New approaches to studying epilepsy and comorbidities

The MINI involves asking subjects five open-ended questions to harvest the language that would serve as the input to the machine-learning model: Do you have hope? Do you have any secrets? Are you angry? Do you have any fear? Does it hurt emotionally?¹ All interviews were conducted by one of two experienced interviewers (a certified clinical research professional or a research nurse practitioner).¹

References:

1.Glauser T, Santel D, DelBello M, et al. Identifying epilepsy psychiatric comorbidities with machine learning. Acta Neurol Scand 2020; 141 (5): 388–396.

2.Lo Barco T, Garcelon N, Neuraz A, Nabbout R. Natural history of rare diseases using natural language processing of narrative unstructured electronic health records: the example of Dravet syndrome. Epilepsia 2024; 65 (2): 350–361.

3.Biondi A, Winsor AA, Ebelthite C, et al. A comprehensive digital mental health screening tool for people with epilepsy: a feasibility study in UK outpatients. Epilepsy Behav 2024; 160: 110103.