Skip to content

About

No description, website, or topics provided.

Resources

Stars

0 stars

Watchers

0 watching

Forks

Latest commit

 

History

7 Commits

Folders and files

NameName
Last commit message
Last commit date
 
 

Repository files navigation

Hi, I’m Aravind Kumar 👋

Data Analyst | Research Data Specialist | AI/ML & Bio-Image Analytics

📍 Louisville, KY | Open to Data Analyst, AI Analyst, Computer Vision, and Risk Analytics roles


About Me

I am a Data Analyst and Research Data Specialist with a Master’s in Business Analytics and 7+ years of experience across healthcare research, risk analytics, fraud detection, and business intelligence.

My work focuses on transforming complex data into actionable insights through automated pipelines, advanced SQL analytics, Python-based workflows, dashboard development, and AI/ML-driven image analysis. I have worked with high-volume transactional data at Amazon and biomedical imaging data at UofL Health, where I support research teams through data processing, segmentation, validation, reporting, and reproducible analytics workflows.

I enjoy building end-to-end solutions that connect raw data, automation, model development, deployment-ready workflows, monitoring, and stakeholder-facing insights.


Technical Skills

Programming & Analytics: Python, SQL, Pandas, NumPy, Jupyter Notebook Business Intelligence: Tableau, Power BI, Amazon QuickSight Data Engineering: ETL Pipelines, Data Cleaning, Data Validation, Workflow Automation AI/ML & Computer Vision: Deep Learning, CNNs, 3D Image Segmentation, Transfer Learning, Model Evaluation Bio-Image Analysis: FIJI/ImageJ, Ilastik, Cellpose, StarDist, OME-TIFF, 3D/5D Microscopy Data Cloud & Tools: AWS, Git, GitHub, SSMS, Excel, MATLAB Domain Expertise: Healthcare Research, Fraud Detection, Risk Analytics, Process Optimization


Featured Project

Deep Learning-Based 3D Microscopy Image Segmentation Pipeline

Developed an end-to-end computer vision workflow for automated segmentation of volumetric biomedical microscopy data using deep learning and Python-based image processing.

This project focuses on processing 3D/5D microscopy datasets, generating ground-truth annotations, training segmentation models, evaluating model performance, and creating deployment-ready inference workflows for research use.

Key Contributions

  • Built a 3D deep learning segmentation pipeline for volumetric microscopy image data.
  • Processed multi-dimensional OME-TIFF and TIFF imaging datasets using Python-based workflows.
  • Designed preprocessing steps including filtering, contrast enhancement, noise reduction, intensity normalization, and voxel-aware scaling.
  • Generated and validated ground-truth masks for supervised model training.
  • Trained and optimized deep learning-based segmentation models for cellular structure detection.
  • Evaluated model performance using IoU, Dice coefficient, precision, recall, and visual quality checks.
  • Automated preprocessing and inference steps to reduce manual workload and improve reproducibility.
  • Created deployment-ready scripts for running trained models on new microscopy datasets.
  • Added monitoring outputs such as prediction summaries, segmentation quality checks, model performance metrics, and review-ready visualizations.
  • Collaborated with research scientists to compare model outputs against biological ground truth and improve experimental reliability.

Deployment & Monitoring Focus

I designed the workflow to support reproducible model deployment and performance monitoring across new datasets.

The pipeline includes:

  • Standardized input preparation for raw microscopy files
  • Automated preprocessing before model inference
  • Model inference on new image volumes
  • Segmentation output generation and visualization
  • Quality-control checks for prediction accuracy
  • Performance tracking using segmentation metrics
  • Review-ready outputs for researchers and stakeholders

This approach helps ensure that the model is not only trained successfully but can also be reused, evaluated, and monitored consistently when applied to future biomedical imaging data.


What I’m Working On

  • Improving deep learning workflows for biomedical image segmentation
  • Building reproducible AI/ML pipelines for research data
  • Developing dashboards and analytics tools for scientific and business decision-making
  • Expanding skills in MLOps, cloud deployment, and model monitoring

Contact

I am open to opportunities in Data Analytics, AI/ML, Computer Vision, Healthcare Analytics, and Risk Analytics.

📫 Connect with me on LinkedIn : www.linkedin.com/in/aravindkumar2803/ 📧 Email me : Aravindkumarmadavaram@gmail.com

About

No description, website, or topics provided.

Resources

Stars

0 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors