- 3.3Impact Factor
- 6.7CiteScore
- 16 daysTime to First Decision
Journal of Imaging, Volume 8, Issue 5
May 2022 - 32 articles
Cover Story: Tumor segmentation requires a highly trained physician. To reduce human intervention, we propose a full-body 3D positron emission tomography (PET) image with two 2D projections obtained through Maximum Intensity Projections (MIPs). The two projections are then input to two 2D convolutional neural networks (CNNs) trained to classify lung vs. esophageal cancer. A weighted class activation map (CAM) is obtained for each projection, and the intersection of the two 2D orthogonal CAMs serves to detect the 3D region around the tumor. To refine the segmentation, we add a geometric loss based on prior knowledge penalizing the distance between the CAMs and a seed point provided by the user. Finally, the 3D segmentation is fed to a 3D CNN to predict the patient outcome. View this paper
- Issues are regarded as officially published after their release is announced to the table of contents alert mailing list .
- You may sign up for email alerts to receive table of contents of newly released issues.
- PDF is the official format for papers published in both, html and pdf forms. To view the papers in pdf format, click on the "PDF Full-text" link, and use the free Adobe Reader to open them.
Articles
There are no articles in this issue yet.
Get Alerted
Add your email address to receive forthcoming issues of this journal.
J. Imaging - ISSN 2313-433X