Telemedicine Algorithm for the Management of Dizzy Patients

Artificial intelligence for early stroke diagnosis in acute vestibular syndrome

Objective

Measuring 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.

Methods

We 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.

Results

We 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).

Conclusion

AI 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.

Videooculography “HINTS” in Acute Vestibular Syndrome: A Prospective Study

Objective

A three-step bedside test (“HINTS”: Head Impulse-Nystagmus-Test of Skew), is a well-established way to differentiate peripheral from central causes in patients with acute vestibular syndrome (AVS). Nowadays, the use of videooculography gives physicians the possibility to quantify all eye movements. The goal of this study is to compare the accuracy of VOG “HINTS” (vHINTS) to an expert evaluation.

Methods

We performed a prospective study from July 2015 to April 2020 on all patients presenting at the emergency department with signs of AVS. All the patients underwent clinical HINTS (cHINTS) and vHINTS followed by delayed MRI, which served as a gold standard for stroke confirmation.

Results

We assessed 46 patients with AVS, 35 patients with acute unilateral vestibulopathy, and 11 patients with stroke. The overall accuracy of vHINTS in detecting a central pathology was 94.2% with 100% sensitivity and 88.9% specificity. Experts, however, assessed cHINTS with a lower accuracy of 88.3%, 90.9% sensitivity, and 85.7% specificity. The agreement between clinical and video head impulse tests was good, whereas for nystagmus direction was fair.

Conclusions

vHINTS proved to be very accurate in detecting strokes in patients AVS, with 9% points better sensitivity than the expert. The evaluation of nystagmus direction was the most difficult part of HINTS.

Feasibility of Telemedical HINTS (Head Impulse-Nystagmus-Test of Skew) Evaluation in Patients With Acute Dizziness or Vertigo in the Emergency Department of Primary Care Hospitals

Background

Acute dizziness, vertigo and imbalance are common symptoms in emergency departments. Stroke needs to be distinguished from vestibular diseases. A battery of three clinical bedside tests (HINTS: Head Impulse Test, Nystagmus, Test of Skew) has been shown to detect stroke as underlying cause with high reliability, but implementation is challenging in primary care hospitals. Aim of this study is to prove the feasibility of a telemedical HINTS examination via a remotely controlled videooculography (VOG) system.

Methods

The existing video system of our telestroke network TEMPiS (Telemedic Project for Integrative Stroke Care) was expanded through a VOG system. This feature enables the remote teleneurologist to assess a telemedical HINTS examination based on inspection of eye movements and quantitative video head impulse test (vHIT) evaluation. ED doctors in 11 spoke hospitals were trained in performing vHIT, nystagmus detection and alternating cover test. Patients with first time acute dizziness, vertigo or imbalance, whether ongoing or resolved, presented to the teleneurologist were included in the analysis, as long as no focal neurological deficit according to the standard teleneurological examination or obvious internal medicine cause was present and a fully trained team was available. Primary outcome was defined as the feasibility of the telemedical HINTS examination.

Results

From 01.06.2019 to 31.03.2020, 81 consecutive patients were included. In 72 (88.9%) cases the telemedical HINTS examination was performed. The complete telemedical HINTS examination was feasible in 46 cases (63.9%), nystagmus detection in all cases (100%) and alternating covert test in 70 cases (97.2%). The vHIT was recorded and interpretable in 47 cases (65.3%). Results of the examination with the VOG system yielded clear results in 21 cases (45.7%) with 14 central and 7 peripheral lesions. The main reason for incomplete examination was the insufficient generation of head impulses.

Conclusion

In our analysis the telemedical HINTS examination within a telestroke network was feasible in two thirds of the patients. This offers the opportunity to improve specific diagnostics and therapy for patients with acute dizziness and vertigo even in primary care hospitals. Improved training for spoke hospital staff is needed to further increase the feasibility of vHIT.

Teaching Telehealth During a Pandemic and Beyond: an Intern's Survival Guide for Virtual Medicine

BACKGROUND

The rapid transition to telemedicine at the onset of the COVID-19 pandemic required many providers to learn telemedicine "on the fly." As virtual care will likely remain a mainstay of outpatient medicine, it is imperative that telemedicine training be incorporated into graduate medical education.

AIM

Design a telemedicine curriculum for internal medicine residents based on principles of experiential learning.

SETTING

VA-based internal medicine primary care clinic.

PARTICIPANTS

Sixteen first-year internal medicine residents participated in the curriculum.

PROGRAM DESCRIPTION

The curriculum included a didactic session followed by four simulated patient encounters focused on troubleshooting technical issues, performing the virtual physical exam, coordinating team-based care, and tackling emergencies.

PROGRAM EVALUATION

Participants reported minimal previous experience with telemedicine. After completing the training, resident confidence in conducting video visits increased from an average score of four to seven (on a 10-point scale). Residents were more likely to agree that video visits would allow them to build bonds and effectively address their patients' needs. This increased confidence persisted at 3 months after training.

DISCUSSION

Using experiential learning, we identified strategies which increased the confidence of internal medicine trainees in conducting telemedicine visits. Further research is needed to validate our findings across different practice settings.

New Frontiers in Managing the Dizzy Patient

Despite progress in vestibular research in the last 20 years, much remains poorly understood about vestibular pathophysiology and its management. A shared language is a critical first step in understanding vestibular disorders and is under development. Telehealth will continue for patients with dizziness, and ambulatory monitoring of nystagmus will become a diagnostic tool. In the next 2 decades, it is anticipated that vestibular perceptual threshold testing will become common in tertiary centers, imaging with improved spatial resolution will yield better understanding of vestibular pathophysiology, and that vestibular implants will become a part of clinical practice.

Pitfalls in the Diagnosis of Posterior Circulation Stroke in the Emergency Setting

Posterior circulation stroke (PCS), caused by infarction within the vertebrobasilar arterial system, is a potentially life-threatening condition and accounts for about 20–25% of all ischemic strokes. Diagnosing PCS can be challenging due to the vast area of brain tissue supplied by the posterior circulation and, as a consequence, the wide range of—frequently non-specific—symptoms. Commonly used prehospital stroke scales and triage systems do not adequately represent signs and symptoms of PCS, which may also escape detection by cerebral imaging. All these factors may contribute to causing delay in recognition and diagnosis of PCS in the emergency context. This narrative review approaches the issue of diagnostic error in PCS from different perspectives, including anatomical and demographic considerations as well as pitfalls and problems associated with various stages of prehospital and emergency department assessment. Strategies and approaches to improve speed and accuracy of recognition and early management of PCS are outlined.