Ā Ā
Ā Ā
Ā Ā
Ā Ā
Iām a Data Science and Machine Learning Enthusiast with a unique background that bridges:
- ā” Electronics Engineering
- š Applied Statistics
I bring experience from diverse, high-impact industries ā including:
- āļø High-End Electronics Manufacturing
- š³ Fintech
- š§¬ Biotechnology R&D
My work spans across domains where Data-Driven Insights, Engineering Precision, and Statistical Rigor intersect to solve real-world problems.
I'm passionate about Lifelong Learning and creating Innovative, End-to-End Projects that:
- š¤ Harness the power of data and machine learning
- š ļø Deliver practical, real-world solutions
- š Contribute to open knowledge
In my spare time, youāll find me experimenting with cutting-edge ML tools and applying them to meaningful use cases.
Iāve briefly organized my work below ā including:
- Machine Learning Applications: End-to-end predictive modeling projects showcasing complete pipelines from data to deployment with live interactive web apps.
- Machine Learning Case Studies: In-depth, statistically rigorous studies that emphasize exploratory modeling, validation, and interpretability.
- Machine Learning Exploratory Projects: Targeted experiments exploring specific machine learning lifecycle components, techniques, or technologies in isolation.
- Visual Analytics Projects: Interactive dashboards designed for intuitive data exploration and visual storytelling.
- Scientific Research Papers: Peer-reviewed co-authored research contributions applying data science methods in academic and scientific contexts.
- šµ Completed Projects
- š“ Ongoing Work
For a deeper dive into my projects, methodologies, and tools used, check out my š Project Portfolio Website
Thanks for visiting my GitHub! š»
Feel free to š explore, š¤ connect, or š„ collaborate.
| Tools | Project Title | Status | Link |
|---|---|---|---|
| Classifying Brain Tumors from Magnetic Resonance Images by Leveraging Convolutional Neural Network-Based Multilevel Feature Extraction and Hierarchical Representation | šµ | Notebook Repository Application |
|
| Estimating Heart Failure Survival Risk Profiles From Cardiovascular, Hematologic And Metabolic Markers | šµ | Notebook Repository Application |
|
| Estimating Lung Cancer Probabilities From Demographic Factors, Clinical Symptoms And Behavioral Indicators | šµ | Notebook Repository Application |
| Tools | Project Title | Status | Link |
|---|---|---|---|
| Learning Hierarchical Features for Predicting Multiclass X-Ray Images using Convolutional Neural Network Model Variations | šµ | Notebook Repository |
|
| Discovering Global Patterns in Cancer Mortality Across Countries Via Clustering Analysis | šµ | Notebook Repository |
|
| Identifying Contributing Factors for Countries With High Cancer Rates Using Classification Algorithms With Class Imbalance Treatment | šµ | Notebook Repository |
|
| Uncovering Underlying Constructs of Chronic Disease Indicators Across US States Using Exploratory and Confirmatory Factor Analyses | šµ | Notebook Repository |
|
| Characterizing Life Expectancy Drivers Across Countries Using Model-Agnostic Interpretation Methods for Black-Box Models | šµ | Notebook Repository |
| Tools | Project Title | Status | Link |
|---|---|---|---|
| Detecting and Analyzing Machine Learning Model Drift Using Open-Source Monitoring Tools | šµ | Notebook Repository |
|
| Machine Learning Model Experiment Logging and Tracking Using Open-Source Frameworks | šµ | Notebook Repository |
|
| Detecting and Evaluating Anomalies in Categorical Data Under Supervised and Unsupervised Settings | šµ | Notebook Repository |
|
| Containerizing and Deploying Machine Learning API Endpoints on Open-Source Platforms | šµ | Notebook Repository |
|
| Leveraging Ensemble Learning With Bagging, Boosting, Stacking and Blending Approaches | šµ | Notebook Repository |
|
