Antong Chen

Antong Chen

Auto-generated profile, based on publicly available information

Antong Chen

Merck & Co., Inc

Research Focus & Metrics

Academic Career Length

16Years (2011-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
15 March 2018
Verified
Book Chapter
Closed access
PDF file
ACAntong Chen
Antong Chen
TZTian Zhou
Tian Zhou
IIIlknur Icke
Ilknur Icke
SPSarayu Parimal
Sarayu Parimal
BDBelma Dogdas
Belma Dogdas
CCChih-Liang Chin
Chih-Liang Chin

31 October 2011 in IEEE Transactions on Plasma Science
Verified
Journal Article
Closed access
XYXueling Yao
Xueling Yao
JCJingliang Chen
Jingliang Chen
ACAntong Chen
Antong Chen
A surge protective gap (SPG) with an active energy-coupling-triggered circuit is designed, and its discharge characteristics are studied. The experimental results show that the SPG with active-energy-coupling-triggered circuit and flashover trigger device has excellent surge protective characteristics. The gap distance is 5 mm, and the produced p · d is 90 Pa · mm. Moreover, with no active-energy-coupling-triggered circuit, the dc breakdown voltage is 2.32 kV, and the pulse breakdown voltage is 5.75 kV. Therefore, the voltage ratio, which is defined as the specific value of the pulse breakdown voltage and the dc breakdown voltage, is 2.48. When using the BaTiO3 flashover trigger device and the capacitor-coupling-triggered circuit and with 5-mm gap distance and p · d = 90 Pa · mm, the pulse breakdown voltage can reduce to 2.16 kV, the voltage ratio is close to one, and the response time is less than 100 ns. The results of the study lay a theoretical foundation for further research on enhancing the voltage protection level of the SPG.

23 February 2012
Verified
Conference Paper
Closed access
PDF file
ACAntong Chen
Antong Chen
JNJack H Noble
Jack H Noble
KNKenneth J Niermann
Kenneth J Niermann
MDMatthew A Deeley
Matthew A Deeley
BDBenoit M Dawant
Benoit M Dawant
Automatic segmentation of parotid glands in head and neck CT images for IMRT planning has drawn attention in recent years. Although previous approaches have achieved substantial success by reaching high overall volume-wise accuracy, suboptimal segmentations are observed on the interior boundary of the gland where the contrast is poor against the adjacent muscle groups. Herein we propose to use a constrained active shape model with landmark uncertainty to improve the segmentation in this area. Results obtained using this method are compared with results obtained using a regular active shape model through a leave-one-out experiment.

7 February 2024 in IEEE Access
Verified
Journal Article
Open access
YLYiqiao Liu
Yiqiao Liu
SHSarah Halek
Sarah Halek
RCRandolph Crawford
Randolph Crawford
KPKeith Persson
Keith Persson
MTMichal Tomaszewski
Michal Tomaszewski
ACAntong Chen
Antong Chen
Deep convolutional neural networks (CNN) are often trained on 2D annotations created by radiologists following RECIST guidelines to segment lesions in 3D medical images. Three-dimensional segmentation is conducted by segmenting each lesion slice-by-slice on the axial direction and stacking the 2D segmentation masks into 3D. However, the performance of such models is inherently biased by the appearance of most of the lesions in the training dataset. Herein we propose an approach to generate accurate 3D segmentations of underrepresented necrotic lung lesions. Our proposed approach applies two novel augmentation techniques for contrastive learning pretraining: dependency augmentation that captures inter-slice and inter-timepoint dependencies, and distance transform-based mask-out augmentation imitating necrotic lesions. In dependency augmentation, cosine similarity within RECIST bounding box is applied to construct positive pairs from 2D image slices of the same lesion in the current 3D volume and across longitudinal scans. We further compared contrastive learning architectures, Momentum Contrast (MoCo) and Bootstrap Your Own Latent (BYOL), based upon two internal 3D testing sets, one with regular lung lesions and the other with necrotic lung lesions, and a public 3D DeepLesion lung lesion testing set. MoCo with both proposed augmentations demonstrated the best performance among all methods that were compared. Specifically, it 1) improved Dice similarity coefficient (DSC) by 8.42% over baseline model trained from scratch and 2.40% over ImageNet pretrained model on the 3D necrotic lung lesion set; 2) achieved better segmentation performance on necrotic lesions with 10% of labeled data for supervised fine-tuning compared with the baseline model trained with all labels from scratch.

2 March 2018
Verified
Conference Paper
Closed access
PDF file
ACAntong Chen
Antong Chen
DZDongqing Zhang
Dongqing Zhang
IIIlknur Icke
Ilknur Icke
BDBelma Dogdas
Belma Dogdas
SPSarayu Parimal
Sarayu Parimal
CCChih-Liang Chin
Chih-Liang Chin
In the development of treatments for cardiovascular diseases, short axis cardiac cine MRI is important for the assessment of various structural and functional properties of the heart. In short axis cardiac cine MRI, Cardiac properties including the ventricle dimensions, stroke volume, and ejection fraction can be extracted based on accurate segmentation of the left ventricle (LV) myocardium. One of the most advanced segmentation methods is based on fully convolutional neural networks (FCN) and can be successfully used to do segmentation in cardiac cine MRI slices. However, the temporal dependency between slices acquired at neighboring time points is not used. Here, based on our previously proposed FCN structure, we proposed a new algorithm to segment LV myocardium in porcine short axis cardiac cine MRI by incorporating convolutional long short-term memory (Conv-LSTM) to leverage the temporal dependency. In this approach, instead of processing each slice independently in a conventional CNN-based approach, the Conv-LSTM architecture captures the dynamics of cardiac motion over time. In a leave-one-out experiment on 8 porcine specimens (3,600 slices), the proposed approach was shown to be promising by achieving average mean Dice similarity coefficient (DSC) of 0.84, Hausdorff distance (HD) of 6.35 mm, and average perpendicular distance (APD) of 1.09 mm when compared with manual segmentations, which improved the performance of our previous FCN-based approach (average mean DSC=0.84, HD=6.78 mm, and APD=1.11 mm). Qualitatively, our model showed robustness against low image quality and complications in the surrounding anatomy due to its ability to capture the dynamics of cardiac motion.

28 January 2022
Verified
Conference Paper
Open access
PDF file
ACAntong Chen
Antong Chen
Hong Kong films with Cantonese as the main language have long been a significant part of Hong Kong’s popular culture. The main themes of action movies with Kung fu films, gangster gunfights, and police-and-bandit films have become a major representative. As an important embodiment of violent aesthetics, a literary form in the history of films, shooting, fighting, chasing, and other scenes that can be seen in most actions films bring a multi-sensory feast to the audience. As time passed, this artistically presenting cruel and bloody scenes has changed in Hong Kong action films. Taking police and bandit films as an example, compared with the last century, the presentation of violent aesthetics in them seems to be gradually weakened. The reasons accounting for this phenomenon are complex and diverse. Political, economical and value factors have played an important role. For example, the signing of the Closer Economic Partnership Arrangement in 2003 has brought closer ties between mainland China and Hong Kong Special Administrative Region (HKSAR). With the mainland market’s increasing importance, filmmakers in this special administrative region have been facing a new market environment and censorship, and they have to adjust their products to meet a larger market needs. In addition, Hong Kong society’s pursuit of the rule of law and harmonious and stable values also played an influential role in declining the direct representation of violence in films. The new action film racing fire, released in China on July 30, 2021, seems to bring some different voices.