Importance: Misdiagnosis of epilepsy is common. Video electroencephalogram provides a definitive diagnosis but is impractical for many patients referred for evaluation of epilepsy.

Objective: To evaluate the accuracy of outpatient smartphone videos in epilepsy.

Design, setting, and participants: This prospective, masked, diagnostic accuracy study (the OSmartViE study) took place between August 31, 2015, and August 31, 2018, at 8 academic epilepsy centers in the United States and included a convenience sample of 44 nonconsecutive outpatients who volunteered a smartphone video during evaluation and subsequently underwent video electroencephalogram monitoring. Three epileptologists uploaded videos for physicians from the 8 epilepsy centers to review.

Main outcomes and measures: Measures of performance (accuracy, sensitivity, specificity, positive predictive value, and negative predictive value) for smartphone video-based diagnosis by experts and trainees (the index test) were compared with those for history and physical examination and video electroencephalogram monitoring (the reference standard).

Results: Forty-four eligible epilepsy clinic outpatients (31 women [70.5%]; mean [range] age, 45.1 [20-82] years) submitted smartphone videos (530 total physician reviews). Final video electroencephalogram diagnoses included 11 epileptic seizures, 30 psychogenic nonepileptic attacks, and 3 physiologic nonepileptic events. Expert interpretation of a smartphone video was accurate in predicting a video electroencephalogram monitoring diagnosis of epileptic seizures 89.1% (95% CI, 84.2%-92.9%) of the time, with a specificity of 93.3% (95% CI, 88.3%-96.6%). Resident responses were less accurate for all metrics involving epileptic seizures and psychogenic nonepileptic attacks, despite greater confidence. Motor signs during events increased accuracy. One-fourth of the smartphone videos were correctly diagnosed by 100% of the reviewing physicians, composed solely of psychogenic attacks. When histories and physical examination results were combined with smartphone videos, correct diagnoses rose from 78.6% to 95.2%. The odds of receiving a correct diagnosis were 5.45 times greater using smartphone video alongside patient history and physical examination results than with history and physical examination alone (95% CI, 1.01-54.3; P = .02).

Conclusions and relevance: Outpatient smartphone video review by experts has predictive and additive value for diagnosing epileptic seizures. Smartphone videos may reliably aid psychogenic nonepileptic attacks diagnosis for some people.'s

 Here is what they had to say in the discussion:

"An HP (history + physical exam) had approximately 1 in 5 patients with unknown/incorrect diagnoses, but combined with a smartphone video, the yield rose to 95.2%. A prior VEM study of 5 epileptologists analyzing 41 events (34 epileptic seizures) in 30 patients had an overall video diagnostic accuracy of 65% using charted description and 88% with video, similar to this study. The odds of receiving a correct diagnosis were 5.45 times greater using smartphone video alongside HP than with HP alone (95% CI, 1.01-54.3; P = .02)."

My tuppence worth? The above findings suggest that home videos, when viewed by people who are experts (experienced at video-EEG recordings) are incorrectly diagnosed as epileptic seizures in about 1 in 14 patients and that performance is improved by a detailed history. Based on the videos alone, these experts get it wrong in about 1 in 9 times when viewing the videos of people with no epileptic and non-epileptic seizures (this excludes other events, an artificially restricted choice) and 1 in approximately 20 times when provided with both the history, examination and video findings..

The point is that such videos, even in expert hands, are imperfect. One can see why so many people are misdiagnosed with epilepsy across the world. It is not the fault of the neurologists; it is the limitation of the history and examination, with or without video findings and video-EEG recordings are a scarce resource.

Here is the link to the article Assessment of the Predictive Value of Outpatient Smartphone Videos for Diagnosis of Epileptic Seizures - PMC

 The above study involved only 43 patients, an important limitation of the study. It is remarkable that the study nevertheless made its way into JAMA Neurology. Below are a few more abstracts

We examined the usefulness of video versus EEG in isolation for the differentiation of epileptic seizures (ES) from psychogenic nonepileptic events (PNEE). Video-EEG recordings of 43 events in 43 patients (27 with ES and 16 with PNEE) were analyzed by experienced clinical epileptologists/electroencephalographers blinded to the patients' clinical histories. Both the video and EEG were scored independently by the same reader for each event. Relying on video recordings alone, the readers correctly identified ES with a sensitivity of 93% and specificity of 94%. Based on EEG data alone, the readers correctly identified ES with a sensitivity of 89% and specificity of 94%. Semiologically, a gradual evolving buildup of visible symptoms, reaching maximal behavioral intensity within 70 seconds of event onset, was a reliable indicator of ES. No patient with ES demonstrated eye closure at the time of peak behavioral manifestations. Although several additional semiologic features were statistically associated with either ES or PNEE, they were less reliably present and, hence, less clinically useful. Correct categorization of some neurobehavioral events can be made by experienced epileptologists on the basis of video or EEG recordings during an event, without simultaneous review of both provided that the full event is recorded. Home video recordings may represent a useful screening tool for a subset of patients with neurobehavioral events of unclear etiology.

The above may appear fairly impressive but bear in mind that the interpretations were done by experts with a lot of experience in reviewing video-EMG recordings. However, a specificity of 94% means that approximately one in 17 patients will be told by an expert that they have had an epileptic seizure, when they have had a non-epileptic seizure. Similar findings pertain to the EEG viewed in isolation. For people who don't work in the field of video-EEG recordings, the error rates are likely to be higher for both videos and EEGs (and even the combination). There are good reasons why the two are done together and why this combination is regarded as the gold standard. 

But wait, there is more evidence, and it is more recent: