Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Exploiting the Segment Anything Model (SAM) for Lung Segmentation in Chest X-ray Images

Gabriel Bellon de Carvalho, Jurandy Almeida

TL;DR

The approach adopted for this work, with the aim of improving the model's performance for lung segmentation, involved a transfer learning process, specifically the fine-tuning technique, and a substantial improvement was observed in the evaluation metrics used to assess SAM's performance compared to the masks provided by the datasets.

Abstract

Segment Anything Model (SAM), a new AI model from Meta AI released in April 2023, is an ambitious tool designed to identify and separate individual objects within a given image through semantic interpretation. The advanced capabilities of SAM are the result of its training with millions of images and masks, and a few days after its release, several researchers began testing the model on medical images to evaluate its performance in this domain. With this perspective in focus -- i.e., optimizing work in the healthcare field -- this work proposes the use of this new technology to evaluate and study chest X-ray images. The approach adopted for this work, with the aim of improving the model's performance for lung segmentation, involved a transfer learning process, specifically the fine-tuning technique. After applying this adjustment, a substantial improvement was observed in the evaluation metrics used to assess SAM's performance compared to the masks provided by the datasets. The results obtained by the model after the adjustments were satisfactory and similar to cutting-edge neural networks, such as U-Net.

Exploiting the Segment Anything Model (SAM) for Lung Segmentation in Chest X-ray Images

TL;DR

The approach adopted for this work, with the aim of improving the model's performance for lung segmentation, involved a transfer learning process, specifically the fine-tuning technique, and a substantial improvement was observed in the evaluation metrics used to assess SAM's performance compared to the masks provided by the datasets.

Abstract

Segment Anything Model (SAM), a new AI model from Meta AI released in April 2023, is an ambitious tool designed to identify and separate individual objects within a given image through semantic interpretation. The advanced capabilities of SAM are the result of its training with millions of images and masks, and a few days after its release, several researchers began testing the model on medical images to evaluate its performance in this domain. With this perspective in focus -- i.e., optimizing work in the healthcare field -- this work proposes the use of this new technology to evaluate and study chest X-ray images. The approach adopted for this work, with the aim of improving the model's performance for lung segmentation, involved a transfer learning process, specifically the fine-tuning technique. After applying this adjustment, a substantial improvement was observed in the evaluation metrics used to assess SAM's performance compared to the masks provided by the datasets. The results obtained by the model after the adjustments were satisfactory and similar to cutting-edge neural networks, such as U-Net.

Paper Structure

This paper contains 30 sections, 2 equations, 7 figures, 9 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Segment Anything Model (SAM)
  4. Model Development
  5. Components of SAM
  6. Image Encoder
  7. Prompt Encoder
  8. Mask Decoder
  9. Materials and Methods
  10. Datasets
  11. Montgomery dataset dados
  12. Shenzhen dataset dados
  13. Technical Approach
  14. Evaluation Metrics
  15. Intersection over Union (IoU)
  16. ...and 15 more sections

Figures (7)

  • Figure 1: Example of a given image and its corresponding masks after a panoptic segmentation. Adapted from Google Images.
  • Figure 2: Example of a given image with areas of interest selected by bounding boxes and its corresponding masks after an instance segmentation. Adapted from Google Images.
  • Figure 3: Organization of the Segment-Anything Model. Adapted from sam.
  • Figure 4: Example of a mean image obtained on the Montgomery dataset.
  • Figure 5: Example of a learning curve obtained on the Shenzhen dataset.
  • ...and 2 more figures