Upwork home
Sign up
Project Catalog
DesignDesign
Video & AudioVideo & Audio
Development & ITDevelopment & IT
MarketingMarketing
Writing & TranslationWriting & Translation
MoreMore

You will get Medical Image Segmentation Pipeline with MONAI and PyTorch

Houssem E.Status: Offline
Houssem E.
Play video
Preview image

Let a pro handle the details

Buy Machine Learning services from Houssem, priced and ready to go.
Houssem E.Status: Offline
Houssem E.
Play video
Preview image

Let a pro handle the details

Buy Machine Learning services from Houssem, priced and ready to go.

Project details

I build production-grade medical image segmentation pipelines using MONAI and PyTorch, the same stack used in regulated clinical environments. My last deployment achieved Dice scores of 0.88–0.91 for cardiac MRI segmentation (LV, RV, myocardium) with 95%+ clinical acceptance in a blinded study with senior cardiologists, delivered under EU MDR/IVDR documentation standards.
What you get is a complete, deployment-ready segmentation system. Deliverables include a trained model on your imaging data, ONNX export for cross-platform deployment, GradCAM explainability maps, validation metrics (Dice, HD95), and full technical documentation. Source code included.
I work with DICOM, NIfTI, and standard image formats, and can adapt to cardiac, brain, abdominal, or other anatomical structures depending on your dataset.
Machine Learning Tools
Databricks Platform, Databricks MLflow, GitHub Copilot, Google Data Studio, Keras, MLflow, Open Neural Network Exchange, OpenCV, pandas, Python, Python Scikit-Learn, PyTorch, scikit-learn, TensorFlow, Vertex AI, XGBoost

What's included $600

These options are included with the project scope.

$600
  • Delivery Time 21 days
  • Number of Revisions 2
  • Number of Model Variations 1
  • Number of Scenarios 2
  • Number of Graphs/Charts 3
    • Model Validation/Testing
    • Model Documentation
    • Data Source Connectivity
    • Source Code

Frequently asked questions

Houssem E.Status: Offline

About Houssem

Houssem E.Status: Offline
Medical Imaging, AI Engineer, YOLOv8, MONAI, Production System
Ariana, Tunisia - 1:37 pm local time
I build production-grade computer vision and medical imaging systems — not prototypes. Recent work includes a cardiac MRI segmentation pipeline (Dice 0.88–0.91, 95%+ clinical acceptance) and an industrial AOI system that reduced manufacturing defect rates from 6.8% to 0.2%.

Steps for completing your project

After purchasing the project, send requirements so Houssem can start the project.

Delivery time starts when Houssem receives requirements from you.

Houssem works on your project following the steps below.

Revisions may occur after the delivery date.

Data review and preprocessing

Review imaging data, apply normalization, voxel spacing, and augmentation

Model training and validation

Train segmentation model, evaluate Dice and HD95, iterate to target

Review the work, release payment, and leave feedback to Houssem.