Enhancing Automatic Placenta Analysis Through Distributional Feature Recomposition in Vision-Language Contrastive Learning

The placenta is a valuable organ that can aid in understanding adverse events during pregnancy and predicting issues post-birth. Manual pathological examination and report generation, however, are laborious and resource-intensive. Limitations in diagnostic accuracy and model efficiency have impeded previous attempts to automate placenta analysis. This study presents a novel framework for the automatic analysis of placenta images that aims to improve accuracy and efficiency. Building on previous vision-language contrastive learning (VLC) methods, we propose two enhancements, namely Pathology Report Feature Recomposition and Distributional Feature Recomposition, which increase representation robustness and mitigate feature suppression. In addition, we employ efficient neural networks as image encoders to achieve model compression and inference acceleration. Experiments validate that the proposed approach outperforms prior work in both performance and efficiency by significant margins. The benefits of our method, including enhanced efficacy and deployability, may have significant implications for reproductive healthcare, particularly in rural areas or low- and middle-income countries.

Conference paper from the 2023 International Conference on Medical Image Computing and Computer-Assisted Intervention

Files

  • pan.pdf

    size: 1.44 MB | mime_type: application/pdf | date: 2024-03-04 | sha256: 5eaea2f

Metadata

Work Title Enhancing Automatic Placenta Analysis Through Distributional Feature Recomposition in Vision-Language Contrastive Learning
Access
Open Access
Creators
  1. Yimu Pan
  2. Tongan Cai
  3. Manas Mehta
  4. Alison D. Gernand
  5. Jeffery A. Goldstein
  6. Leena Mithal
  7. Delia Mwinyelle
  8. Kelly Gallagher
  9. James Z. Wang
Keyword
  1. Placenta analysis
  2. Representation
  3. Vision-language
License In Copyright (Rights Reserved)
Work Type Conference Proceeding
Publisher
  1. Medical Image Computing and Computer Assisted Intervention – MICCAI 2023
Publication Date October 1, 2023
Publisher Identifier (DOI)
  1. https://doi.org/10.1007/978-3-031-43987-2_12
Deposited March 04, 2024

Versions

Analytics

Collections

This resource is currently not in any collection.

Work History

Version 1
published

  • Created
  • Added pan.pdf
  • Added Creator Yimu Pan
  • Added Creator Tongan Cai
  • Added Creator Manas Mehta
  • Added Creator Alison D. Gernand
  • Added Creator Jeffery A. Goldstein
  • Added Creator Leena Mithal
  • Added Creator Delia Mwinyelle
  • Added Creator Kelly Gallagher
  • Added Creator James Z. Wang
  • Published
  • Updated Keyword, Publisher, Description, and 1 more Show Changes
    Keyword
    • Placenta analysis, Representation, Vision-language
    Publisher
    • Medical Image Computing and Computer Assisted Intervention – MICCAI 2023
    Description
    • <p>The placenta is a valuable organ that can aid in understanding adverse events during pregnancy and predicting issues post-birth. Manual pathological examination and report generation, however, are laborious and resource-intensive. Limitations in diagnostic accuracy and model efficiency have impeded previous attempts to automate placenta analysis. This study presents a novel framework for the automatic analysis of placenta images that aims to improve accuracy and efficiency. Building on previous vision-language contrastive learning (VLC) methods, we propose two enhancements, namely Pathology Report Feature Recomposition and Distributional Feature Recomposition, which increase representation robustness and mitigate feature suppression. In addition, we employ efficient neural networks as image encoders to achieve model compression and inference acceleration. Experiments validate that the proposed approach outperforms prior work in both performance and efficiency by significant margins. The benefits of our method, including enhanced efficacy and deployability, may have significant implications for reproductive healthcare, particularly in rural areas or low- and middle-income countries.</p>
    • <p>The placenta is a valuable organ that can aid in understanding adverse events during pregnancy and predicting issues post-birth. Manual pathological examination and report generation, however, are laborious and resource-intensive. Limitations in diagnostic accuracy and model efficiency have impeded previous attempts to automate placenta analysis. This study presents a novel framework for the automatic analysis of placenta images that aims to improve accuracy and efficiency. Building on previous vision-language contrastive learning (VLC) methods, we propose two enhancements, namely Pathology Report Feature Recomposition and Distributional Feature Recomposition, which increase representation robustness and mitigate feature suppression. In addition, we employ efficient neural networks as image encoders to achieve model compression and inference acceleration. Experiments validate that the proposed approach outperforms prior work in both performance and efficiency by significant margins. The benefits of our method, including enhanced efficacy and deployability, may have significant implications for reproductive healthcare, particularly in rural areas or low- and middle-income countries.</p>
    • Conference paper from the 2023 International Conference on Medical Image Computing and Computer-Assisted Intervention
    Publication Date
    • 2023-01-01
    • 2023-10-01
  • Updated