20240023879. A COMPUTER-IMPLEMENTED MODEL FOR PREDICTING OCCURRENCE OF A SEIZURE AND TRAINING METHOD THEREOF simplified abstract (INSERM (INSTITUT NATIONAL DE LA SANTÉ ET DE LA RECHERCHE MÉDICALE))
Contents
A COMPUTER-IMPLEMENTED MODEL FOR PREDICTING OCCURRENCE OF A SEIZURE AND TRAINING METHOD THEREOF
Organization Name
INSERM (INSTITUT NATIONAL DE LA SANTÉ ET DE LA RECHERCHE MÉDICALE)
Inventor(s)
A COMPUTER-IMPLEMENTED MODEL FOR PREDICTING OCCURRENCE OF A SEIZURE AND TRAINING METHOD THEREOF - A simplified explanation of the abstract
This abstract first appeared for US patent application 20240023879 titled 'A COMPUTER-IMPLEMENTED MODEL FOR PREDICTING OCCURRENCE OF A SEIZURE AND TRAINING METHOD THEREOF
Simplified Explanation
The invention is a method for training a model to predict the occurrence of an epileptic seizure. The method involves using a supervised training approach with a nonlinear binary classification model. The model takes as input the evaluation of the intensity of each prodromal symptom, as reported by the patient, from a predefined set of prodromal symptoms. It then outputs a classification of the patient belonging to either a pre-ictal or inter-ictal state. The training dataset consists of data inputs obtained from multiple epileptic patients, where each data input includes the evaluation of the intensity of the predefined set of prodromal symptoms and an indication of the patient's state at the time of the evaluation.
- The method trains a model to predict epileptic seizures based on the evaluation of prodromal symptoms.
- The model is a nonlinear binary classification model.
- The model takes as input the intensity evaluation of each prodromal symptom from a predefined set.
- The model outputs a classification of the patient's state as pre-ictal or inter-ictal.
- The training dataset includes data inputs from multiple epileptic patients.
- Each data input includes the evaluation of prodromal symptoms and the patient's state at the time of evaluation.
Potential Applications:
- Early warning system for epileptic seizures.
- Personalized treatment plans for epileptic patients.
- Monitoring and management of epilepsy in clinical settings.
Problems Solved:
- Predicting the occurrence of epileptic seizures.
- Identifying the pre-ictal state in epileptic patients.
- Providing timely interventions to prevent or mitigate seizures.
Benefits:
- Improved quality of life for epileptic patients through better seizure prediction.
- Reduced risk of injury or complications associated with seizures.
- More effective use of healthcare resources by targeting interventions to high-risk periods.
Original Abstract Submitted
the invention relates to a method for training a model for predicting occurrence of an epileptic seizure, the method comprising performing a supervised training over a training dataset of a nonlinear binary classification model configured to receive as input the evaluation, by a patient, of the intensity of each prodromal symptom among a predefined set of prodromal symptoms, and to output a classification of said patient belonging either to a pre-ictal or inter-ictal state, and the training dataset comprises data inputs obtained from a plurality of epileptic patients, each data input comprising an evaluation, by a patient, of the intensity of each of the predefined set of prodromal symptoms, each data input being further associated to an indication of said patient belonging to a pre-ictal or inter-ictal state at the time of the evaluation. the invention also relates to a prediction model obtained accordingly, and a computing device for implementing said prediction model.
