Multiphase Computed Tomographic Angiography with Bone Subtraction Using 3D Multichannel Convolution Neural Networks

Huang, Adam; Cheng, Wen-Hsiang; Lee, Chung-Wei; Yang, Chung‐Yi; Liu, Hon‐Man

doi:10.1109/embc44109.2020.9176844

Cited by 3 publications

(5 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The CTA images needed to be preprocessed. The coarse-to-fine (C2F) method [ 12 ] was adopted to segment the arteries in CTA images. After the upsampling of the original CTA image I in the coarse segmentation, the obtained coarse segmentation output image is expressed as equation ( 1 ):

…”

Section: Methodsmentioning

confidence: 99%

“…Sensitivity (Sen) was utilized to measure the probability that a voxel that was positive in the gold standard was also positive in the segmentation result, and it is calculated using equation (12):…”

Section: Evaluation Indicators Of Cta Image Segmentationmentioning

confidence: 99%

“…Inclusion criteria were as follows: (1) patients were between the ages of 18 and 60 years old, regardless of gender; and (2) patients had skin and soft tissue defects of the limbs. e exclusion criteria were as follows: (1) patients had a history of iodine allergy; (2) patients had the positive result of iodine allergy test; (3) patients' lateral circumflex femoral artery was determined as the type without thick branches; (4) patients suffered from liver or kidney dysfunction (glutamic-pyruvic transaminase >80 U/L); (5) patients got pulmonary infection shown in chest X-ray examination; (6) patients had renal insufficiency (creatinine >200 mg/L); (7) patients had a cardiac function > grade II; (8) local ulcers in patients were diagnosed as malignant tumors; (9) patients had a history of mental illness once; (11) patients got severe allergic reactions during CTA examination; (12) patients had an anesthesia accident during anesthesia, which made the surgery could not be performed; and (13) the lateral femoral circumflex artery was classified with no thick branch during the surgery, or they went with poor vascular conditions.…”

Section: Research Objects and Groupsmentioning

confidence: 99%

See 2 more Smart Citations

Computed Tomography Angiography and B‐Mode Ultrasonography under Artificial Intelligence Plaque Segmentation Algorithm in the Perforator Localization for Preparation of Free Anterolateral Femoral Flap

Shen

Huang

et al. 2022

Contrast Media & Molecular Imaging

View full text Add to dashboard Cite

This research was aimed to investigate the accuracy of U-shaped network (UNet)-based computed tomography angiography (CTA) and B-mode ultrasonography (US) in the perforator localization of free anterolateral thigh flap (ALTF). Based on UNet, a fusion of deep supervision mechanism, squeeze-and-excitation module, and attention mechanism was introduced to optimize the algorithm. Then, a CTA segmentation model, DA-UNet, was established. The segmentation performance of DA-UNet and other algorithms was compared under the same conditions. 30 patients who were planned to receive ALTF surgery were selected as the research objects. According to different preoperative localization methods, they were divided into group A (CTA) and group B (B-mode US), 15 cases in each group. Combined with the actual situation during surgery, the diagnostic accordance rate, sensitivity (Sen), specificity, and the distance between the perforator location and the actual location were compared between the two groups. The Dice coefficient, Jaccard index, Sen, the area under curve (AUC), and average Hausdorff distance (AVD) of the DA-UNet segmentation algorithm were 80.70%, 69.97%, 77.56%, 0.887, and 2.48, respectively. These results were significantly better than those of other algorithms ( P < 0.05 ). In group A, the diagnostic accordance rate, Sen, and specificity of patients were 96.55%, 90.52%, and 73.58%, respectively, which were higher than 91.53%, 81.36%, and 15.60% of patients in group B significantly ( P < 0.05 ). There was no statistical difference in the distance between the perforator location and the actual location ( P > 0.05 ). It showed that the accuracy of CTA under the UNet-based DA-UNet segmentation model in the perforator localization of ALTF was better than that of B-mode US. Thus, a reference could be provided for the preparation of free ALTF and its clinical application.

show abstract

…”

Section: Methodsmentioning

confidence: 99%

Section: Evaluation Indicators Of Cta Image Segmentationmentioning

confidence: 99%

Section: Research Objects and Groupsmentioning

confidence: 99%

See 1 more Smart Citation

Computed Tomography Angiography and B‐Mode Ultrasonography under Artificial Intelligence Plaque Segmentation Algorithm in the Perforator Localization for Preparation of Free Anterolateral Femoral Flap

