Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Improve hourglass performances #28

Open
NathanMolinier opened this issue Sep 19, 2023 · 4 comments
Open

Improve hourglass performances #28

NathanMolinier opened this issue Sep 19, 2023 · 4 comments

Comments

@NathanMolinier
Copy link
Collaborator

Description

This issue is meant to discuss about strategies to improve the performances of the hourglass network.

@NathanMolinier
Copy link
Collaborator Author

Strategy 1: Improve the loss function

Problem

Since the addition of new images (representing different field of view of the spine) performances of the hourglass have dropped significantly (mainly due to the addition of new classes ~25 discs) leading to an increasing number of false positive detections.

This problem was partially solved here but still numerous false positive detection are still occurring.

Idea

Split the loss function into two main components:

  • an accuracy loss (MSE) for masks where a disc should be predicted
  • a true detection loss (MSE) to ensure that non present discs will lead to an empty prediction mask

@NathanMolinier
Copy link
Collaborator Author

Strategy 2: Use random crop with larger fields of view

Idea

To improve the performances and help the model to generalize, a new data augmentation function random_crop was added to randomly vary the discs that are present in the image.

Results

The task still is too difficult for the model, the number of different classes is maybe to important.

@NathanMolinier
Copy link
Collaborator Author

NathanMolinier commented Feb 1, 2024

Strategy 3: Train 2 different models

Idea

Finding and identifying all the 25 discs classes seems to be to difficult for one model, therefore, I will try to train 2 models:

  • one able to identify all the discs present in the image (Training 1)
  • one able to assign classes to these detected discs (Training 2)

These models will be train with and without the random cropping to evaluate their ability to generalize.

Results

...

@NathanMolinier
Copy link
Collaborator Author

NathanMolinier commented Feb 7, 2024

Strategy 4: Fixed window size with BCE loss

Idea

based on https://www.sciencedirect.com/science/article/pii/S1361841521003030

Retry with the 25 classes but this time with a fixed size sliding window.

The use of the Binary Cross Entropy will also be investigated to avoid having multiple classes for the same disc.

Results

...

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant