Putting harmful brain activity into groups has become an important area of study that has a big effect on finding and stopping neurological disorders (ND). Because about 3 billion people around the world have some kind of ND, and it can take long time and hard to figure out what’s wrong with them. On average, they have to wait two years for a final diagnosis. It takes a long time and a lot of different tests to confirm that someone has an ND. Some of these tests are non-invasive, like EEGs, ultrasounds or simple reflex tests. Others, like spine taps or biopsies, are more invasive. Electroencephalogram (EEG) is one of the first tests that is usually done on a patient. It is done by putting several devices on the head and measuring the electrical activity of the brain. This shows patterns of neural activity happening inside the brain. It is hard to decode brain patterns or put them into groups. So, the goal of this article is to shorten the time to diagnose using a long short-term memory (LSTM) model to find harmful brain activity patterns in spectrogram data. This could lead to better outcomes for patients.