为人工智能研究中的儿科语音数据库开发针对年龄的协议

IF 1.3 4区医学 Q3 OTORHINOLARYNGOLOGY

International journal of pediatric otorhinolaryngology Pub Date : 2025-07-05 DOI:10.1016/j.ijporl.2025.112455

Laurie Russell , Yael Bensoussan , Evan Ng , Allistair Johnson , Siyu Miao , Nikolaus E. Wolter , Evan J. Propst , Jennifer M. Siu

{"title":"为人工智能研究中的儿科语音数据库开发针对年龄的协议","authors":"Laurie Russell , Yael Bensoussan , Evan Ng , Allistair Johnson , Siyu Miao , Nikolaus E. Wolter , Evan J. Propst , Jennifer M. Siu","doi":"10.1016/j.ijporl.2025.112455","DOIUrl":null,"url":null,"abstract":"<div><h3>Introduction</h3><div>Children's voice and communication abilities evolve with age, necessitating tailored protocols for accurate analysis. Distinct vocal properties and communication styles across developmental stages require specific tasks. The absence of standardized pediatric protocols impedes reliable voice data collection, limits robust acoustic measurements, and hinders AI-driven voice research. This study aims to develop age-specific voice data collection protocols, creating a pediatric voice database to advance research and enhance AI applications.</div></div><div><h3>Methods</h3><div>Protocols were developed iteratively over six months through a multidisciplinary, multi-institutional team involving speech-language pathologists and otolaryngologists. A comprehensive literature review was completed. Emphasis was placed on designing protocols capable of generating high-quality, consistent data for AI modeling while minimizing measurement variability.</div></div><div><h3>Results</h3><div>Four developmentally appropriate protocols were created for children aged 2 to <4 years, 4 to <6 years, 6 to <10 years, and 10+ years respectively. Each protocol targeted key acoustic parameters while aligning with the linguistic and cognitive abilities of the respective age group. The iterative design incorporated tasks that engaged children while ensuring clinically robust data collection. Standardized protocols were deployed via a user-friendly iPad application, using child-sized headset microphones at fixed distances from the mouth. The protocols were administered to 100 pediatric patients in an otolaryngology clinic between the ages of 2–18.</div></div><div><h3>Conclusions</h3><div>The development of age specific protocols aims to standardize the collection of pediatric voice and acoustic sound data, which will serve as a valuable resource for advancing artificial intelligence applications within clinical sciences. Future validation will ensure feasibility, adherence, and reliability of data collection to support AI-driven innovations.</div></div>","PeriodicalId":14388,"journal":{"name":"International journal of pediatric otorhinolaryngology","volume":"196 ","pages":"Article 112455"},"PeriodicalIF":1.3000,"publicationDate":"2025-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Developing age-specific protocols for pediatric voice databases in artificial intelligence research\",\"authors\":\"Laurie Russell , Yael Bensoussan , Evan Ng , Allistair Johnson , Siyu Miao , Nikolaus E. Wolter , Evan J. Propst , Jennifer M. Siu\",\"doi\":\"10.1016/j.ijporl.2025.112455\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><h3>Introduction</h3><div>Children's voice and communication abilities evolve with age, necessitating tailored protocols for accurate analysis. Distinct vocal properties and communication styles across developmental stages require specific tasks. The absence of standardized pediatric protocols impedes reliable voice data collection, limits robust acoustic measurements, and hinders AI-driven voice research. This study aims to develop age-specific voice data collection protocols, creating a pediatric voice database to advance research and enhance AI applications.</div></div><div><h3>Methods</h3><div>Protocols were developed iteratively over six months through a multidisciplinary, multi-institutional team involving speech-language pathologists and otolaryngologists. A comprehensive literature review was completed. Emphasis was placed on designing protocols capable of generating high-quality, consistent data for AI modeling while minimizing measurement variability.</div></div><div><h3>Results</h3><div>Four developmentally appropriate protocols were created for children aged 2 to <4 years, 4 to <6 years, 6 to <10 years, and 10+ years respectively. Each protocol targeted key acoustic parameters while aligning with the linguistic and cognitive abilities of the respective age group. The iterative design incorporated tasks that engaged children while ensuring clinically robust data collection. Standardized protocols were deployed via a user-friendly iPad application, using child-sized headset microphones at fixed distances from the mouth. The protocols were administered to 100 pediatric patients in an otolaryngology clinic between the ages of 2–18.</div></div><div><h3>Conclusions</h3><div>The development of age specific protocols aims to standardize the collection of pediatric voice and acoustic sound data, which will serve as a valuable resource for advancing artificial intelligence applications within clinical sciences. Future validation will ensure feasibility, adherence, and reliability of data collection to support AI-driven innovations.</div></div>\",\"PeriodicalId\":14388,\"journal\":{\"name\":\"International journal of pediatric otorhinolaryngology\",\"volume\":\"196 \",\"pages\":\"Article 112455\"},\"PeriodicalIF\":1.3000,\"publicationDate\":\"2025-07-05\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"International journal of pediatric otorhinolaryngology\",\"FirstCategoryId\":\"3\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S0165587625002423\",\"RegionNum\":4,\"RegionCategory\":\"医学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q3\",\"JCRName\":\"OTORHINOLARYNGOLOGY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"International journal of pediatric otorhinolaryngology","FirstCategoryId":"3","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0165587625002423","RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"OTORHINOLARYNGOLOGY","Score":null,"Total":0}

