Biomedical ML Bootcamp
Completed Fall 2023Biomedical Data Analytics & Machine Learning Bootcamp - Stony Brook Medicine & CEAS Department intensive program
Program Overview
Intensive bootcamp by Stony Brook Medicine and CEAS Department in Fall 2023, focusing on machine learning for biomedical data analytics. Hands-on experience with supervised and unsupervised learning using real biomedical datasets.
Technical Focus
Master various ML approaches and apply them to real biomedical datasets with clinical relevance for effective data analysis and model optimization.
Technical Implementation
Applied various ML techniques to the Iris dataset:
Machine Learning Techniques
- K-Means clustering: Unsupervised classification with silhouette score evaluation
- SelectKBest feature selection: Optimized model efficiency through relevant feature selection
- Random Forest and Decision Trees: Classification with ensemble methods
- Neural Networks: PyTorch models with optimized layer depth and node count
Advanced Techniques
- OpenCV Edge Detection: Image processing with custom convolution kernels
- Optuna Hyperparameter Tuning: Automated model optimization
- Model Evaluation: Performance assessment using multiple metrics
- Data Preprocessing: Feature scaling, normalization, and cleaning
Technology Stack
Core ML Libraries
- PyTorch neural networks
- Scikit-learn ML algorithms
- Pandas data manipulation
- NumPy numerical computing
Specialized Tools
- OpenCV computer vision
- Optuna hyperparameter optimization
- Seaborn statistical visualization
- Matplotlib custom plotting
Technical Challenges
Neural Network Architecture Design
Optimized network architectures by balancing layer depth, node count, and activation functions to prevent overfitting while maintaining generalization.
Parameter Selection Efficiency
Overcame large hyperparameter search space challenges using Optuna's automated algorithms, reducing training time while improving performance.
Future Applications
Seeking opportunities to apply these techniques to real-world datasets and industry applications at the intersection of biomedical engineering and AI/ML.
Medical Imaging
Computer vision for medical imaging analysis and diagnostics.
Clinical Data Analysis
ML models for clinical research predictive analytics and patient outcome modeling.
Biomedical Automation
ML integration into biomedical devices and automated diagnostic systems.
Professional Development Impact
Earned formal credentials in ML and biomedical data analytics from Stony Brook Medicine. Demonstrated commitment to continuous learning and established foundation for applying AI/ML to engineering challenges at the intersection of mechanical engineering and intelligent systems.