Dataset Information

Developing a Diagnostic Decision Support System for Benign Paroxysmal Positional Vertigo Using a Deep-Learning Model.

ABSTRACT: Background: Diagnosis of benign paroxysmal positional vertigo (BPPV) depends on the accurate interpretation of nystagmus induced by positional tests. However, difficulties in interpreting eye-movement often can arise in primary care practice or emergency room. We hypothesized that the use of machine learning would be helpful for the interpretation. Methods: From our clinical data warehouse, 91,778 nystagmus videos from 3467 patients with dizziness were obtained, in which the three-dimensional movement of nystagmus was annotated by four otologic experts. From each labeled video, 30 features changed into 255 grid images fed into the input layer of the neural network for the training dataset. For the model validation, video dataset of 3566 horizontal, 2068 vertical, and 720 torsional movements from 1005 patients with BPPV were collected. Results: The model had a sensitivity and specificity of 0.910 ± 0.036 and 0.919 ± 0.032 for horizontal nystagmus; of 0.879 ± 0.029 and 0.894 ± 0.025 for vertical nystagmus; and of 0.783 ± 0.040 and 0.799 ± 0.038 for torsional nystagmus, respectively. The affected canal was predicted with a sensitivity of 0.806 ± 0.010 and a specificity of 0.971 ± 0.003. Conclusions: As our deep-learning model had high sensitivity and specificity for the classification of nystagmus and localization of affected canal in patients with BPPV, it may have wide clinical applicability.

SUBMITTER: Lim EC

PROVIDER: S-EPMC6571642 | biostudies-literature | 2019 May

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Developing a Diagnostic Decision Support System for Benign Paroxysmal Positional Vertigo Using a Deep-Learning Model.

Lim Eun-Cheon EC Park Jeong Hye JH Jeon Han Jae HJ Kim Hyung-Jong HJ Lee Hyo-Jeong HJ Song Chang-Geun CG Hong Sung Kwang SK

Journal of clinical medicine 20190508 5

<i>Background</i>: Diagnosis of benign paroxysmal positional vertigo (BPPV) depends on the accurate interpretation of nystagmus induced by positional tests. However, difficulties in interpreting eye-movement often can arise in primary care practice or emergency room. We hypothesized that the use of machine learning would be helpful for the interpretation. <i>Methods</i>: From our clinical data warehouse, 91,778 nystagmus videos from 3467 patients with dizziness were obtained, in which the three- ...[more]

PMID: 31072056

Similar Datasets

Project description:BackgroundBenign paroxysmal positional vertigo is the most common disease in which vertigo is the main clinical manifestation, and it has become a global medical problem, affecting a wide range of areas and seriously affecting the quality of human life.ObjectiveThis article presents an analysis of the current characteristics of BPPV-related research and summarizes the current hot topics and trends, with the goal of inspiring future research into the prevention and treatment of BPPV, thereby improving the differential diagnosis and prevention of peripheral vertigo.MethodsA bibliometric approach was used to collect 1,219 eligible studies on BPPV from four databases-PubMed, Embase, Scopus, and Web of Science-published between 1974 and 2022. The characteristics and status of the accumulated scientific output were processed using R and VOSviewer so that we could visualize any trends or hotspots.ResultsThe results showed a significant increase in the annual number of publications, with an average annual growth rate of 21.58%. A possible reason for the especially pronounced peak in 2021 was an increase in the prevalence of BPPV as a result of COVID-19. The new coronavirus became a focus of research in 2021. A total of 3,876 authors (of whom 1,097 were first authors) published articles in 307 different journals; 15.7% of the articles were published in Acta Oto-Larygologica, Otology and Neurotology, and Frontiers in Neurology. Acta Oto-Laryngologica was well ahead of the other journals in terms of growth rate and number of articles published. American scholars generated the largest number of articles overall, and the USA was involved in the greatest number of international collaborations, followed by Italy and China. The themes of the research centered around three topics, namely the treatment of BPPV, its influencing factors, and diagnosis.ConclusionsThere has been a major increase in BPPV-related research over the last 50 years, leading to an increase in related articles and rapid development of the field. Key directions for future research include the improvement of individualized treatment for residual symptoms after initial treatment of BPPV among the elderly; effective control of comorbidities such as osteoporosis; and secondary inner ear disease, such as Ménière's disease.

Dataset Information

Developing a Diagnostic Decision Support System for Benign Paroxysmal Positional Vertigo Using a Deep-Learning Model.

Publications

Developing a Diagnostic Decision Support System for Benign Paroxysmal Positional Vertigo Using a Deep-Learning Model.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets