Virtual digital subtraction angiography using multizone patch-based U-Net
- PDF / 1,434,300 Bytes
- 11 Pages / 595.276 x 790.866 pts Page_size
- 40 Downloads / 189 Views
SCIENTIFIC PAPER
Virtual digital subtraction angiography using multizone patch‑based U‑Net Ryusei Kimura1 · Atsushi Teramoto1 · Tomoyuki Ohno2 · Kuniaki Saito1 · Hiroshi Fujita3 Received: 13 April 2020 / Accepted: 25 September 2020 © Australasian College of Physical Scientists and Engineers in Medicine 2020
Abstract Digital subtraction angiography (DSA) is a powerful technique for visualizing blood vessels from X-ray images. However, the subtraction images obtained with this technique suffer from artifacts caused by patient motion. To avoid these artifacts, a new method called “Virtual DSA” is proposed, which generates DSA images directly from a single live image without using a mask image. The proposed Virtual DSA method was developed using the U-Net deep learning architecture. In the proposed method, a virtual DSA image only containing the extracted blood vessels was generated by inputting a single live image into U-Net. To extract the blood vessels more accurately, U-Net operates on each small area via a patch-based process. In addition, a different network was used for each zone to use the local information. The evaluation of the live images of the head confirmed accurate blood vessel extraction without artifacts in the virtual DSA image generated with the proposed method. In this study, the NMSE, PSNR, and SSIM indices were 8.58%, 33.86 dB, and 0.829, respectively. These results indicate that the proposed method can visualize blood vessels without motion artifacts from a single live image. Keywords Digital subtraction angiography · Blood vessel extraction · Deep learning · U-Net · Angiogram · Registration
Introduction Background In modern medicine, digital subtraction angiography (DSA) is a powerful technique for visualizing blood vessels from X-ray images [1, 2]. This technique plays an important role in the diagnosis and treatment of vascular diseases, such as cerebral thrombosis and coronary heart disease. DSA removes unnecessary components such as bones and organs from an X-ray fluoroscopic image, while enhancing the blood vessels in the images. Image acquisition in DSA involves subtraction of the mask images from an image obtained during the passage of the contrast material through * Atsushi Teramoto teramoto@fujita‑hu.ac.jp 1
Graduate School of Health Sciences, Fujita Health University, 1‑98 Dengakugakubo, Kutsukake‑cho, Toyoake‑city, Aichi 470‑1192, Japan
2
Fujita Health University Bantane Hospital, 3‑6‑10 Otobashi Nakagawa‑ku, Nagoya‑city, Aichi 454‑8509, Japan
3
Faculty of Engineering, Gifu University, 1‑1 Yanagido, Gifu‑city, Gifu 501‑1194, Japan
the imaging region (referred to as the live or contrast image). All images are generated in real time using a computer [3]. Ideally, the resulting subtraction images should only represent a high-contrast visualization of the blood vessels. However, the time lag between the live and mask images presents a challenge in DSA, resulting in the misalignment of images due to patient movement. Some patient movement is inevitable, including breat
Data Loading...
