Cargando…

Deep Learning for Automatically Visual Evoked Potential Classification During Surgical Decompression of Sellar Region Tumors

PURPOSE: Detection of the huge amount of data generated in real-time visual evoked potential (VEP) requires labor-intensive work and experienced electrophysiologists. This study aims to build an automatic VEP classification system by using a deep learning algorithm. METHODS: Patients with sellar reg...

Descripción completa

Detalles Bibliográficos
Autores principales: Qiao, Nidan, Song, Mengju, Ye, Zhao, He, Wenqiang, Ma, Zengyi, Wang, Yongfei, Zhang, Yuyan, Shou, Xuefei
Formato: Online Artículo Texto
Lenguaje:English
Publicado: The Association for Research in Vision and Ophthalmology 2019
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6871542/
https://www.ncbi.nlm.nih.gov/pubmed/31788350
http://dx.doi.org/10.1167/tvst.8.6.21
_version_ 1783472426555801600
author Qiao, Nidan
Song, Mengju
Ye, Zhao
He, Wenqiang
Ma, Zengyi
Wang, Yongfei
Zhang, Yuyan
Shou, Xuefei
author_facet Qiao, Nidan
Song, Mengju
Ye, Zhao
He, Wenqiang
Ma, Zengyi
Wang, Yongfei
Zhang, Yuyan
Shou, Xuefei
author_sort Qiao, Nidan
collection PubMed
description PURPOSE: Detection of the huge amount of data generated in real-time visual evoked potential (VEP) requires labor-intensive work and experienced electrophysiologists. This study aims to build an automatic VEP classification system by using a deep learning algorithm. METHODS: Patients with sellar region tumor and optic chiasm compression were enrolled. Flash VEP monitoring was applied during surgical decompression. Sequential VEP images were fed into three neural network algorithms to train VEP classification models. RESULTS: We included 76 patients. During surgical decompression, we observed 68 eyes with increased VEP amplitude, 47 eyes with a transient decrease, and 37 eyes without change. We generated 2,843 sequences (39,802 images) in total (887 sequences with increasing VEP, 276 sequences with decreasing VEP, and 1680 sequences without change). The model combining convolutional and recurrent neural network had the highest accuracy (87.4%; 95% confidence interval, 84.2%–90.1%). The sensitivity of predicting no change VEP, increasing VEP, and decreasing VEP was 92.6%, 78.9%, and 83.7%, respectively. The specificity of predicting no change VEP, increasing VEP, and decreasing VEP was 80.5%, 93.3%, and 100.0%, respectively. The class activation map visualization technique showed that the P2-N3-P3 complex was important in determining the output. CONCLUSIONS: We identified three VEP responses (no change, increase, and decrease) during transsphenoidal surgical decompression of sellar region tumors. We developed a deep learning model to classify the sequential changes of intraoperative VEP. TRANSLATIONAL RELEVANCE: Our model may have the potential to be applied in real-time monitoring during surgical resection of sellar region tumors.
format Online
Article
Text
id pubmed-6871542
institution National Center for Biotechnology Information
language English
publishDate 2019
publisher The Association for Research in Vision and Ophthalmology
record_format MEDLINE/PubMed
spelling pubmed-68715422019-11-29 Deep Learning for Automatically Visual Evoked Potential Classification During Surgical Decompression of Sellar Region Tumors Qiao, Nidan Song, Mengju Ye, Zhao He, Wenqiang Ma, Zengyi Wang, Yongfei Zhang, Yuyan Shou, Xuefei Transl Vis Sci Technol Articles PURPOSE: Detection of the huge amount of data generated in real-time visual evoked potential (VEP) requires labor-intensive work and experienced electrophysiologists. This study aims to build an automatic VEP classification system by using a deep learning algorithm. METHODS: Patients with sellar region tumor and optic chiasm compression were enrolled. Flash VEP monitoring was applied during surgical decompression. Sequential