Normalization techniques for CNN based analysis of surgical cone beam CT volumes

Martín Vicario C, Kordon FJ, Denzinger F, El Barbari JS, Privalov M, Franke J, Maier A, Kunze H (2022)

Publication Language: English

Publication Type: Conference contribution, Conference Contribution

Publication year: 2022

Publisher: SPIE

Book Volume: 120322G

Conference Proceedings Title: Medial Imaging 2022, Image Processing

Event location: San Diego, California

DOI: 10.1117/12.2606973

Abstract

Purpose: The choice of input normalization has effects on the generalization and performance of deep neural networks. While this topic is explored for 2D imaging applications, the influence of different normalization techniques on medical imaging modalities, e.g. cone-beam CT (CBCT), differs due to a different value range and distribution. In this paper a good normalization technique for intra-operatively acquired surgical CBCT volumes is presented. Methods: A set of normalization strategies, namely histogram equalization, min-max scaling, z-score normalization, linear look up table (LUT) with clipping and sigmoid function with clipping is compared on a CBCT volume classification task. Results: The results show that a combination of parameterized LUTs and clipping with the range [-710, 1640] HU independent of the underlying intensity histogram provides the best performance for the task at hand. Conclusions: The clipping based normalization technique helps to compress the feature space to the relevant range. By this approach, most of the information about the intensity values of soft tissue and bone is retained. The clipping range presented in this paper is valid for surgical CBCTs.

Authors with CRIS profile

Involved external institutions

BG Unfallklinik Ludwigshafen (BGU) / BG Hospital Ludwigshafen Germany (DE)

How to cite

APA:

Martín Vicario, C., Kordon, F.J., Denzinger, F., El Barbari, J.S., Privalov, M., Franke, J.,... Kunze, H. (2022). Normalization techniques for CNN based analysis of surgical cone beam CT volumes. In Išgum I, Colliot O (Eds.), Medial Imaging 2022, Image Processing. San Diego, California: SPIE.

MLA:

Martín Vicario, Celia, et al. "Normalization techniques for CNN based analysis of surgical cone beam CT volumes." Proceedings of the SPIE Medical Imaging 2022, San Diego, California Ed. Išgum I, Colliot O, SPIE, 2022.

BibTeX: Download