Artificial intelligence for early stroke diagnosis in acute vestibular syndrome
ObjectiveMeasuring the Vestibular-Ocular-Reflex (VOR) gains with the video head impulse test (vHIT) allows for accurate discrimination between peripheral and central causes of acute vestibular syndrome (AVS). In this study, we sought to investigate whether the accuracy of artificial intelligence (AI...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2022-09-01
|
| Series: | Frontiers in Neurology |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fneur.2022.919777/full |
| _version_ | 1811275070507581440 |
|---|---|
| author | Athanasia Korda Wilhelm Wimmer Wilhelm Wimmer Thomas Wyss Efterpi Michailidou Ewa Zamaro Franca Wagner Marco D. Caversaccio Georgios Mantokoudis |
| author_facet | Athanasia Korda Wilhelm Wimmer Wilhelm Wimmer Thomas Wyss Efterpi Michailidou Ewa Zamaro Franca Wagner Marco D. Caversaccio Georgios Mantokoudis |
| author_sort | Athanasia Korda |
| collection | DOAJ |
| description | ObjectiveMeasuring the Vestibular-Ocular-Reflex (VOR) gains with the video head impulse test (vHIT) allows for accurate discrimination between peripheral and central causes of acute vestibular syndrome (AVS). In this study, we sought to investigate whether the accuracy of artificial intelligence (AI) based vestibular stroke classification applied in unprocessed vHIT data is comparable to VOR gain classification.MethodsWe performed a prospective study from July 2015 until April 2020 on all patients presenting at the emergency department (ED) with signs of an AVS. The patients underwent vHIT followed by a delayed MRI, which served as a gold standard for stroke confirmation. The MRI ground truth labels were then applied to train a recurrent neural network (long short-term memory architecture) that used eye- and head velocity time series extracted from the vHIT examinations.ResultsWe assessed 57 AVS patients, 39 acute unilateral vestibulopathy patients (AUVP) and 18 stroke patients. The overall sensitivity, specificity and accuracy for detecting stroke with a VOR gain cut-off of 0.57 was 88.8, 92.3, and 91.2%, respectively. The trained neural network was able to classify strokes with a sensitivity of 87.7%, a specificity of 88.4%, and an accuracy of 87.9% based on the unprocessed vHIT data. The accuracy of these two methods was not significantly different (p = 0.09).ConclusionAI can accurately diagnose a vestibular stroke by using unprocessed vHIT time series. The quantification of eye- and head movements with the use of machine learning and AI can serve in the future for an automated diagnosis in ED patients with acute dizziness. The application of different neural network architectures can potentially further improve performance and enable direct inference from raw video recordings. |
| first_indexed | 2024-04-12T23:31:27Z |
| format | Article |
| id | doaj.art-4d200fdbac49400780330b1188df4d3a |
| institution | Directory Open Access Journal |
| issn | 1664-2295 |
| language | English |
| last_indexed | 2024-04-12T23:31:27Z |
| publishDate | 2022-09-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Neurology |
| spelling | doaj.art-4d200fdbac49400780330b1188df4d3a2022-12-22T03:12:16ZengFrontiers Media S.A.Frontiers in Neurology1664-22952022-09-011310.3389/fneur.2022.919777919777Artificial intelligence for early stroke diagnosis in acute vestibular syndromeAthanasia Korda0Wilhelm Wimmer1Wilhelm Wimmer2Thomas Wyss3Efterpi Michailidou4Ewa Zamaro5Franca Wagner6Marco D. Caversaccio7Georgios Mantokoudis8Department of Otorhinolaryngology, Head and Neck Surgery, Inselspital, University Hospital Bern and University of Bern, Bern, SwitzerlandDepartment of Otorhinolaryngology, Head and Neck Surgery, Inselspital, University Hospital Bern and University of Bern, Bern, SwitzerlandHearing Research Laboratory, ARTORG Center, University of Bern, Bern, SwitzerlandDepartment of Otorhinolaryngology, Head and Neck Surgery, Inselspital, University Hospital Bern and University of Bern, Bern, SwitzerlandDepartment of Otorhinolaryngology, Head and Neck Surgery, Inselspital, University Hospital Bern and University of Bern, Bern, SwitzerlandDepartment of Otorhinolaryngology, Head and Neck Surgery, Inselspital, University Hospital Bern and University of Bern, Bern, SwitzerlandUniversity Institute