|
|
REVIEW ARTICLE |
|
Year : 2019 | Volume
: 32
| Issue : 3 | Page : 763-769 |
|
Role of functional MRI in assessment of voice, language, and speech disorders
Mohamed Baraka1, Hossam El-Dessouky2, Eman Ezzat3, Marwa G A El-Hameed4
1 Phoniatrics Unit, Department of Otorhinolaryngology, Faculty of Medicine, Ain Shams University, Cairo, Egypt 2 Phoniatrics Unit, Department of Otorhinolaryngology, Faculty of Medicine, Cairo University, Cairo, Egypt 3 Phoniatrics Unit, Department of Otorhinolaryngology, Faculty of Medicine, Menoufia University, Shebin El-Kom, Menoufia Governorate, Egypt 4 Phoniatrics Unit, Department of Otorhinolaryngology, Shebin El-Kom Teaching Hospital, Shebin El-Kom, Menoufia Governorate, Egypt
Date of Submission | 31-Oct-2017 |
Date of Acceptance | 17-Dec-2017 |
Date of Web Publication | 17-Oct-2019 |
Correspondence Address: Marwa G A El-Hameed Phoniatrics Unit, Department of Otorhinolaryngology, Shebin El-Kom Teaching Hospital, Shebin El-Kom, Menoufia Governorate Egypt
Source of Support: None, Conflict of Interest: None | Check |
DOI: 10.4103/mmj.mmj_719_17
Objective The objective of this study was to review the role of functional MRI (fMRI) in assessment of voice, language, and speech disorders. Materials and Methods Medline databases (PubMed, Medscape, and ScienceDirect) and all materials available in the Internet from 1993 to 2017 were the sources of data. The initial search presented 267 articles of which 62 met the inclusion criteria. The articles studied the role of fMRI in assessment of communication disorders as regards voice, language, and speech. If the studies did not fulfill the inclusion criteria, they were excluded. Study quality assessment included whether ethical approval was gained, eligibility criteria specified, appropriate controls, adequate information, and defined assessment measures. Results Significant data were collected. Data were heterogeneous. Thus, a structured review was performed with the results tabulated. Conclusion fMRI brain imaging provides a new perspective on the organization of language, speech, and voice in the human brain. It can provide basic information about brain disease, and determine and monitor treatment outcomes of communication disorders. It gives a better definition of the distributed nature of the brain circuits involved and appreciation of the flexibility of these circuits in adapting to the different aspects of language, speech, and voice production.
Keywords: autism, brain mapping, dyslexia, dysphasia, dysphonia, stuttering
How to cite this article: Baraka M, El-Dessouky H, Ezzat E, El-Hameed MG. Role of functional MRI in assessment of voice, language, and speech disorders. Menoufia Med J 2019;32:763-9 |
How to cite this URL: Baraka M, El-Dessouky H, Ezzat E, El-Hameed MG. Role of functional MRI in assessment of voice, language, and speech disorders. Menoufia Med J [serial online] 2019 [cited 2024 Mar 28];32:763-9. Available from: http://www.mmj.eg.net/text.asp?2019/32/3/763/268836 |
Introduction | | |
The brain requires a steady supply of oxygen to metabolize glucose to provide energy. This oxygen is supplied by the component of the blood called hemoglobin. It was shown that the magnetic properties of hemoglobin depended on the amount of oxygen it carried. This dependency has given rise to the method for measuring activation using MRI, commonly known as functional MRI (fMRI) [1].
fMRI is a noninvasive method that aimed to measure neural activity while a person engages in cognitive tasks. Unlike MRI, which provides a static picture of the structure of the brain, fMRI provides both structural and functional images of the brain [2].
The primary form of fMRI used the blood-oxygen-level dependent as contrast. This was used to scan and map neural activity in the brain or spinal cord of humans by imaging the change in blood flow (hemodynamic response) related to energy use by brain cells, and thus fMRI action is based on studying the metabolic activity judged by changes in blood flow and oxygen consumption. Deoxygenated hemoglobin is more magnetic (paramagnetic) than oxygenated hemoglobin, which is virtually resistant to magnetism (diamagnetic). This difference leads to an improved MR signal as the diamagnetic blood interferes less with the magnetic MR signal. This improvement can be mapped to show which neurons are active at a time [3].
