Prediction of Unilateral Vocal Cord Paralysis Patients Through Machine Learning Analysis of Acoustic Parameters: A Preliminary Study

Background and Objectives@#The purpose of this study is to evaluate value of diagnostic tool for vocal cord palsy utilizing artificial intelligence without laryngoscopeMaterials and Method A dataset consisting of recordings from patients with unilateral vocal cord paralysis (n=54) as well as normal individuals (n=163). The dataset included prolonged pronunciations of the vowels /ah/, /u/, /i/, and vocal cord data from paralyzed patients. Various acoustic parameters such as Mel-frequency cepstral coefficients, jitter, shimmer, harmonics-to-noise ratio, and fundamental frequency statistics were analyzed. The classification of vocal cord paralysis encompassed paralysis status, paralysis degree, and paralysis location. The deep learning model employed the leave-one-out method, and the feature set with the highest performance was selected using the following methods. @*Results@#Vocal Cord Paralysis Classifier: The classifier accurately distinguished normal voice from vocal cord paralysis, achieving an accuracy and F1 score of 1.0. Paralysis Location Classifier: The classifier accurately differentiated between median and paramedian vocal cord paralysis, achieving an accuracy and micro F1 score of 1.0. Breathiness Degree Classifier: The classifier achieved an accuracy of 0.795 and a mean absolute error of 0.2857 in distinguishing different degrees of breathiness. @*Conclusion@#Although the small sample size raises concerns of potential overfitting, this preliminary study highlights distinctive acoustic features in cases of unilateral vocal fold paralysis compared to those of normal individuals. These findings suggest the feasibility of determining the presence, degree, and location of paralysis through the utilization of acoustic parameters. Further research is warranted to validate and expand upon these results.

Reliability and Validity of the Comprehensive Limb and Oral Apraxia Test: Standardization and Clinical Application in Korean Patients With Stroke

OBJECTIVE: To develop and standardize the Limb and Oral Apraxia Test (LOAT) for Korean patients and investigate its reliability, validity, and clinical usefulness for patients with stroke. METHODS: We developed the LOAT according to a cognitive neuropsychological model of limb and oral praxis. The test included meaningless, intransitive, transitive, and oral praxis composed of 72 items (56 items on limb praxis and 16 items on oral praxis; maximum score 216). We standardized the LOAT in a nationwide sample of 324 healthy adults. Intra-rater and inter-rater reliability and concurrent validity tests were performed in patients with stroke. We prospectively applied the LOAT in 80 patients and analyzed the incidence of apraxia. We also compared the clinical characteristics between the apraxia and non-apraxia groups. RESULTS: The internal consistency was high (Cronbach’s alpha=0.952). The inter-rater and intra-rater reliability and concurrent validity were also high (r=0.924–0.992, 0.961–0.999, and 0.830, respectively; p0.05). Among the 80 patients with stroke, 19 (23.8%) had limb apraxia and 21 (26.3%) had oral apraxia. Left hemispheric lesions and aphasia were significantly more frequently observed in the limb/oral apraxia group than in the non-apraxia group (p<0.001). CONCLUSION: The LOAT is a newly developed comprehensive test for limb and oral apraxia for Korean patients with stroke. It has high internal consistency, reliability, and validity and is a useful apraxia test for patients with stroke.

Cued and Un-Cued Semantic Category Fluency in Older Adults with Mild Cognitive Impairment

BACKGROUND AND PURPOSE: Patients with switching deficits reportedly benefit more from categorical cueing for semantic category fluency than do patients with clustering deficits. We explored the contribution of language ability and executive control on the performance of semantic category fluency in older adults with mild cognitive impairment (MCI) by examining the effects of categorical cueing on the task. METHODS: Order adults with MCI (n=10) and normal controls (n=25) were compared on two versions of a semantic fluency task: a standard, un-cued version (SF) and a version in which subjects were cued (C-SF) with 4 subordinated categories. The scores and error types of SF and C-SF tasks were analyzed between two groups. Also, the correlation among the SF task, the C-SF task, and the confrontation naming task were examined. RESULTS: The performance of the sematic fluency task improved when categorical cues were included in both groups. However, the normal group showed significantly more improvement than the MCI group. Self-repetition errors in the SF task and categorical errors in the C-SF task occurred most frequently. The normal group showed significantly more errors than the MCI group in the C-SF task. There was a positive correlation among the SF task, the C-SF task, and the confrontation naming task. CONCLUSIONS: The results of the present study suggested that the MCI group has more difficulty in the semantic memory store rather than in the use of retrieval strategies. A combination of standard and cued semantic fluency tasks may help to confirm the underlying deficit of semantic fluency impairment.

