in either the duration of cry utterance or the total cry duration. In previous study, the duration of phonation in low-birth-weight infants has been found to be longer than in normal weight infants [5]. Interestingly, also Bayley Psychomotor Index and 5-min Apgar scores were associated with cry quality. In our study, the only variable that differed significantly between study groups and was also affected significantly by patient characteristics was the fourth formant frequency. Other cry variables that differed significantly between study groups were not affected significantly by the patient characteristics. Because only one infant had a hearing deficit, the effect of poor hearing on cry was impossible to study further. This one infant was classified, blinded to patient characteristics, to the moderately abnormal cry group. In earlier studies, the cry utterance duration has been found to be longer in hearing impaired infant in comparison to normally hearing infants. In addition, some differences in fundamental frequency and in the number of unvoiced utterances have been found [28]. The three children with CP were all classified to the moderately abnormal cry group. This study is limited by the relatively small sample size, which restricts the generalizability of findings, reduces the power of statistical methods, and makes chance findings possible. In conclusion, some effect of premature birth on cry quality can still be seen at the age of 1 1/2 years. It seems that the differences in the quality of cry of VLBWI compared to full-term infants relate to prematurity itself, not primarily to brain injury or differences in other patient characteristics. The specific factors in VLBWI causing the differences in cry sound remain unsolved. Potential differences, genetic as well as environmental, in the pathways controlling cry acoustics include several steps beginning from the central nervous system down to the anatomy of the vocal cords. The significance of the most abnormal cry responses will be elucidated by the further follow-up of the children. Acknowledgements Members of PIPARI study group are Satu Ekblad, RN, Eeva Ekholm, MD, PhD, Leena Haataja, MD, PhD, Pentti Kero, MD, PhD, Jarkko Kirjavainen, MD, PhD, Harry Kujari, MD, Helena Lapinleimu MD, PhD, Liisa Lehtonen, MD, PhD, Hanna Manninen, MD, Jaakko Matomäki, Student of Statistics, Jonna Maunu, MD, Pekka Niemi, PhD, Pertti Palo, MD, PhD, Riitta Parkkola, MD, PhD, Riikka Korja, MSc, Jorma Piha, MD, PhD, Annika Pihlgren, MSc, Liisi Rautava, Medical Student, Päivi Rautava, MD, PhD, Milla Reiman, Medical Student, Hellevi Rikalainen, MD, PhD, Taija Saanisto, MSc (statistician), Katriina Saarinen, Physiotherapist, Elina Savonlahti, MD, Matti Sillanpää, MD, PhD, Suvi Stolt, Phil.Lic, Petriina Munck, MA, Päivi Tuomikoski-Koiranen, RN, and Tuula Äärimaa, MD, PhD. References [1]. [1]Golub HL, Corwin MJ. A physioacoustic model of the infant cry. In: Lester BM, Boukydis CFZ editor. Infant crying: theoretical and research perspectives. New York: Plenum Press; 1985;p. 59–82. [2]. [2]Porter FL, Porges SW, Marshall RE. Newborn pain cries and vagal tone: parallel changes in response to circumcision. Child Dev. 1988;59(2):495–505. MEDLINE |
CrossRef
[3]. [3]Karelitz S, Fisichelli VR. The cry thresholds of normal infants and those with brain damage. J Pediatr. 1962;61:679–685. Abstract |
Full-Text PDF (806 KB)
|
CrossRef
[4]. [4]Fisichelli VR, Karelitz S. The cry latencies of normal infants and those with brain damage. J Pediatr. 1963;62:724–734. Abstract |
Full-Text PDF (493 KB)
|
CrossRef
[5]. [5]Michelsson K. Cry analyses of symptomless low birth weight neonates and of asphyxiated newborn infants. Acta Paediatr Scand. 1971;(Suppl 216):10–45. [6]. [6]Michelsson K, Raes J, Thoden CJ, Wasz-Höckert O. Sound spectrographic cry analysis in neonatal diagnostics. An evaluative study. J Phon. 1982;10:79–88. [7]. [7]Golub HL, Corwin MJ. Infant cry: a clue to diagnosis. Pediatrics. 1982;69(2):197–201. [8]. [8]Rapisardi G, Vohr B, Peucker M, Lester B. Assessment of infant cry variability in high-risk infants. Int J Pediatr Otorhinolaryngol. 1989;17:19–29. MEDLINE |
CrossRef
[9]. [9]Rothenberg SJ, Cansino S, Sepkoski C, Torres LM, Medina S, Schnaas L, et al. Prenatal and perinatal lead exposures alter acoustic cry parameters of neonate. Neurotoxicol Teratol. 1995;17(2):151–160. MEDLINE |