During fMRI experiment, an individual is placed in the magnet of an MRI machine, where various different kinds of stimuli may be administered in a controlled manner. For example, sounds may be played, visual scenes may be presented, and small motor movements or responses can be recorded. Two main experimental designs are in common use: the 'block' design and the 'event-related' design. fMRI is used in diagnosis and follow-up of individuals with language, speech, and voice disorders, which is important for validating the findings and demonstrating the remarkable consistency of the functional anatomy across the brain and its defects [4].
Task activation fMRI studies seek to induce different neural states in the brain as the visual, auditory, or other stimulus is manipulated during the scan, and activation maps are obtained by comparing the signals recorded during the different states. Therefore, it is important to collect each image in a snapshot mode to avoid head motion and physiological processes of respiration and cardiovascular functions from injecting noise signals unrelated to the neural processing being interrogated [5].
fMRI has been used clinically to map functional areas, check left–right hemispherical asymmetry in language and memory regions, check the neural correlates of a seizure, study how the brain recovers partially from a stroke, test how well a drug or behavioral therapy works, detect the onset of Alzheimer's, and note the presence of disorders such as depression.
Materials and Methods | | |
Search strategy
We reviewed papers on fMRI of voice, language, and speech disorders from Medline databases, which are PubMed, Medscape, ScienceDirect, and Autism community, and also materials available in the Internet from 1993 to 2017. We used fMRI/ADHD, fMRI/Aprexia, fMRI/autism (ASDs), fMRI/brain mapping, fMRI/dysclculia, fMRI/dyslexia, fMRI/dysphonia, fMRI/SLI, and fMRI/stuttering as searching terms.
Study selection
All the studies were independently assessed for inclusion. The published studies were included if they fulfilled the following inclusion criteria:
- Published in English language
- Published in peer-reviewed journals
- Focused on fMRI in assessment of communication disorders as regards voice, language, and speech
- Well-randomized cross-sectional studies, case–control studies, and systematic reviews
- If a study had several publications on certain aspects, we used the latest publication giving the most relevant data.
Data extraction
If the studies did not fulfill the above criteria, they were excluded. The analyzed publications were evaluated according to evidence-based medicine criteria using the classification of the US Preventive Services Task Force and UK National Health Service Protocol for evidence-based medicine in addition to the Evidence Pyramid.
US Preventive Services Task Force:
- Level I: evidence obtained from at least one properly designed randomized controlled trial
- Level II-1: evidence obtained from well-designed controlled trials without randomization
- Level II-2: evidence obtained from well-designed cohort or case–control analytic studies, preferably from more than one center or research group
- Level II-3: evidence obtained from multiple times series with or without the intervention. Marked results in uncontrolled trials might also be regarded as this type of evidence
- Level III: opinions of respected authorities, based on clinical experience, descriptive studies, or reports of expert committees.
Quality assessment
The quality of all the studies was assessed using Cochrane Systematic Review checklist type tool that included important factors, study design (prospective randomized cross-sectional studies, case–control studies), attainment of ethical approval, evidence of a power calculation, specified eligibility criteria (assessing the methodological quality or clinical practice guideline that report primary studies and systemic reviews), appropriate controls, adequate information, and specified assessment measures. It was expected that confounding factors would be reported and controlled for and appropriate data analysis made in addition to an explanation of missing data.
Data synthesis
A structured systematic review was performed with the results tabulated.
MRI data acquisition
The fMRI data were acquired using a multislice echo-planar imaging (EPI) sequence with whole-brain coverage (repetition time = 3000 ms, echo time = 30 ms, flip angle 85°, 44 slices, matrix size = 72 × 72, 3-mm isotropic voxels). Two consecutive runs of 120 volumes were collected for each participant, with the same fMRI speech paradigm repeated in each run. EPI data were collected continuously throughout each run [6].
Functional MRI analyses
The fMRI data were processed using a two-level random-effects analysis. At the first level, an event-related analysis was performed for each patient with a contrast of interest comparing the active condition (speaking single word) against the baseline condition (listening to single word). The first-level contrasts were then analyzed at the second level to generate group activation maps and comparisons [6].
Results | | |
A total of 62 studies were selected. The studies were deemed eligible by fulfilling the inclusion criteria. There was a high degree of heterogeneity regarding the role of fMRI in assessment of voice, language, and speech disorders.
fMRI gained the attention of researches in the field of communication disorders (voice, language, and speech). fMRI studies in voice disorders investigated the neural correlates for voluntary exhalation control and oral articulation. It also observed brain area activity during phonation and its malfunction in different forms of voice disorders [Table 1]. Multiple language disorders were investigated using fMRI for activation changes in different brain networks associated with language processing [Table 2]. fMRI also investigated speech highly specialized brain areas and its organized networks that connect several different areas of the brain and changes occurred due to its malfunction [Table 3].
Discussion | | |
The idea of fMRI use came originally from the magnetic principles of hemoglobin that is directly proportional to the amount of oxygen it carries. Areas of the brain with activity express increased hemodynamic response that could be imaged by fMRI. This type of scan is used to map neural activity in the brain or spinal cord by imaging the change in blood flow related to energy use by brain cells [3].
The use of fMRI has dominated other brain mapping researches, because it does not require people to ingest substances, or to be exposed to ionizing radiation. The brain activation can be graphically blotted with color-coding according to the strength of activation across the brain or the specific region [18].
Other modalities of brain imaging (structural and functional) have been used to diagnose diseases of the brain and the central nervous system, such as computed axial tomography scanners and PET. However, fMRI is preferred as it provides equivalent anatomical resolution and superior contrast resolution to that of computed axial tomography scanners. In addition, fMRI produces functional information similar to that of PET scanners but with superior anatomical detail. MRI scanners also provide imaging complementary to radiographic images as it distinguish soft tissue in both normal and diseased states without ionizing radiation [19].
fMRI study measures brain response to different kinds of stimuli sounds, visual scenes, and small motor movement. Two main experimental designs are in common use: the 'block' design and the 'event-related' design, where the brain of the patient is scanned repeatedly, usually using the fast imaging technique of EPI [2].
The capability of fMRI to give precised mapping of different brain areas and regions linked to critical functions such as speaking, moving, sensing, or planning could be useful to plan for surgery and radiation therapy of the brain. Clinicians also use fMRI to anatomically map the brain and detect the effects of tumors, stroke, head and brain injury, or diseases such as Alzheimer's disease, and developmental disabilities such as autism [20].
Human phonation is a laryngeal motor behavior that extends from reflexive laryngeal actions to highly skilled laryngeal sensorimotor control to support speech or singing. A component of normal phonation is the variation of voice pitch (habitual, high, and low). Integration of the sensory input and laryngeal motor output is required for pitch adaptation during vocalization [7].
fMRI helps in diagnosis and follow-up of some voice disorders such as spasmodic dysphonia and psychogenic dysphonia. In cases of spasmodic dysphonia, fMRI showed hyperactivity patterns at the bilateral precentral gyrus, right superior frontal gyrus, middle frontal gyrus, and inferior frontal gyrus, lingual gyrus, cingulate gyrus, superior temporal gyrus, thalamus, and bilateral cerebellum in the spasmodic dysphonia more than the healthy group [8].
Another fMRI study investigated cases with psychogenic dysphonia for pretreatment and post-treatment integrity of the amygdala–prefrontal circuit. They found increased activity in medial prefrontal regions, as well as decreased emotional reactivity in the amygdala during symptoms relative to recovery. In addition, fMRI displayed altered functions during symptoms, although the specific pattern of neural changes varied across tasks, showing higher activation upon scenic stimuli but lower activation upon neutral faces [9].
Language processing is a complex integral process of anatomical, neurological, and physiological functions of different brain regions. Many language disorders were identified with their negative impact on the affected individuals such as autism, specific language impairment (SLI), hearing loss, attention deficit hyperactivity disorder (ADHD), dysphasia, and learning disability.
