Related Experiment Video
Updated: Dec 31, 2025
Analysis of Cerebral Vasospasm in a Murine Model of Subarachnoid Hemorrhage with High Frequency Transcranial Duplex Ultrasound
Published on: June 3, 2021
Localization of common carotid artery transverse section in B-mode ultrasound images using faster RCNN: a deep
Pankaj K Jain1, Saurabh Gupta2, Arnav Bhavsar3
1Indian Institute of Technology Varanasi, Banaras Hindu University, Varanasi, UP, India.
Insights
A new deep learning method accurately locates the common carotid artery (CCA) in ultrasound images. This fast region convolutional neural network (FRCNN) approach aids in real-time cardiac health assessments by automating artery localization.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Cardiovascular Diagnostics
Background:
- Accurate localization of the common carotid artery (CCA) in B-mode ultrasound images is crucial for assessing cardiac health parameters like artery stiffness, lumen diameter (LD), and carotid intima-media thickness (cIMT).
- Manual localization methods are time-consuming and prone to errors, hindering real-time analysis.
Purpose of the Study:
- To develop and evaluate a novel, automated method for the real-time localization of the common carotid artery (CCA) transverse section in B-mode ultrasound images.
- To improve the efficiency and objectivity of CCA localization for subsequent cardiac health parameter measurements.
Main Methods:
- A fast region convolutional neural network (FRCNN) based localization method was designed using convolutional, fully connected, and pooling layers.
- The FRCNN model incorporates a region proposal network (RPN) and an object class detection network (OCDN) to generate bounding boxes with prediction scores for the CCA cross-section.
- A B-mode ultrasound image database of CCA was partitioned using K=2, 5, and 10, with training extended up to 2000 epochs.
Main Results:
- The FRCNN model achieved 95% validation accuracy after 2000 epochs.
- The mean accuracy across K=10 partitions (90% training, 10% testing) up to 2000 epochs was 89.36%.
- Testing on an independent dataset of 433 images yielded an accuracy of 87.99%.
Conclusions:
- The proposed FRCNN-based method demonstrates a promising and accurate approach for automated localization of the carotid artery transverse section.
- This deep learning technique offers an objective and near real-time solution for CCA localization, facilitating timely cardiac health assessments.
Abstract:
Cardiologists can acquire important information related to patients' cardiac health using carotid artery stiffness, its lumen diameter (LD), and its carotid intima-media thickness (cIMT). The sonographers primarily concern about the location of the artery in B-mode ultrasound images. Localization using manual methods is tedious and time-consuming and also may lead to some errors. On the other hand, automated approaches are more objective and can provide the localization of the artery at near real time. Above arterial parameters may be determined after localization of the artery in real time.A novel method of localization of common carotid artery (CCA) transverse section is presented in this work. The method is known as fast region convolutional neural network (FRCNN)-based localization method and is designed using a stack of three layers viz. convolutional layers, fully connected layers, and pooling layers. These organized layers constitute a region proposal network (RPN) and an object class detection network (OCDN). We obtain an outcome as a bounding box along with a score of prediction around the cross-section of the CCA.B-mode ultrasound image database of CCA is split into training and testing set, to accomplish this, three partition methods K = 2, 5, and 10 are used in our work. The training is extended for 30, 200, and 2000 epochs in order to achieve fine-tuned features from the convolutional neural network. After 2000 epochs, we obtain 95% validation accuracy; however, mean of the accuracies up to 2000 epochs is 89.36% for K = 10 partitions protocol (training 90%, testing 10%). Generated CNN model is tested on a different dataset of 433 images and the acquired accuracy is 87.99%. Thus, the proposed method including an advanced deep learning technique demonstrates promising localization for carotid artery transverse section. Graphical abstract.
