• Journal of the Korean Academy of Child and Adolescent Psychiatry
• /
• v.10 no.1
• /
• pp.43-49
• /
• 1999

Objectives：One of the most common developmental problems is communication disorder in which a child appears normal in every way but who has failed to begin speaking or speaks very little. A few studies have examined the temperamental characteristics of children with communication disorders. This study was to investigate the temperamental characteristics of Korean children with communication disorders. Methods：The parents of 20 Korean children with communication disorders and the parents of 50 normal control children, the age of both groups ranges from 3 to 7, completed Korean version of Parental Temperamental Questionnaire developed by Thomas and Chess. Children with a pervasive developmental disorder, mental retardation, or speech-motor or sensory deficit were excluded. The scores of each temperamental scale of two groups and the diagnostic clusters of two groups were compared. Results：The children with communication disorders were characterized by lower mood scores and higher intensity of reaction scores than normal controls. The two groups showed no significant correlation in terms of the temperamental diagnostic clusters. Conclusion：This findings suggest the existence of a distinct temperamental profile of the children with communication disorders. Early detection of the profile may be of great value for parents in understanding the developmental characteristics of the children with communication disorders and in providing appropriate parenting approaches.

• KIPS Transactions on Software and Data Engineering
• /
• v.9 no.7
• /
• pp.221-228
• /
• 2020

Part-of-Speech (POS) tagged corpus is a collection of electronic text in which each word is annotated with a tag as the corresponding POS and is widely used for various training data for natural language processing. The training data generally assumes that there are no errors, but in reality they include various types of errors, which cause performance degradation of systems trained using the data. To alleviate this problem, we propose a novel method for detecting errors in the existing POS tagged corpus using the classifier of XGBoost and cross-validation as evaluation techniques. We first train a classifier of a POS tagger using the POS-tagged corpus with some errors and then detect errors from the POS-tagged corpus using cross-validation, but the classifier cannot detect errors because there is no training data for detecting POS tagged errors. We thus detect errors by comparing the outputs (probabilities of POS) of the classifier, adjusting hyperparameters. The hyperparameters is estimated by a small scale error-tagged corpus, in which text is sampled from a POS-tagged corpus and which is marked up POS errors by experts. In this paper, we use recall and precision as evaluation metrics which are widely used in information retrieval. We have shown that the proposed method is valid by comparing two distributions of the sample (the error-tagged corpus) and the population (the POS-tagged corpus) because all detected errors cannot be checked. In the near future, we will apply the proposed method to a dependency tree-tagged corpus and a semantic role tagged corpus.

• Journal of The Korean Society of Inherited Metabolic disease
• /
• v.23 no.2
• /
• pp.39-44
• /
• 2023

Hereditary tyrosinemia type 1 (HT-1) is a metabolic disorder caused by biallelic pathogenic variants in the fumarylacetoacetate hydrolase (FAH) gene, which impairs the function of the FAH enzyme, resulting in the accumulation of tyrosine's toxic metabolites in hepatocytes and renal tubular cells. As a consequence, individuals with HT-1 exhibit symptomatic manifestations. Rapid diagnosis and treatment of HT-1 can prevent short-term death and long-term complications. A 15-day-old boy presented to the outpatient department with elevated levels of tyrosine on his newborn screening tests conducted at the age of 3 and 10 days, respectively. Further blood tests revealed increased levels of alpha-fetoprotein and amino acids including tyrosine and threonine. Urine organic acid tests indicated a significant elevation in tyrosine metabolites, as well as the presence of succinylacetone (SA), which led to the diagnosis of HT-1. Two pathogenic and likely pathogenic variants of FAH compatible with HT-1 were also detected. He began a tyrosine-restricted diet at one month old and received nitisinone (NTBC) at two months old. With continued treatment, the patient's initially elevated AFP level, detection of SA in the urine, and mild hepatomegaly showed improvement. During four years and seven months of treatment, there were no exceptional complications apart from an increase in tyrosine levels and a delay in speech. We report a case of tyrosinemia type 1 detected through newborn screening, treated with dietary restriction and NTBC, with a good prognosis.

• Clinical and Experimental Pediatrics
• /
• v.49 no.5
• /
• pp.507-512
• /
• 2006

Purpose : Early detection and intervention of hearing impairment is believed to improve speech and language development and behavior of children. The aim of this preliminary study was to determine the prevalence of hearing impairments, and to identify the association of risk factors relating to refer response in high risk neonates who were screened using distortion product otoacoustic emissions (DPOAE). Methods : The subjects included 871 neonates who were admitted to the neonatal intensive care unit of the Pediatric Department in Soonchunhyang University Bucheon Hospital from May, 2001 to December, 2004. They were screened using DPOAE. Based on DPOAE, we divided the neonates in two groups : 'Pass' and 'Refer'. The differences in risk factors between the pass group and the refer group were analyzed. Results : The incidence of the refer group was 12.1 percent(106 out of 871). The bilateral refer rate was 5.4 percent(47 out of 871). And the unilateral refer rate was 6.7 percent(59 out of 871). Gender, birth place, family history of hearing loss, small/large for gestational age, obstetrical factor, hyperbilirubinemia and use of gentamicin were not statistically related to the refer rate. Statistically related to refer rate were birth weight, resuscitated neonates, Apgar score, craniofacial anomaly, mechanical ventilator application, sepsis, using of vancomycin(P<0.05). The prevalence of hearing impairment (${\geq}60dB$) in this study was 2 percent(18 out of 871). Conclusion : This study showed a higher prevalence of hearing impairment in high-risk neonates. Thus neonatal hearing screening should be carried out in high-risk neonates.

• Journal of Service Research and Studies
• /
• v.14 no.1
• /
• pp.13-26
• /
• 2024

Multi-modal generation is the process of generating results based on a variety of information, such as text, images, and audio. With the rapid development of AI technology, there is a growing number of multi-modal based systems that synthesize different types of data to produce results. In this paper, we present an AI system that uses speech and text recognition to describe a person and generate a montage image. While the existing montage generation technology is based on the appearance of Westerners, the montage generation system developed in this paper learns a model based on Korean facial features. Therefore, it is possible to create more accurate and effective Korean montage images based on multi-modal voice and text specific to Korean. Since the developed montage generation app can be utilized as a draft montage, it can dramatically reduce the manual labor of existing montage production personnel. For this purpose, we utilized persona-based virtual person montage data provided by the AI-Hub of the National Information Society Agency. AI-Hub is an AI integration platform aimed at providing a one-stop service by building artificial intelligence learning data necessary for the development of AI technology and services. The image generation system was implemented using VQGAN, a deep learning model used to generate high-resolution images, and the KoDALLE model, a Korean-based image generation model. It can be confirmed that the learned AI model creates a montage image of a face that is very similar to what was described using voice and text. To verify the practicality of the developed montage generation app, 10 testers used it and more than 70% responded that they were satisfied. The montage generator can be used in various fields, such as criminal detection, to describe and image facial features.