https://github.com/akimuddinshaikh/data-governance-2
An ethical AI approach for liver disease detection using machine learning and deep learning models. The project ensures data privacy, fairness, and transparency, using the Indian Liver Patient Records dataset for scalable healthcare diagnostics
https://github.com/akimuddinshaikh/data-governance-2
bias-mitigation data-privacy fairness transparency
Last synced: 4 months ago
JSON representation
An ethical AI approach for liver disease detection using machine learning and deep learning models. The project ensures data privacy, fairness, and transparency, using the Indian Liver Patient Records dataset for scalable healthcare diagnostics
- Host: GitHub
- URL: https://github.com/akimuddinshaikh/data-governance-2
- Owner: Akimuddinshaikh
- Created: 2025-02-04T03:03:08.000Z (over 1 year ago)
- Default Branch: main
- Last Pushed: 2025-02-04T03:06:04.000Z (over 1 year ago)
- Last Synced: 2025-12-21T22:12:20.943Z (5 months ago)
- Topics: bias-mitigation, data-privacy, fairness, transparency
- Homepage:
- Size: 69.3 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
Liver Disease Detection Using Machine Learning and Deep Learning
Ethical Considerations in AI-Based Healthcare Diagnostics
📌 Author: Akimuddin Aslam Shaikh
📍 Institution: National College of Ireland
📌 Project Overview
Liver disease remains a major health concern, affecting millions and burdening healthcare systems. This project explores the use of machine learning (ML) and deep learning (DL) techniques to enhance the accuracy and scalability of liver disease diagnosis.
The project uses the Indian Liver Patient Records dataset, implementing ML and DL models while ensuring that ethical considerations such as data privacy, bias mitigation, and transparency are maintained.
🔍 Key Highlights
✅ Machine Learning & Deep Learning applied to liver disease detection.
✅ Ethical focus on data privacy, fairness, and bias mitigation.
✅ Use of feature selection, preprocessing, and fairness-aware metrics.
✅ Secured data handling with encryption and access control.
✅ Open-source models for transparency while maintaining patient confidentiality.
📊 Ethical Considerations in AI-Based Liver Disease Detection
🔹 Before Research: Ethical Data Sourcing
✔ Anonymized dataset from Kaggle to protect patient privacy.
✔ Bias analysis performed to check if the dataset is demographically skewed.
✔ Exploratory Data Analysis (EDA) conducted to identify and mitigate imbalances in data representation.
🔹 During Research: Ethical Model Development
Ethical Concern Strategy Implemented
Data Privacy & Security Anonymization, secure storage, encryption, and restricted access.
Bias Mitigation Ensured balanced training & test datasets to avoid model bias.
Algorithm Transparency All data preprocessing, feature selection, and model training steps fully documented.
Accountability Peer-reviewed methodologies ensure responsible AI practices.
🔹 After Research: Societal & Ethical Impact
✔ Open-source models for transparency while securing sensitive patient data.
✔ Ethical AI ensures fairness, privacy, and unbiased decision-making.
✔ Potential to integrate ML/DL-based tools into real-world clinical practice for early liver disease detection.
✔ Supports healthcare policymakers in deploying AI-driven diagnostic tools in underprivileged regions.
🔹 Key Takeaway: AI-powered healthcare solutions must balance innovation with ethical responsibility, ensuring fairness, security, and regulatory compliance.
📌 Future Recommendations
✔ Enhance dataset diversity to improve model generalization.
✔ Implement explainable AI (XAI) techniques to improve model interpretability.
✔ Apply federated learning to ensure privacy-preserving AI in healthcare.
✔ Conduct fairness audits to assess ethical risks before clinical deployment.