The Effects of Aging and Mild Cognitive Impairment on the Tip-of-the-Tongue Phenomenon in People Naming Task

BACKGROUND AND PURPOSE: The tip-of-the-tongue (TOT) phenomenon refers to knowing the meaning of the desired information but being unable to produce the phonological information. This study induced the TOT phenomenon through a task involving naming famous people, compared and analyzed results by age, and examined the differences with effects from pathological aging [mild cognitive impairment (MCI)]. METHODS: Young (n=10), middle-aged (n=10), older (n=10), oldest (n=10), and MCI (n=10) subjects participated in this study. To examine TOT rate, rate of TOT response, voluntary TOT solution rate, and TOT solution rate after presenting syllabic cue, each group participated in the experiment through a task of naming famous people. RESULTS: First, as subject age increased, TOT rate (the number of TOTs divided by total trials) increased. There was no significant difference with pathological aging. Second, normal aging and pathological aging had no significant effect on the rate of TOT response (the number of each response type divided by total response types). Third, the voluntary TOT solution rate (the number of voluntary TOT solutions divided by total TOTs) on occurrence of the TOT phenomenon was very low. Fourth, the TOT solution rate (the number of TOT solutions after presenting syllabic cue divided by total TOTs) had a negative correlation to normal aging. In examining pathological aging, the TOT solution rate of the MCI group was significantly lower than normal oldest people. CONCLUSION: Normal aging and pathological aging showed differences regarding TOT rate and TOT solution rate when naming famous individuals. The results of this study could be helpful as a differential tool between normal and pathological aging.

Lateralization of Cognitive Functions in Aphasia after Right Brain Damage

PURPOSE: The lateralization of cognitive functions in crossed aphasia in dextrals (CAD) has been explored and compared mainly with cases of aphasia with left hemisphere damage. However, comparing the neuropsychological aspects of CAD and aphasia after right brain damage in left-handers (ARL) could potentially provide more insights into the effect of a shift in the laterality of handedness or language on other cognitive organization. Thus, this case study compared two cases of CAD and one case of ARL. MATERIALS AND METHODS: The following neuropsychological measures were obtained from three aphasic patients with right brain damage (two cases of CAD and one case of ARL); language, oral and limb praxis, and nonverbal cognitive functions (visuospatial neglect and visuospatial construction). RESULTS: All three patients showed impaired visuoconstructional abilities, whereas each patient showed a different level of performances for oral and limb praxis, and visuospatial neglect. CONCLUSION: Based on the analysis of these three aphasic patients' performances, we highlighted the lateralization of language, handedness, oral and limb praxis, visuospatial neglect and visuospatial constructive ability in aphasic patients with right brain damage.

Standardization of Korean Version of Frenchay Aphasia Screening Test in Normal Adults

OBJECTIVE: To develop Korean version of Frenchay Aphasia Screening Test (K-FAST) and to standardize K-FAST in normal adult population in Korea. METHOD: The 'river scene' of stimulus set in original version of FAST was adapted for K-FAST. English version of the test instructions and scoring methods were translated into Korean and reverse-translation was performed by English- Korean bilinguals. The test structures and language domains (comprehension, expression, reading, and writing) were unchanged and possible maximum score was 30 points. We standardized K-FAST in 240 normal adult populations (male 102, female 138) whose ages were above 44. Basic personal information was collected through an interview and we performed Edinburgh handedness inventory (EHI), K- MMSE and K-FAST. Subjects who had history of brain disease, cognitive communicative disorders, or K-MMSE scores less than 2 percentile of same age group were excluded. K-FAST scores were analyzed according to the age and education groups. RESULTS: Mean EHI, K-MMSE, and K-FAST scores in total subjects were 9.4+/-1.2, 25.9+/-2.8, 25.4+/-3.3 points, respectively. Post-hoc analysis of K-FAST scores according to age groups classified into 3 age groups, 45~64, 65~74, and > or =74 years and education groups into 0, 1~9, > or =10 years of total education. K-FAST scores decreased significantly as increase of age (r=-0.441, p=0.000) and decrease of total years of education (r=0.580, p=0.000). CONCLUSION: Newly adapted K-FAST can be used for screening of aphasia in Korea and the standardized data according to age and education levels may provide useful reference values for interpretation of the results of K-FAST.