引用次数: 0

摘要

儿童的声音和沟通能力随着年龄的增长而发展，需要量身定制的方案来进行准确的分析。在不同的发育阶段，不同的声音特性和沟通方式需要特定的任务。缺乏标准化的儿科协议阻碍了可靠的语音数据收集，限制了强大的声学测量，并阻碍了人工智能驱动的语音研究。该研究旨在制定针对特定年龄的语音数据收集协议，创建儿科语音数据库，以推进研究和增强人工智能应用。方法由语言病理学家和耳鼻喉科医生组成的多学科、多机构团队在六个月的时间里反复制定协议。完成了全面的文献综述。重点是设计能够为人工智能建模生成高质量、一致数据的协议，同时最大限度地减少测量变异性。结果针对2 ~ 4岁、4 ~ 6岁、6 ~ 10岁、10+岁儿童分别制定了4个发育适宜方案。每个方案都针对关键的声学参数，同时与相应年龄组的语言和认知能力保持一致。迭代设计纳入了儿童参与的任务，同时确保临床可靠的数据收集。标准化协议通过一个用户友好的iPad应用程序部署，使用儿童大小的耳机麦克风，与嘴保持固定距离。该方案在一家耳鼻喉科诊所对100名年龄在2-18岁之间的儿科患者实施。结论制定针对儿童年龄的协议旨在规范儿童语音和声学数据的收集，这将为推进人工智能在临床科学中的应用提供宝贵的资源。未来的验证将确保数据收集的可行性、依从性和可靠性，以支持人工智能驱动的创新。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

Developing age-specific protocols for pediatric voice databases in artificial intelligence research

Introduction

Children's voice and communication abilities evolve with age, necessitating tailored protocols for accurate analysis. Distinct vocal properties and communication styles across developmental stages require specific tasks. The absence of standardized pediatric protocols impedes reliable voice data collection, limits robust acoustic measurements, and hinders AI-driven voice research. This study aims to develop age-specific voice data collection protocols, creating a pediatric voice database to advance research and enhance AI applications.

Methods

Protocols were developed iteratively over six months through a multidisciplinary, multi-institutional team involving speech-language pathologists and otolaryngologists. A comprehensive literature review was completed. Emphasis was placed on designing protocols capable of generating high-quality, consistent data for AI modeling while minimizing measurement variability.

Results

Four developmentally appropriate protocols were created for children aged 2 to <4 years, 4 to <6 years, 6 to <10 years, and 10+ years respectively. Each protocol targeted key acoustic parameters while aligning with the linguistic and cognitive abilities of the respective age group. The iterative design incorporated tasks that engaged children while ensuring clinically robust data collection. Standardized protocols were deployed via a user-friendly iPad application, using child-sized headset microphones at fixed distances from the mouth. The protocols were administered to 100 pediatric patients in an otolaryngology clinic between the ages of 2–18.

Conclusions

The development of age specific protocols aims to standardize the collection of pediatric voice and acoustic sound data, which will serve as a valuable resource for advancing artificial intelligence applications within clinical sciences. Future validation will ensure feasibility, adherence, and reliability of data collection to support AI-driven innovations.

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

International journal of pediatric otorhinolaryngology 医学-耳鼻喉科学

CiteScore

3.20

自引率

6.70%

发文量

276

审稿时长

62 days

期刊介绍： The purpose of the International Journal of Pediatric Otorhinolaryngology is to concentrate and disseminate information concerning prevention, cure and care of otorhinolaryngological disorders in infants and children due to developmental, degenerative, infectious, neoplastic, traumatic, social, psychiatric and economic causes. The Journal provides a medium for clinical and basic contributions in all of the areas of pediatric otorhinolaryngology. This includes medical and surgical otology, bronchoesophagology, laryngology, rhinology, diseases of the head and neck, and disorders of communication, including voice, speech and language disorders.