| Exploring Modular Application Programming Interface Frameworks For Serving Model Predictions | šµ | Notebook Repository |
|
| Exploring Parametric Accelerated Failure Time Models for Estimating Lifetimes in Survival Data | šµ | Notebook Repository |
|
| Implementing Shapley Additive Explanations for Interpreting Feature Contributions in Penalized Cox Regression | šµ | Notebook Repository |
|
| Modelling Right-Censored Survival Time and Status Responses for Prediction | šµ | Notebook Repository |
|
| Exploring Regularization Approaches for Controlling Model Complexity Through Weight Penalization for Neural Network Classification | šµ | Notebook Repository |
|
| Comparing Optimization Algorithms in Parameter Updates and Loss Function Minimization for Neural Network Classification | šµ | Notebook Repository |
|
| Exploring Activation Functions And Backpropagation Gradient Updates for Neural Network Classification | šµ | Notebook Repository |
|
| Comparing Batch, Stochastic and Mini-Batch Approaches to Gradient Descent in Estimating Regression Coefficients | šµ | Notebook Repository |
|
| Implementing Backpropagation In Updating Weights for Neural Network Classification | šµ | Notebook Repository |
|
| Implementing Gradient Descent Algorithm in Estimating Regression Coefficients | šµ | Notebook Repository |
|
| Data Quality Assessment, Preprocessing and Exploration for a Classification Modelling Problem | šµ | Notebook Repository |
|
| Exploring Penalized Models for Predicting Numeric Responses | šµ | Notebook Repository |
|
| Data Quality Assessment, Preprocessing and Exploration for a Regression Modelling Problem | šµ | Notebook Repository |
|
| Exploring Boosting, Bagging and Stacking Algorithms for Ensemble Learning | šµ | Notebook Repository |
|
| Discovering Latent Variables in High-Dimensional Data using Exploratory Factor Analysis | šµ | Notebook Repository |
|
| Exploring Penalized Models for Handling High-Dimensional Survival Data | šµ | Notebook Repository |
|
| Exploring and Visualizing Extracted Dimensions from Principal Component Algorithms | šµ | Notebook Repository |
|
| Sample Size and Power Calculations for Tests Comparing Proportions in Clinical Research | šµ | Notebook Repository |
|
| Sample Size and Power Calculations for Tests Comparing Means in Clinical Research | šµ | Notebook Repository |
|
| Comparing Oversampling and Undersampling Algorithms for Class Imbalance Treatment | šµ | Notebook Repository |
|
| Exploring Performance Evaluation Metrics for Survival Prediction | šµ | Notebook Repository |
|
| Exploring Robust Logistic Regression Models for Handling Quasi-Complete Separation | šµ | Notebook Repository |
|
| Estimating Outlier Scores Using Density and Distance-Based Anomaly Detection Algorithms | šµ | Notebook Repository |
|
| Estimating Outlier Scores Using Isolation Forest-Based Anomaly Detection Algorithms | šµ | Notebook Repository |
|
| Identifying Multivariate Outliers Using Density-Based Clustering Algorithms | šµ | Notebook Repository |
|
| Exploring Dichotomization Thresholding Strategies for Optimal Classification | šµ | Notebook Repository |
|
| Implementing Gradient Descent Algorithm in Estimating Regression Coefficients | šµ | Notebook Repository |
|
| Formulating Segmented Groups Using Clustering Algorithms | šµ | Notebook Repository |
|
| Extracting Information Using Dimensionality Reduction Algorithms | šµ | Notebook Repository |
|
| Remedial Procedures for Skewed Data with Extreme Outliers | šµ | Notebook Repository |
|
| Selecting Informative Predictors Using Simulated Annealing and Genetic Algorithms | šµ | Notebook Repository |
|
| Selecting Informative Predictors Using Univariate Filters | šµ | Notebook Repository |
|
| Selecting Informative Predictors Using Recursive Feature Elimination | šµ | Notebook Repository |
|
| Evaluating Model-Independent Feature Importance for Predictors with Dichotomous Categorical Responses | šµ | Notebook Repository |
