By Marcello Cherchi, MD PhD
For many years we have used modular intake questionnaires in order to elicit information from a patient, with the goal of constructing a preliminary history. The information is quite detailed, but also rather formulaic.
Clinicians and researchers who notice the formulaic nature of eliciting history have sought to systematize history-taking in such a way as to algorithmize the diagnostic process. In the 1990s simple computer programs were used for this purpose (Gavilan, Gallego et al. 1990). More recent attempts have employed more complex algorithms, such as “constellatory reasoning” (Feil, Feuerecker et al. 2018). Subsequent literature attempted to validate a variety of questionnaire-based approaches (Jacobson, Piker et al. 2019, Bamiou, Kikidis et al. 2022, Filippopulos, Strobl et al. 2022, Yu, Wu et al. 2022).
These approaches appropriately aim to diagnose some of the most common otoneurological diseases correctly, such as benign paroxysmal positional vertigo, vestibular neuritis, labyrinthitis, Ménière’s disease and migraine associated vertigo. We believe that these algorithms likely hold promise at the level of triaging patients.
We also think that at their present level, these algorithms are unlikely to be very useful in the setting of a tertiary-care otoneurology clinic whose clinicians are expected to recognize less common diseases correctly. It is certainly possible that with accumulated data, and with more advanced techniques (such as neural networks and other forms of machine learning), it may become feasible to devise algorithms that are sufficiently granular to recognize more esoteric pathology.
References
Bamiou DE, Kikidis D, Bibas T, Koohi N, Macdonald N, Maurer C, Wuyts FL, Ihtijarevic B, Celis L, Mucci V, Maes L, Van Rompaey V, Van de Heyning P, Nazareth I, Exarchos TP, Fotiadis D, Koutsouris D, Luxon LM (2022) Diagnostic accuracy and usability of the EMBalance decision support system for vestibular disorders in primary care: proof of concept randomised controlled study results. J Neurol 269: 2584-2598. doi: 10.1007/s00415-021-10829-7
Feil K, Feuerecker R, Goldschagg N, Strobl R, Brandt T, von Muller A, Grill E, Strupp M (2018) Predictive Capability of an iPad-Based Medical Device (medx) for the Diagnosis of Vertigo and Dizziness. Front Neurol 9: 29. doi: 10.3389/fneur.2018.00029
Filippopulos FM, Strobl R, Belanovic B, Dunker K, Grill E, Brandt T, Zwergal A, Huppert D (2022) Validation of a comprehensive diagnostic algorithm for patients with acute vertigo and dizziness. Eur J Neurol 29: 3092-3101. doi: 10.1111/ene.15448
Gavilan C, Gallego J, Gavilan J (1990) ‘Carrusel’: an expert system for vestibular diagnosis. Acta Otolaryngol 110: 161-7. doi: 10.3109/00016489009122532
Jacobson GP, Piker EG, Hatton K, Watford KE, Trone T, McCaslin DL, Bennett ML, Rivas A, Haynes DS, Roberts RA (2019) Development and Preliminary Findings of the Dizziness Symptom Profile. Ear Hear 40: 568-576. doi: 10.1097/AUD.0000000000000628
Yu F, Wu P, Deng H, Wu J, Sun S, Yu H, Yang J, Luo X, He J, Ma X, Wen J, Qiu D, Nie G, Liu R, Hu G, Chen T, Zhang C, Li H (2022) A Questionnaire-Based Ensemble Learning Model to Predict the Diagnosis of Vertigo: Model Development and Validation Study. J Med Internet Res 24: e34126. doi: 10.2196/34126