CrossRef
[10]. [10]Lester BM, Corwin MJ, Sepkoski C, Seifer R, Peucker M, McLaughlin S, et al. Neurobehavioral syndromes in cocaine-exposed newborn infants. Child Dev. 1991;62(4):694–705. MEDLINE |
CrossRef
[11]. [11]Corwin MJ, Lester BM, Sepkoski C, McLauglin S, Kayne H, Golub H. Effects of in utero cocaine exposure on newborn acoustical cry characteristics. Pediatrics. 1992;89(6):1199–1203. [12]. [12]Zeskind PS, Oberlander TF, Grunau RE, Fitzgerald C, Garber KA. Continuing effect of prenatal SSRI exposure detected by spectral analysis of infant cry sounds at two months of age [abstract]. Pediatric Academic Societies' Annual Meeting, May 14–17, Washington, DC. [13]. [13]Michelsson K, Järvenpää AL, Rinne A. Sound spectrographic analysis of pain cry in preterm infants. Early Hum Dev. 1983;8:141–149. MEDLINE |
CrossRef
[14]. [14]Robb MP. Bifurcation and chaos in the cries of full-term and preterm infants. Folia Phoniatr Logop. 2003;55:233–240. MEDLINE |
CrossRef
[15]. [15]Goberman AM, Robb MP. Acoustic examination of preterm and full-term infant cries: the long time average spectrum. J Speech Lang Hear Res. 1999;42:850–861. MEDLINE [16]. [16]Lester BM. Developmental outcome from acoustic cry analysis in term and preterm infants. Pediatrics. 1987;80(4):529–534. [17]. [17]Gilbert HR, Robb MP. Vocal fundamental frequency characteristics of infant hunger cries: birth to 12 months. Int J Pediatr Otorhinolaryngol. 1996;34:237–243. Abstract |
Full-Text PDF (475 KB)
|
CrossRef
[18]. [18]National Health Institute of Finland, webpage: http://www.ktl.fi/portal/suomi/osiot/terveyden_ammattilaisille/rokottaminen/rokotuskattavuus. [19]. [19]Boersma P, Weenink D. Praat Software, 2003, Institute of Phonetic Sciences, University of Amsterdam, Netherlands. Webpage: http://www.fon.hum.uva.nl/praat/. [20]. [20]Maunu J, Kirjavainen J, Rikalainen H, Parkkola R, Lapinleimu H, Haataja L et al., and the PIPARI Study Group. Relation of brain injury to infant crying in very preterm infants. Pediatrics, in press. [21]. [21]Frisone MF, Mercuri E, Laroche S, Foglia C, Maalouf EF, Haataja L, et al. Prognostic value of the neurologic optimality score at 9 and 18 months on preterm infants born before 31 weeks gestation. J Pediatr. 2002;140(1):57–60. Abstract | Full Text |
Full-Text PDF (83 KB)
|
CrossRef
[22]. [22]Haataja L, Mercuri E, Regev R, Cowan F, Rutherford M, Dubowitz V, et al. Optimality score of the neurological examination of the infant at 12 and 18 months of age. J Pediatr. 1999;135:153–161. Abstract | Full Text |
Full-Text PDF (320 KB)
|
CrossRef
[23]. [23]Haataja L, Mercuri E, Guzzetta A, Rutherford M, Counsell S, Frisone MF, et al. Neurological examination in infants with hypoxic-ischaemic encephalopathy at age 9–14 months; use of optimality scores and correlation with MRI findings. J Pediatr. 2000;138(3):332–337. Abstract | Full Text |
Full-Text PDF (74 KB)
|
CrossRef
[24]. [24]Bayley N. Bayley scales of infant development manual. 2nd edition. San Antonio: Psychological Corporation; 1993;. [25]. [25]Ipp M, Cohen E, Goldbach M, Macarthur C. Effect of choice of measles–mumps–rubella vaccine on immediate vaccination pain in infants. Arch Pediatr Adolesc Med. 2004;158(4):323–326. MEDLINE |
CrossRef
[26]. [26]Rothgänger H. Analysis of the sounds of the child in the first year of age and a comparison to the language. Early Hum Dev. 2003;75:55–69. Abstract | Full Text |
Full-Text PDF (405 KB)
|
CrossRef
[27]. [27]Cacace AT, Robb MP, Saxman JH, Risemberg H, Koltai P. Acoustic features of normal-hearing pre-term infant cry. Int J Pediatr Otorhinolaryngol. 1995;33:213–224. Abstract |
Full-Text PDF (896 KB)
|
CrossRef
[28]. [28]Clement CJ, Koopmans-van Beinum FJ, Pols LCW. Acoustical characteristics of sound production of deaf and normally hearing infants. In: Proceedings of the Fourth International Conference on Spoken Language Processing, 1996 Oct 3–6; Philadelphia, PA, USA. 1996;. a Department of Pediatrics, Turku University Hospital, Kiinanmyllynkatu 4-8, 20520 Turku, Finland b Department of Phonetics, University of Turku, Finland c Department of Radiology, Turku University Hospital, Finland d Department of Pediatric Neurology, Turku University Hospital, Finland  Corresponding author.
PII: S0378-3782(06)00088-0 doi:10.1016/j.earlhumdev.2006.03.004 © 2006 Elsevier Ireland Ltd. All rights reserved. | |
|
FULL TEXT