Deep Interactive Segmentation of Medical Images: A Systematic Review and Taxonomy

TL;DR

This review of interactive segmentation in medical image analysis provides a structured overview of this emerging field featuring a comprehensive taxonomy, a systematic review of existing methods, and an in-depth analysis of current practices.

Abstract

Interactive segmentation is a crucial research area in medical image analysis aiming to boost the efficiency of costly annotations by incorporating human feedback. This feedback takes the form of clicks, scribbles, or masks and allows for iterative refinement of the model output so as to efficiently guide the system towards the desired behavior. In recent years, deep learning-based approaches have propelled results to a new level causing a rapid growth in the field with 121 methods proposed in the medical imaging domain alone. In this review, we provide a structured overview of this emerging field featuring a comprehensive taxonomy, a systematic review of existing methods, and an in-depth analysis of current practices. Based on these contributions, we discuss the challenges and opportunities in the field. For instance, we find that there is a severe lack of comparison across methods which needs to be tackled by standardized baselines and benchmarks.

Figures (6)

• Figure 1: Tendencies in medical interactive segmentation in recent years.
• Figure 2: Search strategy in our systematic review for selecting relevant studies. The logos in steps 1 and 4 are illustrated only as examples for literature databases and venues respectively. A full list is given in the Appendixes.
• Figure 3: Our proposed taxonomy tree for all the reviewed studies. The references for studies associated with a node are listed beneath the respective node.
• Figure 4: Taxonomy blueprints for our proposed taxonomy nodes. The human annotator is involved during training, application, or both in online learning.
• Figure 5: Distribution of segmentation targets in the anatomy (top left) and in the pathology (top right), imaging modalities (bottom left), and evaluation metrics (bottom right) among all reviewed papers. The numbers represent the number of papers in that category. The icons on the top row are designed by https://www.flaticon.com.