Brain Tumor Detection using Machine Learning Techniques with Internet of Things

Epilepsy is a neurological condition that is rather common and is thought to afflict around 70 million individuals all over the globe. If epilepsy is to be monitored properly and successfully treated, seizures have to be recorded and logged. The present therapy for epilepsy involves the use of seizure diaries kept by caregivers; nevertheless, the use of clinical seizure detection may sometimes miss events. Wearable technologies may, in the long term, prove to be less intrusive, more pleasant, and simpler to use for ambulatory monitoring. Using biosensors placed on the wrist and ankle, custom-built machine learning (ML) algorithms are tested to see whether or not they are able to correctly recognise seizures over a broad spectrum of epileptic episodes. In this article, an automated method known as a new wireless sensor-based system is developed for the purpose of detecting and monitoring epileptic patients in an environment that is not a clinical setting. The goal of this technique is to cut down on the amount of time spent by neurologists diagnosing seizures. Using a biosensor that is worn on the wrist, a method was devised in this study for recording multi-modal data such as electroencephalogram (EEG) readings. However, excluding noise and extracting features are two important challenges that must be overcome when attempting to foresee epileptic episodes. Support Vector Machine (SVM) was used as a classifier to obtain statistical values and Lyapunov features rather than raw data in order to detect epileptic seizure activity in a shorter amount of time. This resulted in a significant improvement when compared to the methods that are currently considered to be state-of-the-art.