scholarly journals Automatic detection of vessel structure by deep learning using intravascular ultrasound images of the coronary arteries

PLoS ONE ◽  
2021 ◽  
Vol 16 (8) ◽  
pp. e0255577
Author(s):  
Hiroki Shinohara ◽  
Satoshi Kodera ◽  
Kota Ninomiya ◽  
Mitsuhiko Nakamoto ◽  
Susumu Katsushika ◽  
...  
Keyword(s):  
Percutaneous Coronary Intervention ◽  
Intravascular Ultrasound ◽  
Ground Truth ◽  
Coronary Intervention ◽  
Lumen Area ◽  
Test Set ◽  
Percutaneous Coronary ◽  
The Mean ◽  
Complex Lesions ◽  
Severe Calcification

Intravascular ultrasound (IVUS) is a diagnostic modality used during percutaneous coronary intervention. However, specialist skills are required to interpret IVUS images. To address this issue, we developed a new artificial intelligence (AI) program that categorizes vessel components, including calcification and stents, seen in IVUS images of complex lesions. When developing our AI using U-Net, IVUS images were taken from patients with angina pectoris and were manually segmented into the following categories: lumen area, medial plus plaque area, calcification, and stent. To evaluate our AI’s performance, we calculated the classification accuracy of vessel components in IVUS images of vessels with clinically significantly narrowed lumina (< 4 mm2) and those with severe calcification. Additionally, we assessed the correlation between lumen areas in manually-labeled ground truth images and those in AI-predicted images, the mean intersection over union (IoU) of a test set, and the recall score for detecting stent struts in each IVUS image in which a stent was present in the test set. Among 3738 labeled images, 323 were randomly selected for use as a test set. The remaining 3415 images were used for training. The classification accuracies for vessels with significantly narrowed lumina and those with severe calcification were 0.97 and 0.98, respectively. Additionally, there was a significant correlation in the lumen area between the ground truth images and the predicted images (ρ = 0.97, R2 = 0.97, p < 0.001). However, the mean IoU of the test set was 0.66 and the recall score for detecting stent struts was 0.64. Our AI program accurately classified vessels requiring treatment and vessel components, except for stents in IVUS images of complex lesions. AI may be a powerful tool for assisting in the interpretation of IVUS imaging and could promote the popularization of IVUS-guided percutaneous coronary intervention in a clinical setting.

P3587Tip detection method using the new short-tip IVUS with pull-back system which facilitates the 3D wiring technique in percutaneous coronary intervention for chronic total occlusion

2019 ◽  
Vol 40 (Supplement_1) ◽  
Author(s):  
K Tanaka ◽  
A Okamura ◽  
M Iwakura ◽  
H Nagai ◽  
A Sumiyoshi ◽  
...  
Keyword(s):  
Percutaneous Coronary Intervention ◽  
Intravascular Ultrasound ◽  
Chronic Total Occlusion ◽  
Distal Part ◽  
Detection Method ◽  
Coronary Intervention ◽  
Procedure Time ◽  
True Lumen ◽  
Total Occlusion ◽  
Percutaneous Coronary

Abstract Background The strategy of intravascular ultrasound (IVUS)-guided wiring for CTO PCI, that is, leading the second guidewire into the true lumen under observing by IVUS from subintimal space, is the last resort. We developed the angiography-based 3D wiring method. During establishment of the angiography-based 3D wiring method, we deduced that observation of the guidewire tip as well as the shaft named “The tip detection method” simplifies and facilitates 3D wiring under IVUS-guided wiring. Therefore, we produced New CTO IVUS which is the upgraded version of Navifocus WR IVUS by adding the pull-back transducer system. This pull-back system enables us to detect the tip as well as the shaft of the second guidewire in real time (tip detection method), which facilitates the 3D wiring technique under IVUS-guided wiring. Objective We evaluated the efficacy of the tip detection method during 3D wiring for CTO PCI with New CTO IVUS. Method We created a target pinpoint penetration model and performed the procedures using an experimental heartbeat model. The target (a tube with a lumen 0.6 mm in diameter) was placed in the distal part of a CTO 20 mm in length made of 2.5% agar. After the second guidewire (Conquest-12g) was advanced into the CTO lesion to within 5mm of the target using the angiography-based wiring, IVUS-guided wiring was performed by using Navifocus WR or New CTO IVUS each five times. Result The frequency of the puncture time was reduced using the new CTO IVUS compared to the Navifocus WR (1.7±0.8 vs. 28.8±23.2, p=0.17). The procedure time was significantly shorter using the new CTO IVUS compared to the Navifocus WR (103±61 vs. 459±373 seconds, p=0.04). Conclusion The tip detection method during 3D wiring with the new short tip IVUS with the pull-back system enables us to easily perform 3D wiring and will change the CTO PCI strategy.

IMMEDIATE AND 4-YEAR OUTCOME OF PATIENTS WITH ACUTE CORONARY SYNDROME UNDERGOING PERCUTANEOUS CORONARY INTERVENTION FOR LEFT MAIN CORONARY ARTERY DISEASE

2021 ◽  
pp. 8-11
Author(s):  
Saroj Mandal ◽  
Sidnath Singh ◽  
Kaushik Banerjee ◽  
Aditya Verma ◽  
Vignesh R.
Keyword(s):  
Percutaneous Coronary Intervention ◽  
Acute Coronary Syndrome ◽  
Coronary Artery ◽  
Ejection Fraction ◽  
Coronary Intervention ◽  
Left Ventricular Ejection ◽  
Coronary Syndrome ◽  
Percutaneous Coronary ◽  
The Mean

Background: The treatment of LMCAD has shifted from coronary artery bypass grafting (CABG) to Percutaneous coronary intervention (PCI). However, data on long-term outcomes of PCI for LMCA disease, especially in patients with acute coronary syndrome (ACS) remains limited and conicting. This study aims to nd the association of the immediate and 4-year mortality in ACS patients with LMCA disease treated by PCI based on ejection fractions at admission. Methods: A retrospective analytical study was conducted. Patients were divided at admission into those with reduced left ventricular ejection fraction and those with preserved ejection fraction. Results: Forty (58.8%) of the patients presented with preserved EF. The mean age of the patients was 71.6±7.1 years. The mean LVEF of the preserved group was 61.6±4.3% and signicantly higher than that of the reduced group. Age and cardiovascular risk factor prole was similar between the two groups. Patients with reduced ejection fraction had signicantly higher levels of serum creatinine and signicantly lower levels of Hb and HDL. Mean hospital stay was signicantly longer for patients with preserved EF. In-hospital deaths were also similar between the two groups. The reduced EF group had a signicantly higher allcause mortality in the 4-year follow-up period. The mean years of follow-up for all participants was 4.2±1.3 years. Conclusion: It was seen that in patients presenting with ACS and undergoing PCI due to LMCAD, LVEF at admission, singly and in in multivariate regression is an important predictor of in hospital and 4-year mortality

Sign in / Sign up

Export Citation Format

Share Document