Search Our Scientific Publications & Authors

Publications
find publications by category +
Translate page:

Accuracy of auto-identification of the posteroanterior cephalometric landmarks using cascade convolution neural network algorithm and cephalometric images of different quality from nationwide multiple centers.

Authors:
Soo-Min Gil Inhwan Kim Jin-Hyoung Cho Mihee Hong Minji Kim Su-Jung Kim Yoon-Ji Kim Young Ho Kim Sung-Hoon Lim Sang Jin Sung Seung-Hak Baek Namkug Kim Kyung-Hwa Kang

Am J Orthod Dentofacial Orthop 2022 Apr 21;161(4):e361-e371. Epub 2022 Jan 21.

Department of Orthodontics, School of Dentistry, Wonkwang University, Iksan-si, South Korea. Electronic address:

Introduction: The purpose of this study was to evaluate the accuracy of auto-identification of the posteroanterior (PA) cephalometric landmarks using the cascade convolution neural network (CNN) algorithm and PA cephalogram images of a different quality from nationwide multiple centers nationwide.

Methods: Of the 2798 PA cephalograms from 9 university hospitals, 2418 images (2075 training set and 343 validation set) were used to train the CNN algorithm for auto-identification of 16 PA cephalometric landmarks. Subsequently, 99 pretreatment images from the remaining 380 test set images were used to evaluate the accuracy of auto-identification of the CNN algorithm by comparing with the identification by a human examiner (gold standard) using V-Ceph 8.0 (Ostem, Seoul, South Korea). Pretreatment images were used to eliminate the effects of orthodontic bracket, tube and wire, surgical plate, and surgical screws. Paired t test was performed to compare the x- and y-coordinates of each landmark. The point-to-point error and the successful detection rate (range, within 2.0 mm) were calculated.

Results: The number of landmarks without a significant difference between the location identified by the human examiner and by auto-identification by the CNN algorithm were 8 on the x-coordinate and 5 on the y-coordinate, respectively. The mean point-to-point error was 1.52 mm. The low point-to-point error (<1.0 mm) was observed at the left and right antegonion (0.96 mm and 0.99 mm, respectively) and the high point-to-point error (>2.0 mm) was observed at the maxillary right first molar root apex (2.18 mm). The mean successful detection rate of auto-identification was 83.3%.

Conclusions: Cascade CNN algorithm for auto-identification of PA cephalometric landmarks showed a possibility of an effective alternative to manual identification.

Download full-text PDF

 Source
http://dx.doi.org/10.1016/j.ajodo.2021.11.011DOI Listing
April 2022

Publication Analysis

Top Keywords

cnn algorithm
20
cephalometric landmarks
16
point-to-point error
12
accuracy auto-identification
12
evaluate accuracy
8
convolution neural
8
pretreatment images
8
nationwide multiple
8
multiple centers
8
neural network
8
auto-identification cephalometric
8
cascade convolution
8
algorithm auto-identification
8
images quality
8
human examiner
8
landmarks cascade
8
detection rate
8
auto-identification cnn
8
posteroanterior cephalometric
8
auto-identification posteroanterior
8

Keyword Occurance

Similar Publications

Deep learning for blind structured illumination microscopy.

Authors:
Emmanouil Xypakis Giorgio Gosti Taira Giordani Raffaele Santagati Giancarlo Ruocco Marco Leonetti

Sci Rep 2022 May 21;12(1):8623. Epub 2022 May 21.

Center for Life Nano- and Neuro-Science, Istituto Italiano di Tecnologia, Viale Regina Elena 291, 00161, Rome, Italy.

Blind-structured illumination microscopy (blind-SIM) enhances the optical resolution without the requirement of nonlinear effects or pre-defined illumination patterns. It is thus advantageous in experimental conditions where toxicity or biological fluctuations are an issue. In this work, we introduce a custom convolutional neural network architecture for blind-SIM: BS-CNN. Read More

View Article and Full-Text PDF
May 2022
Similar Publications

Convolutional neural network-based reconstruction for positronium annihilation localization.

Authors:
Jin Jegal Dongwoo Jeong Eun-Suk Seo HyeoungWoo Park Hongjoo Kim

Sci Rep 2022 May 20;12(1):8531. Epub 2022 May 20.

Department of Physics, Kyungpook National University, Daegu, 41566, Korea.

A novel hermetic detector composed of 200 bismuth germanium oxide crystal scintillators and 393 channel silicon photomultipliers has been developed for positronium (Ps) annihilation studies. This compact 4π detector is capable of simultaneously detecting γ-ray decay in all directions, enabling not only the study of visible and invisible exotic decay processes but also tumor localization in positron emission tomography for small animals. In this study, we investigate the use of a convolutional neural network (CNN) for the localization of Ps annihilation synonymous with tumor localization. Read More

View Article and Full-Text PDF
May 2022
Similar Publications

A deep learning framework to classify breast density with noisy labels regularization.

Authors:
Hector Lopez-Almazan Francisco Javier Pérez-Benito Andrés Larroza Juan-Carlos Perez-Cortes Marina Pollan Beatriz Perez-Gomez Dolores Salas Trejo María Casals Rafael Llobet

Comput Methods Programs Biomed 2022 May 12;221:106885. Epub 2022 May 12.

Instituto Tecnológico de la Informática, Universitat Politècnica de València,Camino de Vera, s/n, 46022 València, Spain. Electronic address:

Background And Objective: Breast density assessed from digital mammograms is a biomarker for higher risk of developing breast cancer. Experienced radiologists assess breast density using the Breast Image and Data System (BI-RADS) categories. Supervised learning algorithms have been developed with this objective in mind, however, the performance of these algorithms depends on the quality of the ground-truth information which is usually labeled by expert readers. Read More

View Article and Full-Text PDF
May 2022
Similar Publications

Masked Contrastive Representation Learning for Reinforcement Learning.

Authors:
Jinhua Zhu Yingce Xia Lijun Wu Jiajun Deng Wengang Zhou Tao Qin Tie-Yan Liu Houqiang Li

IEEE Trans Pattern Anal Mach Intell 2022 May 20;PP. Epub 2022 May 20.

In pixel-based reinforcement learning (RL), the states are raw video frames, which are mapped into hidden representation before feeding to a policy network. To improve sample efficiency of state representation learning, the most recent state-of-the-art work is based on contrastive unsupervised representation. Witnessing that consecutive video frames in a game are highly correlated, to further improve data efficiency, we propose a new algorithm, masked contrastive representation learning for RL (M-CURL), which takes the correlation among consecutive inputs into consideration. Read More

View Article and Full-Text PDF
May 2022
Similar Publications

Modeling and Fault Detection of Brushless Direct Current Motor by Deep Learning Sensor Data Fusion.

Authors:
Priscile Suawa Tenia Meisel Marcel Jongmanns Michael Huebner Marc Reichenbach

Sensors (Basel) 2022 May 5;22(9). Epub 2022 May 5.

Department of Computer Engineering, Brandenburg University of Technology Cottbus-Senftenberg, 03046 Cottbus, Germany.

Only with new sensor concepts in a network, which go far beyond what the current state-of-the-art can offer, can current and future requirements for flexibility, safety, and security be met. The combination of data from many sensors allows a richer representation of the observed phenomenon, e.g. Read More

View Article and Full-Text PDF
May 2022
Similar Publications
}