- Autism spectrum disorders (ASDs): fMRI research in ASDs has focused on the neurobiology of core social impairment in which face perception is one of the earliest emerging social deficits. The study of autistic patients versus healthy controls in response to a neutral face task found decreased activation in the fusiform face area, occipital face area, and superior temporal sulcus [21]. In contrast to these findings, Hadjikhani et al. [22] reported significant activation in the fusiform face area in adults with ASDs and healthy controls
Piggot et al. [10] used fMRI with emotional face task to compare activation in adults with autism versus healthy controls. The autism group did not activate a cortical 'face area' when explicitly assessing emotions or the left amygdala and left cerebellum when implicitly processing facial emotions
- SLI is deficit in the production or comprehension of language despite normal cognitive development and educational opportunities that involves poor vocabulary, syntactic, morphological deficits, impairment in language comprehension, and phonological problems. fMRI study for SLI cases reported underactivation for Broca's area (BA 44/45), as well as in speech-related cortical and subcortical brain regions, whereas overactivation was observed in the posterior middle temporal gyrus and the left anterior insular (AI) cortex [23]
Badcock et al. [11] assessed the relationship between brain structure and function in 10 individuals with SLI; the left inferior frontal cortex showed increased gray matter and decreased functional activation, whereas the posterior temporal cortex showed both decreased gray matter and functional activation. Vannest et al. [24] reported that the use of a single Language paradigm with fMRI may not adequately reveal hemispheric and regional organization of language, particularly in the developing brain
- ADHD is an age-inappropriate problem with inattention, impulsivity, and hyperactivity affecting school-aged children. fMRI studies of ADHD response and interference inhibition reported consistent underactivation relative to controls in the right and left ventrolateral prefrontal cortex (VLPFC) and AI, supplementary motor area, and caudate while switching also elicits reduced activation in bilateral VLPFC/AI and basal ganglia. Moreover, poor inhibition performance correlates with decreased gray matter volumes in the VLPFC, AI, anterior cingulate cortex, striatal, and temporoparietal regions. However, ADHD child will need sedation to do fMRI properly that compromises the test and fMRI results [12]
Castellanos et al. [25] found that ADHD patients exhibited more significant resting-state brain activities in basic sensory and sensory-related cortices, and that dorsal anterior cingulate cortex had more significant resting-state functional connectivity with several other brain regions in the ADHD patients as compared with the controls
- Hearing loss is associated with delayed language disorder that could be corrected if the offending cause is treated. Tan et al. [26] used fMRI study to investigate the primary auditory cortex (A1) activation pre–post-implant in relation to improvement in hearing thresholds in young cochlear implant recipients. They found that multiple brain regions were more active postimplant at the angular gyrus, the supramarginal gyrus, the middle temporal gyrus, the cingulate gyrus, and regions in the prefrontal cortex
Unilateral sensory neural hearing loss cases showed robust activation in the auditory cortex bilaterally on fMRI. There was unexpected activation in the inferior frontal gyrus bilaterally and the cuneus. These results might represent a general cortical reorganization strategy present in patients with unilateral sensory neural hearing loss on either side [27]
- Developmental dyslexia is a specific disorder of reading development with impairment in the acquisition of reading skills despite normal intelligence, motivation, and adequate schooling. fMRI demonstrated changes in the reading networks depending on reading skill level and an evidence of characteristic hypoactivation of temporoparietal, as well as occipitotemporal brain areas in individuals with developmental dyslexia, an increase in activation in left frontal and right lateralized anterior brain areas. This hyperactivation has been suggested to reflect a compensatory mechanism for the dysfunctional reading system [13]
Evidence from many neuroimaging studies point toward a characteristic hypoactivation of left-hemispheric temporoparietal and occipitotemporal brain regions in children and adults with developmental dyslexia compared with typical reading controls. In developmental dyslexia, a hypoactivation of the left temporoparietal region of the brain seems to reflect an inability to map the sounds of languages (phonemes) to its written counterparts (letters/graphemes). After behavioral remediation, the children with dyslexia improved significantly in language processing and reading ability with coinciding increased cortical activity on fMRI [28]
- Developmental dyscalculia: learning disability in mathematics. Mathematical learning disabilities involves aberrations in multiple functional systems including brain systems implicated in visual form judgement, symbol recognition anchored in the ventral temporal–occipital cortex, quantity and magnitude processing anchored in the intraparietal sulcus region of the parietal cortex, as well as attention and working memory functions supported by a frontoparietal control network. Following acquisition of a variety of arithmetic skills, increased expertise was correlated with decreased frontal and increased parietal fMRI activation. The findings attributed to a switch from calculation to retrieval of learned answers from cortical memory stores [29]
- Aphasia is language limitations that include both expressive and receptive modalities. It is commonly caused by stroke. Several fMRI studies investigated patients of cerebral stroke with aphasia. Lukic et al. [14] on fMRI study showed that patient with stroke showed a shift in activation to the prelesional (parietotemporal) cortex during phonologic task performance.
