[Code Addition Request]: A neurodiagnostic model for Parkinson’s detection

[Sriiishtiiiii]

Have you completed your first issue?

• I have completed my first issue

Guidelines

• I have read the guidelines
• I have the link to my latest merged PR

Latest Merged PR Link

This is my first issue.

Project Description

Project Title

Parkinson’s Disease Prediction using SVM on Biomedical Voice Features

---

Technical Description

• Objective
  Design a machine learning pipeline for binary classification of Parkinson’s Disease using clinical voice biomarker data.

• Dataset Characteristics

  • Source: parkinsons.csv
  • Features: 22 biomedical phonation measures including jitter, shimmer, NHR, HNR, RPDE, DFA, and frequency variation metrics
  • Target: status (1 = Parkinson’s, 0 = healthy)

• Preprocessing Steps

  • Removed identifier column (name) to eliminate data leakage
  • Separated input features (X) and target label (Y)
  • Applied Z-score normalization using StandardScaler to standardize feature distributions
  • Performed 80:20 train-test split with fixed random state for reproducibility

• Model Architecture

  • Employed a Support Vector Machine (SVM) with a linear kernel
  • Constructed a maximum-margin hyperplane to classify high-dimensional input space
  • Implemented using scikit-learn's svm.SVC

• Evaluation Metrics

  • Measured accuracy on both training and testing sets
  • Achieved strong generalization performance, confirming model robustness

• Prediction Module

  • Built a runtime inference function for new patient data inputs
  • Applied trained scaler and SVM model to return binary predictions in real time

• Technical Strengths

  • Complete ML workflow from preprocessing to deployment
  • Effective modeling of neurodegenerative indicators via margin-based classification
  • Clinical relevance through use of non-invasive voice biomarker data

---

Expected Deliverables

• A fully functional, end-to-end Parkinson’s Disease classification model
• Python-based implementation in a Jupyter Notebook
• Clean, commented code with modular structure and reproducible results
• Trained SVM model and saved inference-ready pipeline
• Evaluation metrics (accuracy) reported on both training and testing datasets
• A predictive system capable of real-time diagnosis simulation from raw feature input

Full Name

Srishti Chamoli

Participant Role

I have actively contributed to numerous full-stack open-source projects during programs like GirlScript Summer of Code (GSSoC) and Hacktoberfest previously last year. My contributions spanned both frontend and backend, whereas I am deeply interested in ML research work as well.

[github-actions]

🙌 Thank you for bringing this issue to our attention! We appreciate your input and will investigate it as soon as possible.

Feel free to join our community on Discord to discuss more!

[github-actions]

✅ This issue has been closed. Thank you for your contribution! If you have any further questions, feel free to join our community on Discord to discuss more!