Parkinson Disease Prediction using Machine Learning in Python

Parkinson's Disease is a neurodegenerative disorder that affects millions worldwide, early and accurate diagnosis is crucial for effective treatment which can easily be done using machine learning in Python.

This article explores the application of machine learning techniques in predicting Parkinson's Disease using a dataset from the UCI machine learning repository. By employing the Random Forest Classifier algorithm, we demonstrate how Python can be leveraged to analyze and preprocess data, train a predictive model, and make accurate predictions.

Parkinson Disease Prediction using Machine Learning in Python

We have used Jupyter notebook to run the code in this article.

Below are the steps that we will follow for Parkinson Disease Prediction using Machine Learning in Python −

Step 1:Import necessary libraries

Example

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
from sklearn.metrics import accuracy_score, confusion_matrix
from sklearn.preprocessing import StandardScaler
from sklearn.decomposition import PCA
from sklearn.model_selection import train_test_split
from sklearn.ensemble import RandomForestClassifier

Step 2: Load the Parkinson's Disease dataset

The program reads the dataset from the 'parkinsons.csv' file using the pd.read_csv() function and stores it in the data variable.

Example

# Load the Parkinson's Disease dataset
data = pd.read_csv('parkinsons.csv')

Step 3: Data cleaning

The program below removes the 'name' column from the dataset using the drop() function and assigns the modified dataset back to the data variable.

Example

# Data cleaning
data = data.drop('name', axis=1)  # Remove the 'name' column

Step 4: Data preprocessing

The program below separates the features (X) from the target variable (y) using the drop() function and assigns them to the respective variables.

Example

# Data preprocessing
X = data.drop('status', axis=1)  # Features
y = data['status']  # Target variable

Step 5: Data analysis

The program below provides information about the dataset −

• The shape of the dataset (number of rows and columns) is printed using data.shape.

• The number of samples with Parkinson's Disease and healthy samples is displayed using len(data[data['status'] == 1]) and len(data[data['status'] == 0]), respectively.

• A summary of the dataset is printed using data.describe().

Example

print("Data Shape:", data.shape)
print("Parkinson's Disease Samples:", len(data[data['status'] == 1]))
print("Healthy Samples:", len(data[data['status'] == 0]))
print("\nData Summary:")
print(data.describe())

Output

Data Shape: (195, 23)
Parkinson's Disease Samples: 147
Healthy Samples: 48

Data Summary:
       MDVP:Fo(Hz)  MDVP:Fhi(Hz)  MDVP:Flo(Hz)  MDVP:Jitter(%)  \
count   195.000000    195.000000    195.000000      195.000000   
mean    154.228641    197.104918    116.324631        0.006220   
std      41.390065     91.491548     43.521413        0.004848   
min      88.333000    102.145000     65.476000        0.001680   
25%     117.572000    134.862500     84.291000        0.003460   
50%     148.790000    175.829000    104.315000        0.004940   
75%     182.769000    224.205500    140.018500        0.007365   
max     260.105000    592.030000    239.170000        0.033160   

       MDVP:Jitter(Abs)    MDVP:RAP    MDVP:PPQ  Jitter:DDP  MDVP:Shimmer  \
count        195.000000  195.000000  195.000000  195.000000    195.000000   
mean           0.000044    0.003306    0.003446    0.009920      0.029709   
std            0.000035    0.002968    0.002759    0.008903      0.018857   
min            0.000007    0.000680    0.000920    0.002040      0.009540   
25%            0.000020    0.001660    0.001860    0.004985      0.016505   
50%            0.000030    0.002500    0.002690    0.007490      0.022970   
75%            0.000060    0.003835    0.003955    0.011505      0.037885   
max            0.000260    0.021440    0.019580    0.064330      0.119080   

  
max      0.685151    0.825288   -2.434031    0.450493    3.671155    0.527367  

[8 rows x 23 columns]

Step 6: Data visualization

The histograms are shown using plt.show().

Example

# Data visualization
data.hist(figsize=(12, 12))
plt.tight_layout()
plt.show()

Output

Step 7:Data scaling

The below program scales the features using StandardScaler(), which standardizes the features by subtracting the mean and scaling to unit variance. The scaled features are stored in the X_scaled variable.

Example

scaler = StandardScaler()
X_scaled = scaler.fit_transform(X)

Step 8: Dimensionality reduction

It reduces the features to two principal components using PCA(n_components=2). The reduced features are stored in the X_pca variable.

Example

pca = PCA(n_components=2)
X_pca = pca.fit_transform(X_scaled)

Step 9: Split the dataset into training and testing sets

The program below splits the dataset into training and testing sets using train_test_split().

Example

X_train, X_test, y_train, y_test = train_test_split(X_scaled, y, test_size=0.2, random_state=42)

Step 10: Create a Classifier known as Random Forest Classifier

The program below creates an instance of the Random Forest Classifier using RandomForestClassifier().

Train the model

Example

rf_classifier = RandomForestClassifier()

# Train the model
rf_classifier.fit(X_train, y_train)

Output

RandomForestClassifier()

Step 11: Make predictions on the test set

Calculate the accuracy of the model

Example

# Make predictions on the test set
y_pred = rf_classifier.predict(X_test)

accuracy = accuracy_score(y_test, y_pred)
print("\nAccuracy:", accuracy)

Output

Accuracy: 0.9230769230769231

The program calculates the accuracy of the model by comparing the predicted labels (y_pred) with the actual labels (y_test).

Step 12: Confusion matrix

It uses the confusion_matrix() function from sklearn.metrics and assigns the confusion matrix to the cm variable.

Example

cm = confusion_matrix(y_test, y_pred)
print("\nConfusion Matrix:")
print(cm)

Output

Confusion Matrix:
[[ 5  2]
 [ 1 31]]

Conclusion

In conclusion, this article presented a machine learning approach for Parkinson's Disease prediction using Python. By utilizing the Random Forest Classifier algorithm and a comprehensive dataset, we demonstrated the effectiveness of machine learning in accurately predicting the presence of Parkinson's Disease.

The results highlight the potential of this approach in assisting healthcare professionals with early diagnosis and intervention, leading to improved patient outcomes.