Search

Your search keyword '"Gui, Na"' showing total 12 results
Start Over Author "Gui, Na" Publication Year Range Last 3 years
12 results on '"Gui, Na"'

7. Retinal Layer Segmentation in OCT Images With Boundary Regression and Feature Polarization

Author

Tan, Yubo, Shen, Wen-Da, Wu, Ming-Yuan, Liu, Gui-Na, Zhao, Shi-Xuan, Chen, Yang, Yang, Kai-Fu, and Li, Yong-Jie

Abstract

The geometry of retinal layers is an important imaging feature for the diagnosis of some ophthalmic diseases. In recent years, retinal layer segmentation methods for optical coherence tomography (OCT) images have emerged one after another, and huge progress has been achieved. However, challenges due to interference factors such as noise, blurring, fundus effusion, and tissue artifacts remain in existing methods, primarily manifesting as intra-layer false positives and inter-layer boundary deviation. To solve these problems, we propose a method called Tightly combined Cross-Convolution and Transformer with Boundary regression and feature Polarization (TCCT-BP). This method uses a hybrid architecture of CNN and lightweight Transformer to improve the perception of retinal layers. In addition, a feature grouping and sampling method and the corresponding polarization loss function are designed to maximize the differentiation of the feature vectors of different retinal layers, and a boundary regression loss function is devised to constrain the retinal boundary distribution for a better fit to the ground truth. Extensive experiments on four benchmark datasets demonstrate that the proposed method achieves state-of-the-art performance in dealing with problems of false positives and boundary distortion. The proposed method ranked first in the OCT Layer Segmentation task of GOALS challenge held by MICCAI 2022. The source code is available at https://www.github.com/tyb311/TCCT.

Published
2024
Full Text
View/download PDF

8. Motion of spiral waves induced by local pacing

Author

Chen, Jiang-Xing, Wei, Gui-Na, Li, Bing-Wei, and Xu, Jiang-Rong

Abstract

Abstract: The motion of spiral waves in excitable media driven by a weak pacing around the spiral tip is investigated numerically as well as theoretically. We presented a Bifurcations diagram containing four types of the spiral motion induced by different frequencies of pacing: rigidly rotating, inward-petal meandering, resonant drift, and outward-petal meandering spiral. Simulation shows that the spiral resonantly drifts when the frequency of pacing is close to that of the spiral rotation. We also find that the speed and direction of the drift can be efficiently controlled by means of the strength and phase of the local pacing, which is consistent with analytical results based on the framework of the weak deformation approximation.

Published
2024
Full Text
View/download PDF

12. Utility of ECG-Gated CTA for Detecting and Classifying Coronary Artery Involvement in Patients With Type A Aortic Dissection and Sinus of Valsalva Involvement.

Author

Ma GN, Hong H, Alwalid O, Cao YK, Hu XH, Han P, and Gu J

Subjects
Adult, Aortic Dissection surgery, Coronary Vessels physiopathology, Coronary Vessels surgery, Female, Humans, Male, Middle Aged, Sinus of Valsalva physiopathology, Sinus of Valsalva surgery, Aortic Dissection diagnostic imaging, Computed Tomography Angiography methods, Coronary Angiography methods, Coronary Vessels diagnostic imaging, Electrocardiography methods, Sinus of Valsalva diagnostic imaging
Abstract

BACKGROUND. Extension of type A aortic dissection (TAAD) from the sinus of Valsalva (SV) into the coronary arteries is associated with a poor prognosis and requires direct coronary repair or coronary artery bypass grafting (CABG) depending on the extent of involvement. OBJECTIVE. The purpose of this study was to assess the diagnostic performance of ECG-gated aortic CTA with surgical findings as the reference for detection and classification of coronary artery involvement in patients with TAAD involving the SV. METHODS. This retrospective study included 112 patients who underwent preoperative ECG-gated aortic CTA that showed TAAD with SV involvement. Two radiologists independently reviewed CTA images for right coronary artery (RCA) and left coronary artery (LCA) involvement. Involvement was classified according to a previously described system as type A (coronary ostial dissection), B (dissection with coronary false channel), or C (complete detachment from aortic root with dissection encircling the coronary artery). The diagnostic performance of CTA was calculated with surgical findings as the reference; interreader agreement was assessed; and surgical interventions were summarized. RESULTS. At surgery, the RCA was uninvolved in 33 patients and had type A involvement in 45, type B involvement in 19, and type C involvement in 15 patients. The LCA was uninvolved in 70 patients and had type A involvement in 34 patients, type B involvement in eight patients, and type C involvement in no patient. For the two readers, sensitivity in making the diagnosis in the RCA was 86.7% and 91.1% for type A, 79.0% and 89.5% for type B, 86.7% and 93.3% for type C, and 97.5% and 98.7% for any involvement. Sensitivity for the LCA was 85.3% and 91.2% for type A, 87.5% and 100% for type B, and 100% for any involvement. Specificity for the RCA was 94.0% and 97.0% for type A, 95.7% and 96.7% for type B, 96.9% and 96.9% for type C, and 93.9% and 97.0% for any involvement. Specificity for the LCA was 96.2% and 98.7% for type A, 96.2% and 97.1% for type B, and 97.1% and 98.6% for any involvement. Interreader agreement for types of involvement ranged from a kappa value of 0.85 to 0.96. The most common interventions were aortic repair for SV involvement alone (55.7-63.6%), coronary artery repair for types A and B involvement (53.3-87.5%), and CABG for type C involvement (86.7%). CONCLUSION. ECG-gated CTA has high diagnostic performance in the detection and classification of coronary involvement in TAAD with SV involvement. CLINICAL IMPACT. CTA findings may help guide presurgical planning for patients with TAAD.

Published
2021
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results