{"id":28148866,"url":"https://github.com/niranjanrao07/adhd-ml-project","last_synced_at":"2025-05-15T01:15:25.300Z","repository":{"id":290686549,"uuid":"973968906","full_name":"NiranjanRao07/ADHD-ML-Project","owner":"NiranjanRao07","description":"This project used machine learning to classify ADHD based on EEG data. We preprocessed the EEG signals, extracted various features, and used LDA for dimensionality reduction. A voting ensemble of classifiers achieved 72% accuracy in distinguishing between ADHD and control groups.","archived":false,"fork":false,"pushed_at":"2025-04-30T02:13:21.000Z","size":1414,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-04-30T03:23:54.411Z","etag":null,"topics":["adhd","ensemble","feature-engineering","machine-learning","preprocessing"],"latest_commit_sha":null,"homepage":"","language":"Jupyter Notebook","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/NiranjanRao07.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null}},"created_at":"2025-04-28T04:07:37.000Z","updated_at":"2025-04-30T02:13:24.000Z","dependencies_parsed_at":"2025-04-30T03:23:57.545Z","dependency_job_id":"9a282b92-ddaa-41d6-bf43-5eac8966e887","html_url":"https://github.com/NiranjanRao07/ADHD-ML-Project","commit_stats":null,"previous_names":["niranjanrao07/adhd-ml-project"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/NiranjanRao07%2FADHD-ML-Project","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/NiranjanRao07%2FADHD-ML-Project/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/NiranjanRao07%2FADHD-ML-Project/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/NiranjanRao07%2FADHD-ML-Project/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/NiranjanRao07","download_url":"https://codeload.github.com/NiranjanRao07/ADHD-ML-Project/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":254253937,"owners_count":22039792,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["adhd","ensemble","feature-engineering","machine-learning","preprocessing"],"created_at":"2025-05-15T01:14:52.122Z","updated_at":"2025-05-15T01:15:25.290Z","avatar_url":"https://github.com/NiranjanRao07.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# ADHD Detection from EEG Data\n\nThis repository contains all code, data, and artifacts for our project on automated ADHD detection using 19-channel EEG recordings and classical machine learning techniques.\n\n## 🔍 Overview\n\n1. **Data Exploration**\n   - MATLAB `.mat` EEG files (61 ADHD, 60 control), 19 channels, 128 Hz sampling\n   - Scripts compute channel-wise stats: mean, variance, skewness, kurtosis, ptp\n\n2. **Preprocessing**\n   - Butterworth band-pass filter (0.5–45 Hz) removes drift \u0026 noise\n   - Z-score normalization per channel (mean = 0, std = 1)\n   - Amplitude-based artifact removal (±100 µV threshold)\n\n3. **Feature Extraction**\n   - **Time-domain:** mean, std, skewness, kurtosis, RMS, zero crossings, peak-to-peak\n   - **Frequency-domain:** Welch PSD, delta/theta/alpha/beta/gamma band powers, spectral entropy, SEF, PSD slope\n   - **Non-linear:** approximate entropy, Higuchi fractal dimension, Hjorth mobility/complexity, Hurst exponent\n\n4. **Dimensionality Reduction**\n   - PCA (95 % variance) tested → suboptimal\n   - **Final:** supervised LDA applied only on training split → single discriminant axis\n\n5. **Modeling \u0026 Evaluation**\n   - Classifiers on LDA output: SVM, Decision Tree, Random Forest, KNN, Logistic Regression\n   - Hyperparameter tuning: grid search + 5-fold CV\n   - Held-out 20 % test split for final metrics\n   - Voting ensemble of all five models\n\n6. **Results**\n   - **Ensemble Test Performance:**\n     - Accuracy: 72.0 %\n     - Precision: 68.8 %\n     - Recall: 84.6 %\n     - F1-Score: 75.9 %\n     - ROC AUC: 78.8 %\n\nExplore raw data: run `analyze_files.py`\nPreprocess \u0026 extract features: open and execute `data_preprocess.ipynb` and `feature_extraction.ipynb`\nTrain \u0026 evaluate models: implement LDA on train split, train classifiers, build ensemble\n\n## 🤝 Collaboration \u0026 Version Control\n\nProgress and deliverables were tracked through regular team syncs and shared document updates.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fniranjanrao07%2Fadhd-ml-project","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fniranjanrao07%2Fadhd-ml-project","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fniranjanrao07%2Fadhd-ml-project/lists"}