Applications of Machine Learning in Meniere's Disease Assessment Based on Pure‐Tone Audiometry DOI Creative Commons
Xu Liu, Ping Guo, Dan Wang

и другие.

Otolaryngology, Год журнала: 2024, Номер 172(1), С. 233 - 242

Опубликована: Авг. 28, 2024

Abstract Objective To apply machine learning models based on air conduction thresholds of pure‐tone audiometry for automatic diagnosis Meniere's disease (MD) and prediction endolymphatic hydrops (EH). Study Design Retrospective study. Setting Tertiary medical center. Methods Gadolinium‐enhanced magnetic resonance imaging sequences data were collected. Subsequently, basic multiple analytical features engineered the audiometry. Later, 5 classical trained to diagnose MD using features. The demonstrating excellent performance also selected predict EH. model's effectiveness in was compared with experienced otolaryngologists. Results First, winning light gradient boosting (LGB) model by demonstrates a remarkable MD, achieving an accuracy rate 87%, sensitivity 83%, specificity 90%, robust area under receiver operating characteristic curve 0.95, which compares favorably clinicians. Second, LGB model, 78% EH prediction, outperformed other 3 models. Finally, feature importance analysis reveals pivotal role specific that are essential both prediction. Highlighted include standard deviation mean whole‐frequency hearing, peak audiogram, hearing at low frequencies, notably 250 Hz. Conclusion An efficient produced showed potential subtypes innovative approach demonstrated game‐changing strategy screening promising cost‐effective benefits health care enterprise.

Язык: Английский

Artificial Intelligence in Audiology: A Scoping Review of Current Applications and Future Directions DOI Creative Commons
Andrea Frosolini, Leonardo Franz, Valeria Caragli

и другие.

Sensors, Год журнала: 2024, Номер 24(22), С. 7126 - 7126

Опубликована: Ноя. 6, 2024

The integration of artificial intelligence (AI) into medical disciplines is rapidly transforming healthcare delivery, with audiology being no exception. By synthesizing the existing literature, this review seeks to inform clinicians, researchers, and policymakers about potential challenges integrating AI audiological practice. PubMed, Cochrane, Google Scholar databases were searched for articles published in English from 1990 2024 following query: "(audiology) AND ("artificial intelligence" OR "machine learning" "deep learning")". PRISMA extension scoping reviews (PRISMA-ScR) was followed. database research yielded 1359 results, selection process led inclusion 104 manuscripts. has evolved significantly over succeeding decades, 87.5% manuscripts last 4 years. Most types consistently used specific purposes, such as logistic regression other statistical machine learning tools (e.g., support vector machine, multilayer perceptron, random forest, deep belief network, decision tree, k-nearest neighbor, or LASSO) automated audiometry clinical predictions; convolutional neural networks radiological image analysis; large language models automatic generation diagnostic reports. Despite advances technologies, different ethical professional are still present, underscoring need larger, more diverse data collection bioethics studies field audiology.

Язык: Английский

Процитировано

8
Integration of smartphone technology and artificial intelligence for advanced ophthalmic care: A systematic review DOI Creative Commons
Kai Jin,

Yingyu Li,

Hongkang Wu

и другие.

Advances in Ophthalmology Practice and Research, Год журнала: 2024, Номер 4(3), С. 120 - 127

Опубликована: Март 25, 2024

The convergence of smartphone technology and artificial intelligence (AI) has revolutionized the landscape ophthalmic care, offering unprecedented opportunities for diagnosis, monitoring, management ocular conditions. Nevertheless, there is a lack systematic studies on discussing integration AI in this field. This review includes 52 studies, explores smartphones ophthalmology, delineating its collective impact screening methodologies, disease detection, telemedicine initiatives, patient management. findings from curated indicate promising performance smartphone-based various diseases which encompass major retinal diseases, glaucoma, cataract, visual impairment children surface diseases. Moreover, utilization imaging modalities, coupled with algorithms, able to provide timely, efficient cost-effective pathologies. modality can also facilitate self-monitoring, remote monitoring enhancing accessibility eye care services, particularly underserved regions. Challenges involving data privacy, algorithm validation, regulatory frameworks issues trust are still need be addressed. Furthermore, evaluation real-world implementation imperative as well, prospective currently lacking. Smartphone merged enables earlier, precise diagnoses, personalized treatments, enhanced service care. Collaboration crucial navigate ethical security challenges while responsibly leveraging these innovations, potential revolution access global health equity.