Shen

Huang

et al. 2022

Contrast Media & Molecular Imaging

View full text Add to dashboard Cite

show abstract

“…Consistency loss. In order to form constraints between the original H-phase (or E-phase) image and its style-transferred image and make the segmentation network learn useful knowledge lost due to style transfer, Consistency loss L CON is set between the corresponding prediction results P A and P A→V , as shown in Equation (7).…”

Section: Loss Functionmentioning

confidence: 99%

“…Clinically, multiphase computer tomography angiography (CTA) images are obtained by scanning at different times of peak venous flow. Due to the complementary vessel information on these images presenting a comprehensive status of the collateral circulation, they are routinely used for disease diagnosis and patient prognosis prediction [ 7 ]. For example, the diagnosis of portal vein disease often depends on observing the portal vein on the abdominal CTA images of the hepatic artery phase (H-phase), the portal vein phase (P-phase) and the equilibrium phase (E-phase) simultaneously, and intuitively fusing CTA images of different phases, so as to complement vessel information and implement effective treatment [ 8 ].…”

Section: Introductionmentioning

confidence: 99%

Segmentation of Portal Vein in Multiphase CTA Image Based on Unsupervised Domain Transfer and Pseudo Label

et al. 2023

View full text Add to dashboard Cite

Background: Clinically, physicians diagnose portal vein diseases on abdominal CT angiography (CTA) images scanned in the hepatic arterial phase (H-phase), portal vein phase (P-phase) and equilibrium phase (E-phase) simultaneously. However, existing studies typically segment the portal vein on P-phase images without considering other phase images. Method: We propose a method for segmenting portal veins on multiphase images based on unsupervised domain transfer and pseudo labels by using annotated P-phase images. Firstly, unsupervised domain transfer is performed to make the H-phase and E-phase images of the same patient approach the P-phase image in style, reducing the image differences caused by contrast media. Secondly, the H-phase (or E-phase) image and its style transferred image are input into the segmentation module together with the P-phase image. Under the constraints of pseudo labels, accurate prediction results are obtained. Results: This method was evaluated on the multiphase CTA images of 169 patients. The portal vein segmented from the H-phase and E-phase images achieved DSC values of 0.76 and 0.86 and Jaccard values of 0.61 and 0.76, respectively. Conclusion: The method can automatically segment the portal vein on H-phase and E-phase images when only the portal vein on the P-phase CTA image is annotated, which greatly assists in clinical diagnosis.

show abstract

Revolutionizing Biometrics With AI-Enhanced X-Ray and MRI Analysis

Komperla,

Pokkuluri,

Nomula

et al. 2024

Advances in Medical Technologies and Clinical Practice

View full text Add to dashboard Cite

This chapter explores the innovative integration of artificial intelligence (AI) in enhancing biometric analysis through X-ray and MRI imaging. The recent advancements in AI and machine learning have opened new avenues in medical imaging, providing more accurate, efficient, and detailed analyses. The authors investigate how AI algorithms, particularly deep learning, can significantly improve the interpretation of X-ray and MRI scans for biometric purposes. The study focuses on developing an AI framework capable of identifying unique biometric markers from these imaging modalities. This approach aims to offer a more secure and reliable method for identity verification and medical diagnostics. This methodology involves training AI models with extensive datasets of X-ray and MRI images, incorporating feature extraction and pattern recognition techniques. The results demonstrate a marked improvement in accuracy and speed over traditional biometric systems.

show abstract

Multiphase Computed Tomographic Angiography with Bone Subtraction Using 3D Multichannel Convolution Neural Networks

Cited by 3 publications

References 10 publications

Computed Tomography Angiography and B‐Mode Ultrasonography under Artificial Intelligence Plaque Segmentation Algorithm in the Perforator Localization for Preparation of Free Anterolateral Femoral Flap

Computed Tomography Angiography and B‐Mode Ultrasonography under Artificial Intelligence Plaque Segmentation Algorithm in the Perforator Localization for Preparation of Free Anterolateral Femoral Flap

Segmentation of Portal Vein in Multiphase CTA Image Based on Unsupervised Domain Transfer and Pseudo Label

Revolutionizing Biometrics With AI-Enhanced X-Ray and MRI Analysis

Contact Info

Product

Resources

About