Parkinson's disease analysis using logistic regression

ROC curve plot
to create a confusion matrix

Inspiration

Parkinson's disease is the second most common disease after Alzheimer's and affects 1 million people in India annually it can't be cured early or at a later stage, it can be just predicted.

What it does

Parkinson's disease is a chronic condition that damages brain neurons and includes five phases. According to various research, the progressive nature and symptoms of this disease reduce the quality of life and cause various disabilities to human beings. Symptoms of this disease includes - Changes in speech and Writing, Fatigue and Drowsiness, Poor body balance, Reduction in weight, spasm in voice box, Tension and Depression.

How we built it

Artificial Intelligence and Machine Learning can be effective tools to help early prediction of this disease. The main objective of this proposed model is to understand what Parkinson’s disease is and to detect the early onset of the disease using machine learning algorithms and previous datasets. Data is first cleaned using a data mining process and XGBoost is used to classify the data into testing and training. Using logical regression, one can forecast the likelihood of a target variable, by classifying it into binary values that indicate success or failure, and the classification accuracy is resulted.

Challenges we ran into

Finding a suitable data set was one of the major problems, Initially we faced problems in the mining process after a particular time we overcame it.

Accomplishments that proud of

Train accuracy we're obtained is 3.2%, test accuracy is obtained is 90.13%.

What we learned

To indicate success or failure and the classification accuracy we got in-depth knowledge in logistic regression and to classify the data into testing and training, we are more aware of the XGBoost algorithm.

What's next for Parkinson's disease analysis using logistic regression

these findings improve the information that researchers and medical professionals working with Parkinson's disease have access to, allowing for the disease's early diagnosis.