Crosson et al. [30] found right posterior perisylvian activation for verb argument structure processing in three of five patients with Broca's aphasia who had damage to this region in the left hemisphere. However, data from 12 healthy control participants who were age-matched with the aphasic participants showed activation in the same region bilaterally. The right hemisphere activation noted for the aphasic participants may thus represent a premorbidly available network.
Stuttering is a speech event that contains intraphonemic disruption, part-word repetitions, monosyllabic whole-word repetitions, prolongations, and silent fixations (blocks). Stuttering has a negative impact upon quality of life, interpersonal relationships, employment opportunities, job performance and is associated with significant personal financial costs. The overall cerebellar activation and abnormal right lateralization in stutterers compared with controls during silent and oral word reading increases following fluency inducing therapy but then falls to below pretreatment levels in the long term. This was related to increased sensory or motor monitoring owing to reduced automaticity in articulatory movement sequences, even when reading silently, and there were alternations of resting-state functional connectivity between the basal ganglia and supplementary motor area [15].
The right frontal operculum has been reported to be the only region consistently overactivated in stutterers compared with controls during both reading and passive semantic decision tasks [16].
In contrast, Watkins et al. [17] reported two areas of overactivation in the right AI close to the right frontal operculum and identified decreased functional activity of the white matter underlying ventral premotor and motor cortical areas, which were underactive on fMRI and suggest decreased white matter integrity in tracts, which are important for execution of articulatory movements (by connections with M1) and for integration of articulatory planning and sensory feedback.
Conclusion | | |
fMRI brain imaging provides a new perspective on the organization of language, speech, and voice in the human brain. It can provide basic information about brain disease, and determine and monitor treatment outcomes of communication disorders. It gives a better definition of the distributed nature of the brain circuits involved and appreciation of the flexibility of these circuits in adapting to the different aspects of language, speech, and voice production.
Financial support and sponsorship
Nil.
Conflicts of interest
There are no conflicts of interest.
References | | |
1. | Attwell D, Iadecola C. The neural basis of functional brain imaging signals. Trends Neurosci 2002; 25:621–625. |
2. | Constable RT, McCarthy G, Allison T, Anderson AW, Gore JC. Functional brain imaging at 1.5 T using conventional gradient echo MR imaging techniques. Magn Reson Imaging 1993; 11:451–459. |
3. | Huettel SA, Song AW, McCarthy G. Functional magnetic resonance imaging. 2 nd ed. Sunderland, MA: Sinauer Associates Inc; 2009. |
4. | Cherkassky VL, Kana RK, Keller TA, Just MA. Functional connectivity in a baseline resting-state network in autism. Neuroreport 2006; 17:1687–1690. |
5. | Chang C, Glover GH. Relationship between respiration, end-tidal CO 2, and BOLD signals in resting-state fMRI. Neuroimage 2009; 47:1381–1393. |
6. | Morgan AT, Masterton R, Pigdon L, Connelly A, Liegeois FJ. Functional magnetic resonance imaging of chronic dysarthric speech after childhood brain injury: reliance on a left-hemisphere compensatory network. Brain 2013; 136(Pt 2):646–657. |
7. | Ludlow CL. Laryngeal Reflexes: Physiology, Technique, and Clinical Use. J Clin Neurophysiol 2015; 32:284–293. |
8. | Kryshtopava M, van Lierde K, Meerschman I, D'Haeseleer E, Vandemaele P, Vingerhoets G, et al. Brain activity during phonation in women with muscle tension dysphonia: an fMRI study. J Voice 2017; 31:675–690. |
9. | Spengler FB, Becker B, Kendrick KM, Conrad R, Hurlemann R, Schade G. Emotional dysregulation in psychogenic voice loss. Psychother Psychosom 2017; 86:121–123. |