Язык: Английский

Процитировано

7
Artificial intelligence approaches for tinnitus diagnosis: leveraging high-frequency audiometry data for enhanced clinical predictions DOI Creative Commons
Seyed‐Ali Sadegh‐Zadeh,

Alireza Soleimani Mamalo,

Kaveh Kavianpour

и другие.

Frontiers in Artificial Intelligence, Год журнала: 2024, Номер 7

Опубликована: Май 7, 2024

This research investigates the application of machine learning to improve diagnosis tinnitus using high-frequency audiometry data. A Logistic Regression (LR) model was developed alongside an Artificial Neural Network (ANN) and various baseline classifiers identify most effective approach for classifying presence. The methodology encompassed data preprocessing, feature extraction focused on point detection, rigorous evaluation through performance metrics including accuracy, Area Under ROC Curve (AUC), precision, recall, F1 scores. main findings reveal that LR model, supported by ANN, significantly outperformed other models, achieving accuracy 94.06%, AUC 97.06%, high precision recall These results demonstrate efficacy ANN in accurately diagnosing tinnitus, surpassing traditional diagnostic methods rely subjective assessments. implications this are substantial clinical audiology, suggesting learning, particularly advanced models like ANNs, can provide a more objective quantifiable tool diagnosis, especially when utilizing not typically assessed standard hearing tests. study underscores potential facilitate earlier accurate which could lead improved patient outcomes. Future work should aim expand dataset diversity, explore broader range algorithms, conduct trials validate models' practical utility. highlights transformative paving way advancements treatment tinnitus.

Язык: Английский

Процитировано

7
Web- and app-based tools for remote hearing assessment: a scoping review DOI Creative Commons
Ibrahim Almufarrij, Harvey Dillon, Piers Dawes

и другие.

International Journal of Audiology, Год журнала: 2022, Номер 62(8), С. 699 - 712

Опубликована: Июнь 9, 2022

Objective Remote hearing screening and assessment may improve access to, uptake of, care. This review, the most comprehensive to date, aimed (i) identify assess functionality of remote tools on smartphones online platforms, (ii) determine if assessed were also evaluated in peer-reviewed publications (iii) report accuracy existing validation data.Design Protocol was registered INPLASY reported according PRISMA-Extension for Scoping Reviews.Study sample In total, 187 (using tones, speech, self-report or a combination) 101 studies met inclusion criteria. Quality, functionality, bias applicability each app by at least two authors.Results Assessed showed considerable variability functionality. Twenty-two (12%) 14 had acceptable The results their quality varied greatly, largely depending category tool.Conclusion reliability are unknown. Tone-producing provide approximate thresholds but have calibration background noise issues. Speech less affected these issues mostly do not an estimated pure tone audiogram. Predicting audiograms using filtered language-independent materials could be universal solution.

Язык: Английский

Процитировано

25
Advancing Equitable Hearing Care: Innovations in Technology and Service Delivery DOI
De Wet Swanepoel

Folia Phoniatrica et Logopaedica, Год журнала: 2023, Номер 75(4), С. 201 - 207

Опубликована: Янв. 1, 2023

Hearing loss is a neglected global health priority affecting 1.5 billion persons. Global access to hearing care severely limited with management options, like aids, inaccessible most. The cost and centralised nature of traditional service-delivery approaches in have undermined equitable alongside poor awareness.Recent innovations digital mHealth technologies used by workers through task shifting are enabling novel community-based services across the continuum care. This narrative review explores technology-enabled communities. We provide examples focused on our work over past decade explore more primary, secondary, tertiary levels prevention.Hearing potential increase care, improve quality life for those affected loss, reduce costs associated untreated loss. More that requires scalable models enabled innovative within communities integrated into public initiatives including promotion.

Язык: Английский

Процитировано

14
The Hearing Test App for Android Devices: Distinctive Features of Pure-Tone Audiometry Performed on Mobile Devices DOI Creative Commons
Marcin Masalski

Medical Devices Evidence and Research, Год журнала: 2024, Номер Volume 17, С. 151 - 163

Опубликована: Апрель 1, 2024

Abstract: The popularity of mobile devices, combined with advances in electronic design and internet technology, has enabled home-based hearing tests recent years. purpose this article is to highlight the distinctive aspects pure-tone audiometry performed on a device by means Hearing Test app for Android devices. first version was released decade ago, since then been systematically improved, which required addressing many issues common majority apps testing. discusses techniques calibration, outlines testing procedure how it differs from traditional audiometry, explores potential bone conduction testing, provides considerations interpreting including test duration background noise. concludes detailing clinically relevant requiring special attention during interpretation results are substantial value hundreds thousands active users worldwide, as well other apps. Keywords: self-testing, threshold,

Язык: Английский

Процитировано

6
Performance and reliability evaluation of an improved machine learning‐based pure‐tone audiometry with automated masking DOI Creative Commons

Nicolas Wallaert,

Antoine Perry,

Sandra Quarino

и другие.

