A Conservative Image Boundary Extraction Method with Application to the ILM Tumor Surgery
Primary Area: unsupervised, self-supervised, semi-supervised, and supervised representation learning
Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.
Keywords: Conservative Boundary Extraction, Infant Lymphatic Malformations Tumor, Unsupervised Learning
Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2024/AuthorGuide.
Abstract: While infant lymphatic malformation tumors are benign, they are very difficult to remove. The removal process is very delicate and requires the retention of as much healthy tissue as possible. Commonly utilized boundary extraction methods aim to extract boundaries covering the vast majority of the target area which remove more healthy tissue than is desirable. This paper presents a conservative image boundary extraction (CIBE) approach with well-designed iterative boundary shrinkage procedures which are applied to computerized tomography (CT) images for use in ILM tumor resection operations. CIBE incorporates three primary concepts: Fuzzy Degree, Pixel Deepness and Boundary Smoothness. The proposed algorithm first converts the marked CT image into a 0-1 image matrix. Then it shrinks the boundary according to the estimated PD and BS indices for the image in an iterative fashion until the boundary smoothness meets the desired level. Empirical analysis demonstrates that the smooth, conservative tumor boundaries are obtained using the CIBE algorithm. The proposed method can also be easily extended to the three dimensional studies.
Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors' identity.
No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.
Submission Number: 178
Loading