A Questionnaire-Based Ensemble Learning Model to Predict the Diagnosis of Vertigo: Model Development and Validation Study.

BACKGROUND: Questionnaires have been used in the past 2 decades to predict the diagnosis of vertigo and assist clinical decision-making. A questionnaire-based machine learning model is expected to improve the efficiency of diagnosis of vestibular disorders. OBJECTIVE: This study aims to develop and validate a questionnaire-based machine learning model that predicts the diagnosis of vertigo. METHODS: In this multicenter prospective study, patients presenting with vertigo entered a consecutive cohort at their first visit to the ENT and vertigo clinics of 7 tertiary referral centers from August 2019 to March 2021, with a follow-up period of 2 months. All participants completed a diagnostic questionnaire after eligibility screening. Patients who received only 1 final diagnosis by their treating specialists for their primary complaint were included in model development and validation. The data of patients enrolled before February 1, 2021 were used for modeling and cross-validation, while patients enrolled afterward entered external validation. RESULTS: A total of 1693 patients were enrolled, with a response rate of 96.2% (1693/1760). The median age was 51 (IQR 38-61) years, with 991 (58.5%) females; 1041 (61.5%) patients received the final diagnosis during the study period. Among them, 928 (54.8%) patients were included in model development and validation, and 113 (6.7%) patients who enrolled later were used as a test set for external validation. They were classified into 5 diagnostic categories. We compared 9 candidate machine learning methods, and the recalibrated model of light gradient boosting machine achieved the best performance, with an area under the curve of 0.937 (95% CI 0.917-0.962) in cross-validation and 0.954 (95% CI 0.944-0.967) in external validation. CONCLUSIONS: The questionnaire-based light gradient boosting machine was able to predict common vestibular disorders and assist decision-making in ENT and vertigo clinics. Further studies with a larger sample size and the participation of neurologists will help assess the generalization and robustness of this machine learning method.

[Characteristics of nystagmus and its therapeutic effect in the reduction of posterior semicircular canal benign paroxysmal positional vertigo].

Objective:By observing the nystagmus forms at different positions during Epley reduction in patients with posterior semicircular canal benign parochial positional vertigo（PC-BPPV）, to compare the effects of the first reduction, and to further summarize the possible causes of reduction failure and the prognosis of patients with different nystagmus forms. Methods:A retrospective analysis of 191 PC-BPPV patients diagnosed with PC-BPPV showed that 95 cases had positive nystagmus during the reduction process（group A）, 69 cases had no nystagmus （group B）, and 27 cases had reverse nystagmus （group C）. All the patients were diagnosed by G-Force BPPV and were treated through simulative Epley CRP. The form of nystagmus and the effect of the first reduction were compared. Results:The first reduction effect of group A was better than that of group C（P<0.05）. The reduction effect of group B was better than that of group C（P<0.05）. There was no significant difference in the reduction effect between group A and group B（P>0.05）. Conclusion:The form of nystagmus in the reduction process of PC-BPPV has a certain guiding significance for the evaluation of the first reduction effect and the successful reduction. The reverse nystagmus during the reduction process indicates a poor reduction effect.

[Analysis of reduction effect of the evoked nystagmus in the non-affect side during Dix-Hallpike or Roll-test in unilateral posterior semicircular canal benign paroxysmal positional vertigo].

Objective:Comparative analysis of the reduction effect of the evoked nystagmus in the non-affect side during Dix-Hallpike（D-H） or Roll-test in unilateral posterior semicircular canal benign paroxysmal positional vertigo（PC-BPPV） and PC-BPPV without above evoked nystagmus. Method:Retrospective analysis of 210 patients diagnosed with unilateral PC-BPPV by G-Force BPPV CRP system was made. Among them, 18 patients exhibited positive nystagmus only when the non-affected side was stimulated by D-H test（Group A）, 30 was evoked only when stimulated by Roll-test（Group B）, 26 was evoked when stimulated by both Roll-test and the non-affected side D-H test（Group C）, 136 without nystagmus in the above positions（Group D）. All the patients were diagnosed by G-Force BPPV and were treated through simulative Epley or Semont CRP. Compare the reduction effect among the groups. Result:At the first time ,the reduction effect of nystagmus in Group D was superior to those in Group A and Group C（P<0.05）. There was no difference between Group D and Group B（P>0.05）. The difference between Group A and Group C was also non-significant（P>0.05）. The average CRP times of the four groups（CRP until nystagmus disappeared or no longer alleviated） were 1.44±0.63（Group A）, 1.46±0.65（Group B）, 1.52±0.87（Group C） and 1.48±0.73（Group D）respectively. There were no statistic difference between four groups（P>0.05）. The differences of final reduction effect between groups were the same as those at the first time. Conclusion:The first time reduction effect was less effective when nystagmus evoked the non-affect side during D-H test in unilateral PC-BPPV, while it might be irrelevant to the nystagmus evoked only in Roll-test. Although the times of CRP were similar among the groups, the final reduction effect of groups with nystagmus evoked the non-affect side during D-H was poorer.