World Journal of Otorhinolaryngology - Head and Neck Surgery, Год журнала: 2024, Номер unknown

Опубликована: Сен. 12, 2024

Abstract Objective Automated air‐conduction pure‐tone audiograms through Bayesian estimation and machine learning (ML) classification have recently been proposed in the literature. Although such ML‐based audiometry approaches represent a significant addition to field, they remain unsuited for daily clinical settings, particular listeners with asymmetric or conductive hearing loss, severe cochlear dead zones. The goal here is expand on previously ML assess performance of this improved large sample wide range status. Methods First, we describe changes made method of: (1) safety limits test status, (2) transient responses cater zones nonmeasurable thresholds, importantly, (3) automated contralateral masking loss. Next, compared conventional manual cohort ( n = 109 subjects) both normal‐hearing hearing‐impaired listeners. Results Our results showed that all audiometric frequencies tested, no difference was found between thresholds obtained using audiometer as methods. Furthermore, test–retest not each frequency tested. Finally, when examining cross‐clinic reliability measures, differences were most Conclusions Together, our validate use adult tests audiometry.

Язык: Английский

Процитировано

3
Performance and Reliability Evaluation of an Automated Bone-Conduction Audiometry Using Machine Learning DOI Creative Commons

Nicolas Wallaert,

Antoine Perry,

Hadrien Jean

и другие.

Trends in Hearing, Год журнала: 2024, Номер 28

Опубликована: Сен. 1, 2024

To date, pure-tone audiometry remains the gold standard for clinical auditory testing. However, is time-consuming and only provides a discrete estimate of hearing acuity. Here, we aim to address these two main drawbacks by developing machine learning (ML)-based approach fully automated bone-conduction (BC) tests with forehead vibrator placement. Study 1 examines occlusion effects when headphones are positioned on both ears during BC 2 describes ML-based audiometry, contralateral masking rules, compensation forehead-mastoid corrections. Next, performance ML-audiometry examined in comparison manual conventional mastoid Finally, 3 test-retest reliability ML-audiometry. Our results show no significant difference between audiometry. High achieved Together, our findings demonstrate normal-hearing hearing-impaired adult listeners mild severe losses.

Язык: Английский

Процитировано

3
Audiological profile of first-time hearing aid users – implications for the development of a fast-track fitting protocol DOI Creative Commons
Jukka Kokkonen, Fanni Svärd, Sini Varonen

и другие.

International Journal of Audiology, Год журнала: 2025, Номер unknown, С. 1 - 10

Опубликована: Март 3, 2025

Objective To assess the feasibility of implementing a fast-track process (single-session assess-and-fit appointment with no ENT specialist's examination) in hearing rehabilitation by investigating accuracy protocol assignment applying various cut-off criteria and describing audiometric profile patients being evaluated for their first aids.

Язык: Английский

Процитировано

0
Hearing loss configurations in low- and middle-income countries DOI Creative Commons
John Newall, Rebecca Kim, Piers Dawes

и другие.

International Journal of Audiology, Год журнала: 2025, Номер unknown, С. 1 - 8

Опубликована: Март 6, 2025

The majority of individuals with hearing loss worldwide reside in low- and middle-income countries (LMICs), but there is limited information regarding the characteristics these regions. This descriptive study aims to address this knowledge gap by analysing audiogram patterns LMIC populations. Greater about properties LMICs allows for improved planning interventions. Retrospective data from 23 collaborating centres across 16 were collected. All participants adults seeking help problems. A machine learning approach was utilised classify threshold identify representative profiles. comprised 5773 participants. results revealed mildly sloping audiometric varying severity. differed previous studies conducted high-income regions which included more steeply losses. findings also indicated a higher proportion severe levels loss. These variations could be attributed population-level differences causative mechanisms LMICs, such as prevalence infectious disease-related may reflect health behaviours. highlights need tailored, scalable, interventions LMICs.

Язык: Английский

Процитировано

0