Browse by author
Lookup NU author(s): Dr David Lewis-Smith
Full text for this publication is not currently held within this repository. Alternative links are provided below where available.
The foundation of clinical epileptology is the interpretation of seizures. The recognition of a particular type of seizure enables the clinician to use established knowledge of its associations with epileptic syndromes, underlying pathological processes, co-morbidities, clinical trajectories and treatment responses to improve patient care. Communication of the types of seizure affecting a patient to non-specialist colleagues in an accurate and comprehensible, yet concise manner is a challenge. For example, in the clinical setting the epileptologist may need to categorise the patient’s seizures to guide the clinical geneticist’s selection of tests and subsequent interpretation of their results. In multidisciplinary academic environments, not all researchers can be expected to have detailed knowledge of seizure semiologies and their classification; however, types of seizure (whether considered independently or as the core features of syndromes) are often important variables to consider during analysis. Additionally, in order to maximise the yield of valuable data from people with epilepsy we should ensure that seizure data are recorded in a format that facilitates their meaningful re-use, for example through mega-analysis or re-analysis to validate independently reported findings, or to generate or test new hypotheses. Such re-use may require participants’ seizures to be recategorised at a different degree of specificity from the original study; ideally the format of seizure data should allow this to be automated. The Human Phenotype Ontology (HPO) is commonly used for the concise communication of clinical features between clinicians and non-clinicians in clinical genetic services and in research domains. It provides both standardised terminology for individual clinical phenotypic abnormalities (annotating each with definitions and synonyms) and an acyclic graph (a complex hierarchy-like structure) representing the logical relationship between them. This allows algorithmic interpretation of clinical features at the desired degree of specificity by humans or computers. Within the Epilepsiome Task Force of the International League Against Epilepsy’s (ILAE) Genetics Commission we recognised the importance of an accurate representation of seizure types in the HPO for the age of genomic medicine and big-data analyses. We have redesigned the HPO’s seizure subdivision basing its vocabulary and structure on the ILAE’s current classifications of seizures (2017) and status epilepticus (2015), and the currently proposed classification of neonatal seizures. Using examples, we discuss its advantages for the interpretation of ILAE seizure terminology – particularly by computer algorithms and by colleagues who are not clinical epileptologists – in order to compare individuals to each other, or to categorise them by their seizure types for research and clinical diagnostic services.
Author(s): Lewis-Smith D, Balagura G, Kearney H, Galer P, Ganesan S, Gan G, Krause R, Robinson P, Helbig I
Publication type: Conference Proceedings (inc. Abstract)
Publication status: Published
Conference Name: International League Against Epilepsy British Branch Annual Scientific Meeting
Year of Conference: 2019
Online publication date: 02/10/2019
Acceptance date: 22/07/2019
Publisher: ILAE
URL: https://www.ilae.org/congresses/2019-ilae-british-branch-annual-scientific-meeting
