Calibrate the Inter-Observer Segmentation Uncertainty via Diagnosis-First Principle
- 26 April 2024
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Imaging
- Vol. 43 (9) , 3331-3342
Abstract
Many of the tissues/lesions in the medical images may be ambiguous. Therefore, medical segmentation is typically annotated by a group of clinical experts to mitigate personal bias. A common solution to fuse different annotations is the majority vote, e.g., taking the average of multiple labels. However, such a strategy ignores the difference between the grader expertness. Inspired by the observation that medical image segmentation is usually used to assist the disease diagnosis in clinical practice, we propose the diagnosis-first principle, which is to take disease diagnosis as the criterion to calibrate the inter-observer segmentation uncertainty. Following this idea, a framework named Diagnosis-First segmentation Framework (DiFF) is proposed. Specifically, DiFF will first learn to fuse the multi-rater segmentation labels to a single ground-truth which could maximize the disease diagnosis performance. We dubbed the fused ground-truth as Diagnosis-First Ground-truth (DF-GT). Then, the Take and Give Model (T&G Model) to segment DF-GT from the raw image is proposed. With the T&G Model, DiFF can learn the segmentation with the calibrated uncertainty that facilitate the disease diagnosis. We verify the effectiveness of DiFF on three different medical segmentation tasks: optic-disc/optic-cup (OD/OC) segmentation on fundus images, thyroid nodule segmentation on ultrasound images, and skin lesion segmentation on dermoscopic images. Experimental results show that the proposed DiFF can effectively calibrate the segmentation uncertainty, and thus significantly facilitate the corresponding disease diagnosis, which outperforms previous state-of-the-art multi-rater learning methods.Keywords
All Related Versions
Funding Information
- National Natural Science Foundation of China (82121003, 62072190)
- Beijing Natural Science Foundation (Z190023)
- Program for Guangdong Introducing Innovative, Entrepreneurial Teams (2017ZT07X183)
- Guangdong Provincial Key Laboratory of Human Digital Twin (2022B1212010004)
- Scientific Research Project of National University of Defense Technology (22-ZZCX-07)
- Hefei Comprehensive National Science Center (KY23C503)
- Jianghuai Advance Technology Center (00QK0021)
- Ministry of Education Tier 1 Start-Up Grant, NUS, Singapore (WBS: A-8001267-01-00)
This publication has 50 references indexed in Scilit:
- TNSNet: Thyroid nodule segmentation in ultrasound imaging using soft shape supervisionComputer Methods and Programs in Biomedicine, 2022
- Application of an attention U-Net incorporating transfer learning for optic disc and cup segmentationSignal, Image and Video Processing, 2020
- Skin Lesion Segmentation with Improved Convolutional Neural NetworkJournal of Digital Imaging, 2020
- PHiSeg: Capturing Uncertainty in Medical Image Segmentation2019
- Boundary and Entropy-Driven Adversarial Learning for Fundus Image Segmentation2019
- Attention Guided Network for Retinal Image Segmentation2019
- Robust optic disc and cup segmentation with deep learning for glaucoma detectionComputerized Medical Imaging and Graphics, 2019
- Thyroid Nodule Segmentation in Ultrasound Images Based on Cascaded Convolutional Neural Network2018
- Skin lesion segmentation in dermoscopy images via deep full resolution convolutional networksComputer Methods and Programs in Biomedicine, 2018
- Ultrasound image-based thyroid nodule automatic segmentation using convolutional neural networksInternational Journal of Computer Assisted Radiology and Surgery, 2017