Predictive analytics & modeling for modern health care system for cerebral palsy patients

Abstract

Listen

Translate article

The time-shifting elements and high between individual changeability make early forecast of a seizure express a difficult errand. Numerous examinations have demonstrated that EEG signals do have profitable data that if effectively broke down, could help in the expectation of seizures in epileptic patients before their event. A few numerical changes have been broke down for its connection with seizure beginning forecast and a progression of analyses were done to ensure their qualities. New calculations are exhibited to help elucidate, screen, and cross-approve the order of EEG signs to foresee the ictal (for example seizure) states, explicitly the preictal, interictal, and postictal states in the mind. These new strategies show promising outcomes in distinguishing the nearness of a preictal stage before the ictal state. Artificial Intelligence and Machine Learning are playing major role in Diagnosis and Predicting the diseases, EEG signal were recorded with 16 channel system from brain scalp and stored in system. From recorded database seizure levels were analyzed from each channel by articrafting with ICA and PCA algorithms and processed through Band pass filter to identify the Delta, Theta, Alpha and Beta levels. After collecting the brain signals from each channel they were bided with FFT to get the data in time domain series. A machine learning model is developed using python programming by sampling on obtained EEG data, sampled data is trained on to the system. Machine Learning algorithms were applied to the data to accuracy and predict the accuracy on developed model neural network is applied to identify the True-False values. The analyzed data is integrated with IOT framework to Diagnosis the Cerebral Palsy patient remotely. Remote monitoring system for the persons with Intellectual disabilities and update the health status to care takers time to time while they are at work. It’s also reduces the patient work load by checking his details staying at home rather going to a hospital. If system identifies any changes in patient heart rate, brain signals, body temperature, the system alerts the doctor and respective relatives about patient’s status over IOT and also stores the patient details in the cloud.