10. | Piggot J, Kwon H, Mobbs D, Blasey C, Lotspeich L, Menon V, et al. Emotional attribution in high-functioning individuals with autistic spectrum disorder: a functional imaging study. J Am Acad Child Adolesc Psychiatry 2004; 43:473–480. |
11. | Badcock NA, Bishop DV, Hardiman MJ, Barry JG, Watkins KE. Co-localisation of abnormal brain structure and function in specific language impairment. Brain Lang 2011; 120:310–320. |
12. | Lei D, Du M, Wu M, Chen T, Huang X, Du X, et al. Functional MRI reveals different response inhibition between adults and children with ADHD. Neuropsychology 2015; 29:874–881. |
13. | Houde O, Rossi S, Lubin A, Joliot M. Mapping numerical processing, reading, and executive functions in the developing brain: an fMRI meta-analysis of 52 studies including 842 children. Dev Sci 2010; 13:876–885. |
14. | Lukic S, Barbieri E, Wang X, Caplan D, Kiran S, Rapp B, et al. Right hemisphere grey matter volume and language functions in stroke aphasia. Neural Plast 2017; 2017:5601509. |
15. | Xuan Y, Meng C, Yang Y, Zhu C, Wang L, Yan Q, et al. Resting-state brain activity in adult males who stutter. PLoS One 2012; 7:e30570. |
16. | Neumann K, Preibisch C, Euler HA, von Gudenberg AW, Lanfermann H, Gall V, et al. Cortical plasticity associated with stuttering therapy. J Fluency Disord 2005; 30:23–39. |
17. | Watkins KE, Smith SM, Davis S, Howell P. Structural and functional abnormalities of the motor system in developmental stuttering. Brain 2008; 131(Pt 1):50–59. |
18. | Yan S, Nie L, Wu C, Guo Y. An approximation approach to measurement design in the reconstruction of functional MRI sequences. In: International Conference on Brain and Health Informatics; 29–31 October 2013; Maebashi, Japan. Cham, Switzerland: Springer International Publishing. pp. 115–125. |
19. | Jezzard P, Buxton RB. The clinical potential of functional magnetic resonance imaging. J Magn Reson Imaging 2006; 23:787–793. |
20. | Subbaraju V, Sundaram S, Narasimhan S. Identification of lateralized compensatory neural activities within the social brain due to autism spectrum disorder in adolescent males. Eur J Neurosci 2017; 2017:13634. |
21. | Humphreys K, Hasson U, Avidan G, Minshew N, Behrmann M. Cortical patterns of category-selective activation for faces, places and objects in adults with autism. Autism Res 2008; 1:52–63. |
22. | Hadjikhani N, Joseph RM, Snyder J, Chabris CF, Clark J, Steele S, et al. Activation of the fusiform gyrus when individuals with autism spectrum disorder view faces. Neuroimage 2004; 22:1141–1150. |
23. | Liegeois F, Baldeweg T, Connelly A, Gadian DG, Mishkin M, Vargha-Khadem F. Language fMRI abnormalities associated with FOXP2 gene mutation. Nat Neurosci 2003; 6:1230–1237. |
24. | Vannest J, Karunanayaka PR, Schmithorst VJ, Szaflarski JP, Holland SK. Language networks in children: evidence from functional MRI studies. Am J Roentgenol 2009; 192:1190–1196. |
25. | Castellanos FX, Margulies DS, Kelly C, Uddin LQ, Ghaffari M, Kirsch A, et al. Cingulate-precuneus interactions: a new locus of dysfunction in adult attention-deficit/hyperactivity disorder. Biol Psychiatry 2008; 63:332–337. |
26. | Tan L, Chen Y, Maloney TC, Caré MM, Holland SK, Lu LJ. Combined analysis of sMRI and fMRI imaging data provides accurate disease markers for hearing impairment. Neuroimage Clin 2013; 3:416–428. |
27. | Schmithorst VJ, Holland SK, Ret J, Duggins A, Arjmand E, Greinwald J. Cortical reorganization in children with unilateral sensorineural hearing loss. Neuroreport 2005; 16:463–467. |
28. | Jasinska KK, Petitto LA. Development of neural systems for reading in the monolingual and bilingual brain: new insights from functional near infrared spectroscopy neuroimaging. Dev Neuropsychol 2014; 39:421–439. |
29. | Kucian K, Von AM. Developmental dyscalculia. Eur J Pediatr 2015; 174:1–13. |
30. | Crosson B, McGregor K, Gopinath KS, Conway TW, Benjamin M, Chang YL, et al. Functional MRI of language in aphasia: a review of the literature and the methodological challenges. Neuropsychol Rev 2007; 17:157–177. |
[Table 1], [Table 2], [Table 3]
|