Logo image
Chapter 11 - LOGISMOS-JEI: Segmentation using optimal graph search and just-enough interaction
Book chapter

Chapter 11 - LOGISMOS-JEI: Segmentation using optimal graph search and just-enough interaction

Honghai Zhang, Kyungmoo Lee, Zhi Chen, Satyananda Kashyap and Milan Sonka
Handbook of Medical Image Computing and Computer Assisted Intervention, pp.249-272
Elsevier Inc
2020
DOI: 10.1016/B978-0-12-816176-0.00016-8

View Online

Abstract

LOGISMOS (Layered Optimal Graph Image Segmentation for Multiple Objects and Surfaces) is a general approach for optimally segmenting multiple n-D surfaces that mutually interact within individual and/or between object. JEI (Just-Enough Interaction) extension enables the users to intuitively and efficiently guide LOGISMOS to achieve clinically acceptable segmentation that still maintains global optimality, embedded shape, and anatomy knowledge. This chapter is aimed at both the conceptual designers of quantitative medical image analysis methods and developers adopting LOGISMOS-JEI to perform new segmentation tasks. The basic principles and essential techniques of LOGISMOS-JEI are presented. Using three clinical segmentation problems, the process of designing LOGISMOS-JEI applications is described with a focus on choosing proper techniques at all stages of the analysis pipeline. The capability of JEI is demonstrated by reporting the achieved segmentation accuracy and quantifying the levels of human effort involved.
Graph search Just-Enough Interaction (JEI) LOGISMOS Optimal multisurface segmentation Segmentation

Details

Metrics

165 Record Views
Logo image