VEP images were fed into three neural network algorithms to train VEP classification models. RESULTS: We included 76 patients. During surgical decompression, we observed 68 eyes with increased VEP amplitude, 47 eyes with a transient decrease, and 37 eyes without change. We generated 2,843 sequences (39,802 images) in total (887 sequences with increasing VEP, 276 sequences with decreasing VEP, and 1680 sequences without change). The model combining convolutional and recurrent neural network had the highest accuracy (87.4%; 95% confidence interval, 84.2%–90.1%). The sensitivity of predicting no change VEP, increasing VEP, and decreasing VEP was 92.6%, 78.9%, and 83.7%, respectively. The specificity of predicting no change VEP, increasing VEP, and decreasing VEP was 80.5%, 93.3%, and 100.0%, respectively. The class activation map visualization technique showed that the P2-N3-P3 complex was important in determining the output. CONCLUSIONS: We identified three VEP responses (no change, increase, and decrease) during transsphenoidal surgical decompression of sellar region tumors. We developed a deep learning model to classify the sequential changes of intraoperative VEP. TRANSLATIONAL RELEVANCE: Our model may have the potential to be applied in real-time monitoring during surgical resection of sellar region tumors. The Association for Research in Vision and Ophthalmology 2019-11-20 /pmc/articles/PMC6871542/ /pubmed/31788350 http://dx.doi.org/10.1167/tvst.8.6.21 Text en Copyright 2019 The Authors http://creativecommons.org/licenses/by-nc-nd/4.0/ This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
spellingShingle Articles
Qiao, Nidan
Song, Mengju
Ye, Zhao
He, Wenqiang
Ma, Zengyi
Wang, Yongfei
Zhang, Yuyan
Shou, Xuefei
Deep Learning for Automatically Visual Evoked Potential Classification During Surgical Decompression of Sellar Region Tumors
title Deep Learning for Automatically Visual Evoked Potential Classification During Surgical Decompression of Sellar Region Tumors
title_full Deep Learning for Automatically Visual Evoked Potential Classification During Surgical Decompression of Sellar Region Tumors
title_fullStr Deep Learning for Automatically Visual Evoked Potential Classification During Surgical Decompression of Sellar Region Tumors
title_full_unstemmed Deep Learning for Automatically Visual Evoked Potential Classification During Surgical Decompression of Sellar Region Tumors
title_short Deep Learning for Automatically Visual Evoked Potential Classification During Surgical Decompression of Sellar Region Tumors
title_sort deep learning for automatically visual evoked potential classification during surgical decompression of sellar region tumors
topic Articles
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6871542/
https://www.ncbi.nlm.nih.gov/pubmed/31788350
http://dx.doi.org/10.1167/tvst.8.6.21
work_keys_str_mv AT qiaonidan deeplearningforautomaticallyvisualevokedpotentialclassificationduringsurgicaldecompressionofsellarregiontumors
AT songmengju deeplearningforautomaticallyvisualevokedpotentialclassificationduringsurgicaldecompressionofsellarregiontumors
AT yezhao deeplearningforautomaticallyvisualevokedpotentialclassificationduringsurgicaldecompressionofsellarregiontumors
AT hewenqiang deeplearningforautomaticallyvisualevokedpotentialclassificationduringsurgicaldecompressionofsellarregiontumors
AT mazengyi deeplearningforautomaticallyvisualevokedpotentialclassificationduringsurgicaldecompressionofsellarregiontumors
AT wangyongfei deeplearningforautomaticallyvisualevokedpotentialclassificationduringsurgicaldecompressionofsellarregiontumors
AT zhangyuyan deeplearningforautomaticallyvisualevokedpotentialclassificationduringsurgicaldecompressionofsellarregiontumors
AT shouxuefei deeplearningforautomaticallyvisualevokedpotentialclassificationduringsurgicaldecompressionofsellarregiontumors