Artificial Neural Network in Prediction of Language Delay in Children Under 1 Year: Preliminary Results

Author:

Vasilchenko K. F.1ORCID,Leonova A. V.2ORCID,Usov G. M.1ORCID,Raeva T. V.2ORCID

Affiliation:

1. Omsk State Medical University

2. Tyumen State Medical University

Abstract

Background: the delay in language development is characterized by qualitative and quantitative underdevelopment of the vocabulary and the lack of formation of expressive speech. This violation belongs to the mildest speech pathologies, however, there is a high probability of the presence of concomitant mental pathology and the occurrence of adaptation problems at school age. In the etiology of delayed language development, its multifactorial nature has been established. Thus, there is a need to develop a tool that predicts the formation of a delay in speech development in children for the timely implementation of preventive measures.Aim of the study: to develop a tool for predicting speech development delay in children under one year old using artificial intelligence algorithms.Patients and methods: 196 children were examined. The mean age was 26.9 months (SD ± 5.5 months). The sample was divided into two groups: the first included patients with delayed speech development (n = 98), the second included children with normal speech development (n = 98). Speech status was assessed using a questionnaire to determine the speech development of a child aged 18 to 36 months (Language Development Survey). In assessing the risk factors for the occurrence of speech development delay, the “Anamnestic Card of the child” was used. To create a neural network that predicts speech delay in children under one year old, a model was developed and trained using the Keras library for the Python 3.0 programming language.Results: the analysis of the accuracy of the neural network showed a high result — 89% of the cases during the training of the model were identified correctly. At the same time, the sensitivity of the model on the test sample was 100%, and the specificity was 90%.Conclusions: the developed method can be used to create a tool for predicting speech development delay in children up to 3 years of age, which will allow for differentiated therapeutic and preventive measures that contribute to the harmonious development of the child.

Publisher

Medical Informational Agency Publishers

Subject

Biological Psychiatry,Psychiatry and Mental health,Clinical Psychology

Reference37 articles.

1. Baranov AA, Maslova OI, Namazova-Baranova LS. Ontogenesis of neurocognitive development of children and adolescents. Vestnik Rossijskoj akademii medicinskih nauk. 2012;(8):26–33. (In Russ). doi: 10.15690/vramn.v67i8.346

2. Asmolova GA, Zavadenko AN, Zavadenko EV, Kozlova NN, Medvedev MI, Rogatkin SO. Early diagnostics of speech development disorders. Features of speech development of children with consequences of perinatal pathology of the nervous system. Clinical recommendations. Moscow: Russian association of perinatal medicine specialists. 2014:57. (In Russ.).

3. Kalnak N, Peyrard-Janvid M, Sahlén B, Forssberg H. Family history interview of a broad phenotype in specific language impairment and matched controls. Genes Brain Behav. 2012;11(8):921–927. doi: 10.1111/j.1601-183X.2012.00841.x

4. Hawa VV, Spanoudis G. Toddlers with delayed expressive language: an overview of the characteristics, risk factors and language outcomes. Res Dev Disabil. 2014;35(2):400–407. doi: 10.1016/j.ridd.2013.10.027

5. Raeva TV, Leonova AV, Probotyuk VV. Features of mental impairments, comorbid developmental language delay in early age children. Voprosy psikhicheskogo zdorov’ya detei i podrostkov. 2019;(3):64–69. (In Russ.).

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3