I’m a data scientist focused on digital cardiovascular research and applied ML.
My portfolio spans end-to-end ML pipelines, classic ML from scratch, unsupervised methods, deep learning on tabular data, computer vision transfer learning, and NLP.
Note: The repositories here are curated portfolio demonstrations (clean, reproducible projects). I also share reproducible paper-code repositories that run on publicly available datasets (with data excluded from the repo).
-
📈 PPG-based cf-PWV Estimation (IEEE Access) — Paper Code · feature extraction · correlation analysis · XGBoost regression · diagnostics
Repo: https://github.com/KianaAbrisham/ppg-cfpwv-ieee-access
Paper: https://ieeexplore.ieee.org/abstract/document/11218839 -
🫀 Stroke Prediction — ML Pipeline · impute/scale/one-hot · class imbalance · ROC/PR · permutation importance
https://github.com/KianaAbrisham/stroke-prediction-ml-pipeline -
📐 SVM from Scratch · hinge + L2 · decision boundary · LinearSVC comparison
https://github.com/KianaAbrisham/svm-from-scratch -
🧩 K-means (Breast Cancer) · silhouette · PCA visualization · label mapping
https://github.com/KianaAbrisham/kmeans-breast-cancer-portfolio -
🔥 PyTorch MLP (Tabular) · early stopping · accuracy/F1 · confusion matrix
https://github.com/KianaAbrisham/mlp-pytorch-classifier -
🖼️ CV — Transfer Learning (MobileNetV2) · augmentation · confusion matrix · inference demo
https://github.com/KianaAbrisham/cv-transfer-learning-mobilenetv2 -
🗣️ Naive Bayes (Sentiment) · sklearn pipeline + scratch implementation
https://github.com/KianaAbrisham/naive-bayes-sentiment
- Google Scholar: https://scholar.google.com/citations?user=69IoCyIAAAAJ&hl=en
- ORCID: https://orcid.org/0009-0007-4569-3196
Python · NumPy · pandas · scikit-learn · TensorFlow/Keras · PyTorch
Email: kianaabrisham@gmail.com