I work at the intersection of AI, medical imaging, biosignal processing, and biomedical data science.
My goal is to bridge research-grade machine learning with real-world clinical deployment, building systems that can genuinely improve diagnostics and patient monitoring.

• AI for Healthcare & Biomedicine
• Medical Imaging (CT/X-ray/MRI) | CNNs | Vision Transformers
• Biomarker Discovery (Gene Expression / Glioblastoma)
• ECG Analysis & Wearable Biosignal Intelligence
• Multimodal Clinical AI (Images + Signals + Metadata)
• Explainable AI (Grad-CAM, SHAP, saliency methods)

Machine Learning & Deep Learning:
Python • PyTorch • TensorFlow/Keras • scikit-learn • OpenCV

Biomedical Data Processing:
ECG filtering (FFT, CWT), MRI/CT preprocessing, DICOM, segmentation workflows, PCA/feature selection, gene expression normalization

Deployment & Engineering:
FastAPI • Docker • ONNX • TensorRT • Flask • Streamlit
Edge/IoT computing for wearable and medical sensor data • Mobile inference

Tools & Workflows:
Git • MLOps basics • Model interpretability • Visualization pipelines (Grad-CAM, ROC, UMAP, t-SNE)

• ECG Arrhythmia Classification with 1D CNNs – Research paper (in progress for journal submission)
• Machine learning workflows on Glioblastoma biomarker discovery
• Research experience in medical image-based tumor detection
• Active applicant for PhD programs in AI for Healthcare / Biomedical Engineering

I create two-layer AI solutions:

1. Academic-grade models:

   • Reproducible ML pipelines
   • Experiments, evaluations, ablations
   • Explainability & interpretability
   • Comparison with baselines

2. Industry-grade engineering versions:

   • REST APIs (FastAPI)
   • Dashboards & visualization apps (Streamlit)
   • ONNX/TensorRT for deployment
   • Edge AI for medical IoT devices

This hybrid approach is my identity: research depth + engineering practicality.

Tumor detection, segmentation, lesion classification, contrast enhancement
Tech: CNNs, U-Net, Grad-CAM, radiology preprocessing
Data: CT, MRI, X-ray (DICOM, NIfTI)

Gene expression–based classification and feature selection
Tech: PCA, SVM/KNN feature ranking, statistical testing
Data: TCGA, GEO datasets

ECG arrhythmia detection, real-time event detection for IoT
Tech: 1D CNNs, RNNs, LSTMs, wavelet transforms
Data: MIT-BIH, wearable sensor datasets

• Building multimodal clinical AI models (Imaging + Signal + Tabular)
• Transforming research models into deployable medical software prototypes
• Applying for PhD positions in top labs working on AI & Healthcare
• Developing a strong GitHub portfolio—research rigor + engineering execution

• Email: saeed.amiinii@gmail.com
• LinkedIn: www.linkedin.com/in/saeidamiinii
• GitHub: You’re already here—explore the pinned projects below 👇