of Diagnostic and Interventional Neuroradiology, Inselspital, University Hospital Bern and University of Bern, Bern, SwitzerlandDepartment of Otorhinolaryngology, Head and Neck Surgery, Inselspital, University Hospital Bern and University of Bern, Bern, SwitzerlandDepartment of Otorhinolaryngology, Head and Neck Surgery, Inselspital, University Hospital Bern and University of Bern, Bern, SwitzerlandObjectiveMeasuring the Vestibular-Ocular-Reflex (VOR) gains with the video head impulse test (vHIT) allows for accurate discrimination between peripheral and central causes of acute vestibular syndrome (AVS). In this study, we sought to investigate whether the accuracy of artificial intelligence (AI) based vestibular stroke classification applied in unprocessed vHIT data is comparable to VOR gain classification.MethodsWe performed a prospective study from July 2015 until April 2020 on all patients presenting at the emergency department (ED) with signs of an AVS. The patients underwent vHIT followed by a delayed MRI, which served as a gold standard for stroke confirmation. The MRI ground truth labels were then applied to train a recurrent neural network (long short-term memory architecture) that used eye- and head velocity time series extracted from the vHIT examinations.ResultsWe assessed 57 AVS patients, 39 acute unilateral vestibulopathy patients (AUVP) and 18 stroke patients. The overall sensitivity, specificity and accuracy for detecting stroke with a VOR gain cut-off of 0.57 was 88.8, 92.3, and 91.2%, respectively. The trained neural network was able to classify strokes with a sensitivity of 87.7%, a specificity of 88.4%, and an accuracy of 87.9% based on the unprocessed vHIT data. The accuracy of these two methods was not significantly different (p = 0.09).ConclusionAI can accurately diagnose a vestibular stroke by using unprocessed vHIT time series. The quantification of eye- and head movements with the use of machine learning and AI can serve in the future for an automated diagnosis in ED patients with acute dizziness. The application of different neural network architectures can potentially further improve performance and enable direct inference from raw video recordings.https://www.frontiersin.org/articles/10.3389/fneur.2022.919777/fullvertigoartificial intelligencevideo head impulse teststroke diagnosisemergency department |
| spellingShingle | Athanasia Korda Wilhelm Wimmer Wilhelm Wimmer Thomas Wyss Efterpi Michailidou Ewa Zamaro Franca Wagner Marco D. Caversaccio Georgios Mantokoudis Artificial intelligence for early stroke diagnosis in acute vestibular syndrome Frontiers in Neurology vertigo artificial intelligence video head impulse test stroke diagnosis emergency department |
| title | Artificial intelligence for early stroke diagnosis in acute vestibular syndrome |
| title_full | Artificial intelligence for early stroke diagnosis in acute vestibular syndrome |
| title_fullStr | Artificial intelligence for early stroke diagnosis in acute vestibular syndrome |
| title_full_unstemmed | Artificial intelligence for early stroke diagnosis in acute vestibular syndrome |
| title_short | Artificial intelligence for early stroke diagnosis in acute vestibular syndrome |
| title_sort | artificial intelligence for early stroke diagnosis in acute vestibular syndrome |
| topic | vertigo artificial intelligence video head impulse test stroke diagnosis emergency department |
| url | https://www.frontiersin.org/articles/10.3389/fneur.2022.919777/full |
| work_keys_str_mv | AT athanasiakorda artificialintelligenceforearlystrokediagnosisinacutevestibularsyndrome AT wilhelmwimmer artificialintelligenceforearlystrokediagnosisinacutevestibularsyndrome AT wilhelmwimmer artificialintelligenceforearlystrokediagnosisinacutevestibularsyndrome AT thomaswyss artificialintelligenceforearlystrokediagnosisinacutevestibularsyndrome AT efterpimichailidou artificialintelligenceforearlystrokediagnosisinacutevestibularsyndrome AT ewazamaro artificialintelligenceforearlystrokediagnosisinacutevestibularsyndrome AT francawagner artificialintelligenceforearlystrokediagnosisinacutevestibularsyndrome AT marcodcaversaccio artificialintelligenceforearlystrokediagnosisinacutevestibularsyndrome AT georgiosmantokoudis artificialintelligenceforearlystrokediagnosisinacutevestibularsyndrome |