[fs-toc-h2]Getting Started
Welcome to the Project OME Beta V1.0 tutorial! This guide will help you navigate and make the most of the cutting-edge features introduced in this release. Let’s get started.
Accessing Project OME
1. Sign In:
Use the credentials provided during the beta signup to log in to the Project OME platform
2. Upload Medical Images:
Click the button to start and upload your CT scans via the intuitive upload interface. If you'd like to try it out, anonymized sample datasets are available at this link.
3. Generate 3D Models:
Click the "Generate" button to add your generation task to the automatic pipeline.
4. Processing:
Generating a 3D model may take some time, depending on the request’s complexity and current workload. Please click the Refresh button in a few minutes to check the status.
5. Visualization:
View your 3D models in various formats, including default texture, wireframe, and pathology modes.
[fs-toc-h2]Feature Guide
1. AI-Generated 3D Models
1-1. Creating 3D Models
- Navigate to the "Modeling and Simulation" tab on the left panel.
- Upload your CT scans in ZIP format. (Current beta version supports CT modality only; MRI support is coming soon.)
- Click "Generate Model" and check the status of your task in the "My Pipelines" section. Use the refresh button above the session column to update the task status.
1-2. Exploring Models in My Creations
- Once the session status shows "Success," go to "My Creations" on the left panel.
- Locate your uploaded datasets organized in folders. Each main folder is named after your uploaded file.
- Inside each folder:
- DICOM Subfolder: Contains your original dataset.
- Default Model Subfolder: Houses your generated 3D model.
- Use the "View" button to visualize the models in the 3D viewer or download, rename, or delete them using the respective icons.
1-3. Visualization Portal
Model Selection: Use the dropdown list in the top-left corner to select specific anatomical models or display all 3D models at once.
1-4. Control Panel:
Manipulate the 3D models with these tools:
- Home Icon: Reset the model view to its default layout.
- Zoom Icons: Zoom in or out of the model.
- Hand Icon: Pan the model by holding and dragging with the mouse.
1-5. Texture Panel:
Toggle between different visualization modes:
- Default View: View the model with the standard texture.
- Wireframe View: Examine the polygonal structure of the model.
- Pathology View: Display simulated textures for clinical conditions (if previously generated).
1-6. Predictive ROIs:
Estimate regions of anatomical structures not available in the original dataset:
- Circulatory Icon: Predict nearby arteries and veins.
- Nerve Icon: Predict nearby nerves.
1-7. Model Flashcards
Click on individual models to view terminology and educational content with the automatic labeling feature located in the top-right corner.
2. Medical Chatbot OME
2-1. Using the Chatbot
- Access the chatbot via the "OME Assistant" located in the bottom-right corner.
- Interact with the chatbot powered by Google MedLM:
- Ask questions like, "What are the anatomical landmarks of the liver?"
- Receive detailed and expert-level insights instantly.
3. Texture Simulation for Pathology
3-1. Simulating Pathological Conditions
- Navigate to the "Texture Simulation" section.
- Select a clinical condition (e.g., Costochondritis) from the dropdown menu based on your body region.
- Click "Generate" to start the simulation task, then monitor its progress in the "My Pipelines" section.
- After the task is complete:
- Locate the new subfolder named after the simulated condition within your dataset folder.
- Use the "View" button to visualize the pathology texture or switch back to the default model in the visualization portal.
[fs-toc-h2]Thank You!
Thank you for participating in the Project OME Beta V1.0! Your feedback is invaluable as we refine and enhance this innovative tool. We look forward to seeing how you use Project OME to revolutionize medical education and practice.
Contact us at support@mai.ai for assistance.