|
| Evaluating Model-Independent Feature Importance for Predictors with Numeric Responses | šµ | Notebook Repository |
|
| Cost-Sensitive Learning for Severe Class Imbalance | šµ | Notebook Repository |
|
| Remedial Procedures in Handling Imbalanced Data for Classification | šµ | Notebook Repository |
|
| Evaluating Hyperparameter Tuning Strategies and Resampling Distributions | šµ | Notebook Repository |
|
| Modelling Multiclass Categorical Responses for Prediction | šµ | Notebook Repository |
|
| Modelling Dichotomous Categorical Responses for Prediction | šµ | Notebook Repository |
|
| Modelling Numeric Responses for Prediction | šµ | Notebook Repository |
|
| Resampling Procedures for Model Hyperparameter Tuning and Internal Validation | šµ | Notebook Repository |
|
| Clinical Research Prediction Model Development and Evaluation for Prognosis | šµ | Notebook Repository |
|
| Missing Data Pattern Analysis, Imputation Method Evaluation and Post-Imputation Diagnostics | šµ | Notebook Repository |
|
| Survival Analysis and Descriptive Modelling for a Three-Group Right-Censored Data with Time-Independent Variables Using Cox Proportional Hazards Model | šµ | Notebook Repository |
|
| Survival Analysis and Descriptive Modelling for a Two-Group Right-Censored Data with Time-Independent Variables Using Cox Proportional Hazards Model | šµ | Notebook Repository |
|
| Treatment Comparison Tests Between a Single Two-Level Factor Variable and a Single Numeric Response Variable | šµ | Notebook Repository |
|
| Data Quality Assessment, Preprocessing and Exploration for a Regression Modelling Problem | šµ | Notebook Repository |
|
| Data Quality Assessment, Preprocessing and Exploration for a Classification Modelling Problem | šµ | Notebook Repository |
| Tools | Project Title | Status | Link |
|---|---|---|---|
| Dashboard Development with Slice-and-Dice Exploration Features | šµ | Dashboard | |
| Dashboard Development with Dynamic Filtering Features | šµ | Dashboard | |
| Dashboard Development with Longitudinal Change Tracking Features | šµ | Dashboard | |
| Dashboard Development with What-If Scenario Analysis Features | šµ | Dashboard | |
| Dashboard Development with Period-To-Date Performance Tracking Features | šµ | Dashboard |
| Tools | Project Title | Status | Link |
|---|---|---|---|
| Genomic Imprinting Biomarkers for Cervical Cancer Risk Stratification | šµ | Publication | |
| High Diagnostic Accuracy of Epigenetic Imprinting Biomarkers in Thyroid Nodules | šµ | Publication | |
| Epigenetic Imprinting Alterations as Effective Diagnostic Biomarkers for Early-Stage Lung Cancer and Small Pulmonary Nodules | šµ | Publication | |
| Novel Visualized Quantitative Epigenetic Imprinted Gene Biomarkers Diagnose the Malignancy of Ten Cancer Types | šµ | Publication | |
| New Thyroid Imaging Reporting and Data System (TIRADS) Based on Ultrasonography Features for Follicular Thyroid Neoplasms: A Multicenter Study | šµ | Publication | |
| Advancing Malignancy Risk Stratification for Early-Stage Cancers in Lung Nodules by Combined Imaging and Electrical Impedance Analysis | šµ | Abstract | |
| Intronic Noncoding RNA Expression of DCN is Related to Cancer-Associated Fibroblasts and NSCLC Patientsā Prognosis | šµ | Abstract | |
| Epigenetic Imprinted Genes as Biomarkers for the Proactive Detection and Accurate Presurgical Diagnosis of Small Lung Nodules | šµ | Abstract | |
| Effect of Epigenetic Imprinting Biomarkers in Urine Exfoliated Cells (UEC) on the Diagnostic Accuracy of Low-Grade Bladder Cancer | šµ | Abstract | |
| Epigenetic Imprinted Gene Biomarkers Significantly Improve the Accuracy of Presurgical Bronchoscopy Diagnosis of Lung Cancer | šµ | Abstract | |
| Quantitative Chromogenic Imprinted Gene In Situ Hybridization (QCIGISH) Technique Could Diagnose Lung Cancer Accurately | šµ | Abstract |