Product Documentation
HomepageLoginSign up
  • Welcome to Biodock
  • Why Biodock
  • Quickstart
    • Start here (choose a tutorial)
    • Path 1: AI-assisted analysis
    • Path 2: Train a fully-automated AI model
      • Set up your project
      • Label and train
      • Labeling shortcuts
      • Run your AI model
      • Configuring your model
  • AI Projects
    • Project settings
    • Sharing your AI project
    • Training a model
  • AI Analysis
    • Results dashboard overview
    • Download results and reports
    • Correcting results to improve your model
    • Editing objects (QC)
    • Filters (like flow cytometry)
  • Biodock Scripts
    • Biodock Platform Script Guidelines
  • Files
    • Uploading images
    • AWS S3 integration
    • View and manage data
      • Viewing images
      • File details and download
      • Copy, cut, move
      • Sharing data
      • Merging channels
    • Supported image types
  • Public API (Beta)
    • Overview
    • Authentication
    • Examples
    • Resources
      • Files
      • Analysis Jobs
      • Pipelines
    • Limitations
  • Deep AI models
    • Evaluating Performance
  • User
    • Account registration
    • Change password, login, logout
    • Usage limits and team
      • Filesize based limits
      • Run credits limits
    • Billing
  • Company
    • Academics and startups
    • Contact us
    • Citing Biodock
    • Security and IP
Powered by GitBook
On this page
  • Motivation
  • Importing labels
  • Correcting and retraining
  1. AI Analysis

Correcting results to improve your model

PreviousDownload results and reportsNextEditing objects (QC)

Last updated 2 years ago

Motivation

As you acquire different types of images, from different cell lines, procedures, or other variation, you may notice that for some new images, your results may not be 100% accurate. This is often the case when your initial training set didn't capture the variation seen within the images that you are running.

Biodock provides an easy way to take these results, import them as labels into your project, and only correct labels that may be missing or incorrect to improve performance in your model.

Importing labels

You can find the Add to AI Project button right above the image viewer. Click the button to start importing predictions for this option.

A modal will appear. If you've changed your class names, you may need to respecify which classes will be mapped to which labeling classes. In most cases, you'll want to import them back to the same project.

Correcting and retraining

Once you click the blue button in the bottom right, the modal will close, and you'll soon get a notification in platform that the import succeeded. If you go back to the original project, you'll be able to open the imported image:

You should correct one or more of these imported tiles from this image, especially focusing on areas where the model underperformed. Make sure that every object is labeled correctly in the tile. Then you can mark these as fully-labeled by clicking on the tile checkbox.

Finally, once you've imported enough images and corrected more tiles, you should train another version! The new version should perform much better on images that look like the ones you corrected.