A Label‐Efficient Framework for Automated Sinonasal CT Segmentation in Image‐Guided Surgery
Otolaryngology–Head and Neck Surgery · 2024-10 · 171 (4) · 1217-1225
DOI: 10.1002/ohn.868 ↗At a glance
Journal article in Otolaryngology–Head and Neck Surgery, published online on 2024-06-23. Written by Manish Sahu and 8 co-authors. Institution: Laboratory for Computational Sensing and Robotics, Johns Hopkins University, Baltimore, Maryland, USA. Major MEDLINE topics: Tomography, X-Ray Computed, Surgery, Computer-Assisted, Deep Learning. Cited 6 times so far (Crossref); 22 references. The full text is freely available in PubMed Central. Funded by: NIDCD NIH HHS.
Abstract (original language)
Objective
Segmentation, the partitioning of patient imaging into multiple, labeled segments, has several potential clinical benefits but when performed manually is tedious and resource intensive. Automated deep learning (DL)‐based segmentation methods can streamline the process. The objective of this study was to evaluate a label‐efficient DL pipeline that requires only a small number of annotated scans for semantic segmentation of sinonasal structures in CT scans. Study Design Retrospective cohort study.
Setting
Academic institution.
Methods
Forty CT scans were used in this study including 16 scans in which the nasal septum (NS), inferior turbinate (IT), maxillary sinus (MS), and optic nerve (ON) were manually annotated using an open‐source software. A label‐efficient DL framework was used to train jointly on a few manually labeled scans and the remaining unlabeled scans. Quantitative analysis was then performed to obtain the number of annotated scans needed to achieve submillimeter average surface distances (ASDs).
Results
Our findings reveal that merely four labeled scans are necessary to achieve median submillimeter ASDs for large sinonasal structures—NS (0.96 mm), IT (0.74 mm), and MS (0.43 mm), whereas eight scans are required for smaller structures—ON (0.80 mm).
Conclusion
We have evaluated a label‐efficient pipeline for segmentation of sinonasal structures. Empirical results demonstrate that automated DL methods can achieve submillimeter accuracy using a small number of labeled CT scans. Our pipeline has the potential to improve pre‐operative planning workflows, robotic‐ and image‐guidance navigation systems, computer‐assisted diagnosis, and the construction of statistical shape models to quantify population variations. Level of Evidence N/A
Subjects
MeSH terms (MEDLINE)
Author keywords
Authors and institutions
- Laboratory for Computational Sensing and Robotics, Johns Hopkins University, Baltimore, Maryland, USA.
- Department of Neurosurgery, Johns Hopkins University School of Medicine, Baltimore, Maryland, USA.
- Division of Otolaryngology, Department of Surgery, University of British Columbia, Vancouver, British Columbia, Canada.
- Department of Otolaryngology, Johns Hopkins University School of Medicine, Baltimore, Maryland, USA.
ORCID profiles: Manish Sahu
Funding
- NIDCD NIH HHS
Cite
APA
Sahu, M., Xiao, Y., Porras, J. L., Amanian, A., Jain, A., Thamboo, A., Taylor, R. H., Creighton, F. X., & Ishii, M. (2024). A Label‐Efficient Framework for Automated Sinonasal CT Segmentation in Image‐Guided Surgery. Otolaryngology–Head and Neck Surgery, 171(4), 1217-1225. https://doi.org/10.1002/ohn.868
Vancouver
Sahu M, Xiao Y, Porras JL, Amanian A, Jain A, Thamboo A, et al. A Label‐Efficient Framework for Automated Sinonasal CT Segmentation in Image‐Guided Surgery. Otolaryngology–Head and Neck Surgery. 2024;171(4):1217-1225. doi:10.1002/ohn.868
BibTeX
@article{Sahu2024,
author = {Sahu, Manish and Xiao, Yuliang and Porras, Jose L and Amanian, Ameen and Jain, Aseem and Thamboo, Andrew and Taylor, Russell H and Creighton, Francis X and Ishii, Masaru},
title = {A Label‐Efficient Framework for Automated Sinonasal CT Segmentation in Image‐Guided Surgery},
journal = {Otolaryngology–Head and Neck Surgery},
year = {2024},
volume = {171},
number = {4},
pages = {1217--1225},
doi = {10.1002/ohn.868}
}
EUVIMED