Style harmonization of panoramic radiography using deep learning.

Oral Radiol

Department of Oral and Maxillofacial Radiology, Yonsei University College of Dentistry, 50-1 Yonsei-Ro, Seodaemun-Gu, Seoul, 03722, Republic of Korea.

Published: October 2024

AI Article Synopsis

  • * A total of 15,624 images were analyzed, and a specific subset of 222 image pairs was used to evaluate how well the P-unit images could be transformed to resemble those from the R-unit.
  • * Results showed that the harmonized images had better similarity metrics compared to original P-unit images, with experts identifying a significant portion of them as resembling R-unit images, indicating the effectiveness of CycleGAN for this purpose.

Article Abstract

Objectives: This study aimed to harmonize panoramic radiograph images from different equipment in a single institution to display similar styles.

Methods: A total of 15,624 panoramic images were acquired using two different equipment: 8079 images from Rayscan Alpha Plus (R-unit) and 7545 images from Pax-i plus (P-unit). Among these, 222 image pairs (444 images) from the same patients comprised the test dataset to harmonize the P-unit images with the R-unit image style using CycleGAN. Objective evaluations included Frechet Inception Distance (FID) and Learned Perceptual Image Patch Similarity (LPIPS) assessments. Additionally, expert evaluation was conducted by two oral and maxillofacial radiologists on transformed P-unit and R-unit images. The statistical analysis of LPIPS employed a Student's t-test.

Results: The FID and mean LPIPS values of the transformed P-unit images (7.362, 0.488) were lower than those of the original P-unit images (8.380, 0.519), with a significant difference in LPIPS (p < 0.05). The experts evaluated 43.3-46.7% of the transformed P-unit images as R-unit images, 20.0-28.3% as P-units, and 28.3-33.3% as undetermined images.

Conclusions: CycleGAN has the potential to harmonize panoramic radiograph image styles. Enhancement of the model is anticipated for the application of images produced by additional units.

Download full-text PDF

Source
http://dx.doi.org/10.1007/s11282-024-00782-2DOI Listing

Publication Analysis

Top Keywords

p-unit images
12
images
9
transformed p-unit
8
p-unit
5
style harmonization
4
harmonization panoramic
4
panoramic radiography
4
radiography deep
4
deep learning
4
learning objectives
4

Similar Publications

Style harmonization of panoramic radiography using deep learning.

Oral Radiol

October 2024

Department of Oral and Maxillofacial Radiology, Yonsei University College of Dentistry, 50-1 Yonsei-Ro, Seodaemun-Gu, Seoul, 03722, Republic of Korea.

Article Synopsis
  • * A total of 15,624 images were analyzed, and a specific subset of 222 image pairs was used to evaluate how well the P-unit images could be transformed to resemble those from the R-unit.
  • * Results showed that the harmonized images had better similarity metrics compared to original P-unit images, with experts identifying a significant portion of them as resembling R-unit images, indicating the effectiveness of CycleGAN for this purpose.
View Article and Find Full Text PDF

Objectives: The aim of the current study was to describe a method that assesses the hyperemic microvascular blood plasma volume of the calf musculature. The reversibly albumin binding contrast agent gadofosveset was used in dynamic contrast-enhanced magnetic resonance imaging (DCE MRI) to assess the microvascular status in patients with peripheral arterial disease (PAD) and healthy controls. In addition, the reproducibility of this method in healthy controls was determined.

View Article and Find Full Text PDF

The electric fish Apteronotus leptorhynchus emits a high-frequency electric organ discharge (EOD) sensed by specialized electroreceptors (P-units) distributed across the fish's skin. Objects such as prey increase the amplitude of the EOD over the underlying skin and thus cause an increase in P-unit discharge. The resulting localized intensity increase is called the electric image and is detected by its effect on the P-unit population; the electric image peak value and the extent to its spreads are cues utilized by these fish to estimate the location and size of its prey.

View Article and Find Full Text PDF

Receptive field organization across multiple electrosensory maps. I. Columnar organization and estimation of receptive field size.

J Comp Neurol

October 2009

Department of Cellular and Molecular Medicine, University of Ottawa, Ottawa, Ontario K1H 8M5, Canada.

The electric fish Apteronotus leptorhynchus emits a high-frequency electric organ discharge (EOD) sensed by specialized electroreceptors (P-units). Amplitude modulations (AMs) of the EOD are caused by objects such as prey as well as by social interactions with conspecifics. The firing rate of P-units is modulated by the AMs due to both objects and communication signals.

View Article and Find Full Text PDF

Want AI Summaries of new PubMed Abstracts delivered to your In-box?

Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!