Abstract:Background: To design a 3D nystagmus recognition model and automatic diagnosis system based on deep neural network architectures.Methods: An object detection model was constructed to track the movement of the centre of the pupil; then, convolutional neural network (CNN) based models were trained to detect nystagmus patterns in three dimensions (horizontal/vertical/torsional) using clinic videos from 854 patients, including 3496 videos for horizontal nystagmus and 5962 videos for vertical/torsional nystagmus. Finally, the statistical metrics generated during model predictions were calculated for automatic BPPV diagnosis. An independent test set of 502 cases generated from 457 patients was used to evaluate the overall performance of our BPPV diagnosis system.Findings: Our nystagmus detection models obtained high AUC, 0·982 in horizontal tests, 0·893 in vertical tests, and 0·957 in torsional tests. Compared with previous studies, our nystagmus model was comparable and interpretable. Moreover, our automatic BPPV diagnosis system achieved a sensitivity of 0·8848, specificity of 0·8841, accuracy of 0·8845, and an F1 score of 0·8914.Interpretation: The model has high sensitivity and specificity for the automated detection of BPPV in nystagmus videos. The proposed method can be applied in real-world medical practice.Trial Registration Details: The study was registered in the Chinese Clinical Trials Registry (ChiCTR.org.cn) under the identifier ChiCTR-IOR-1711-506. Funding Information: This work was supported by National Key R&D Project of Ministry of Science and Technology (2019YFC0119900), National Natural Science Foundation of China (82020108008), Shanghai Municipal Education Commission-Gaofeng Clinical Medicine Grant (20191921), Sino-UK Industrial Fund, UK (RP202G0289), Global Challenges Research Fund, UK (P202PF11), Interdisciplinary Program of Shanghai Jiao Tong University (YG2022QN084).Declaration of Interests: The authors declare that they have no competing interests.Ethics Approval Statement: The study was approved by the Ethics Committee of the Shanghai Sixth People’s Hospital and conducted according to the Declaration of Helsinki. Informed consent was obtained from all the participants.

Nystagmus Signal Feature Extraction and Tracking for Diagnosis of the Vestibular System

Deep learning based torsional nystagmus detection for dizziness and vertigo diagnosis

Eye Blink Artifact Detection with Novel Optimized Multi-Dimensional Electroencephalogram Features

Sparse classification of discriminant nystagmus features using combined video-oculography tests and pupil tracking for common vestibular disorder recognition

Torsional nystagmus recognition based on deep learning for vertigo diagnosis

Nystagmus patterns classification framework based on deep learning and optical flow

Hybrid clustering system using Nystagmus parameters discrimination for vestibular disorder diagnosis.

A nystagmus extraction system using artificial intelligence for video-nystagmography

A Deep Learning Model for Three-Dimensional Nystagmus Detection and Its Preliminary Application

Application and research on fast pupil location method used in real-time nystagmus detecting system

Vertical Nystagmus Recognition Based on Deep Learning

A Deep-Learning Model for 3d Nystagmus Detection and its Primary Application

Pupil Detection and Segmentation for Diagnosis of Nystagmus with U-Net

Automatic nystagmus detection and quantification in long-term continuous eye-movement data

[Infrared video-nystagmograph tracing]

Feasibility of video-based real-time nystagmus tracking: a lightweight deep learning model approach using ocular object segmentation

1D Convolutional Neural Networks for Detecting Nystagmus

Digital biomarkers from gaze tests for classification of central and peripheral lesions in acute vestibular syndrome

A New Videonystagmograph Based on Image Segmentation

A Pupil Center Positioning Method For Vertigo Diagnostic