Multiple Sclerosis Disease Prediction

Keywords: Multiple, Sclerosis

Background

This study develops a predictive model using deep learning techniques to accurately predict the occurrence of multiple sclerosis (MS) diseases. By analyzing patient demographics, medical history, and diagnostic tests, the model demonstrates promising accuracy in identifying the presence or likelihood of MS. This research provides valuable insights for early diagnosis, intervention, and improved patient care. Imbalanced datasets, where the number of MS cases is significantly lower than non-MS cases, can pose challenges for training the model. Balancing the dataset or employing specialized techniques to handle class imbalance is necessary for accurate predictions. To develop an accurate predictive model using deep learning techniques for the early detection and prediction of multiple sclerosis (MS) diseases. The model should analyze various patient factors such as demographics, medical history, and diagnostic tests to predict the presence or likelihood of MS. After passing the data into a neural network, the author preprocesses the data using a standard scale. As it is a classification problem statement author used binary cross entropy as a loss. after running the 50 epochs and passing the scaler data in the model fit validation accuracy got 0.58 and loss is -27.

Aim & Objectives

Aim
To predict the Multiple Sclerosis Disease using DL Techniques
Objective
Develop an ANN model for accurate prediction of Multiple Sclerosis disease, aiding in early diagnosis and treatment.
Explore ANN architectures to enhance disease prediction by analyzing medical history and relevant factors.
Evaluate the ANN's performance in predicting Multiple Sclerosis, aiming to improve patient care and outcomes.

Research Methodology

ANN Technique used

Software & Tools

Jupyter, Google Collab

— Preview truncated. The complete document (full chapters, references, diagrams, and appendices) is shared with clients as part of project delivery. —