Download full-text PDF |
Source |
|---|---|
| http://www.ncbi.nlm.nih.gov/pmc/articles/PMC11235067 | PMC |
| http://dx.doi.org/10.1590/0100-3984.2023.0116 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Development and external validation of a multi-task feature fusion network for CTV segmentation in cervical cancer radiotherapy.
Radiother Oncol
December 2024
Department of Digital Medicine, School of Biomedical Engineering and Medical Imaging, Army Medical University, Chongqing 400038, China. Electronic address:
Background And Purpose: Accurate segmentation of the clinical target volume (CTV) is essential to deliver an effective radiation dose to tumor tissues in cervical cancer radiotherapy. Also, although automated CTV segmentation can reduce oncologists' workload, challenges persist due to the microscopic spread of tumor cells undetectable in CT imaging, low-intensity contrast between organs, and inter-observer variability. This study aims to develop and validate a multi-task feature fusion network (MTF-Net) that uses distance-based information to enhance CTV segmentation accuracy.
View Article and Find Full Text PDFTumor aware recurrent inter-patient deformable image registration of computed tomography scans with lung cancer.
Med Phys
November 2024
Department of Medical Physics, Memorial Sloan Kettering Cancer Center, New York, New York, USA.
Background: Voxel-based analysis (VBA) for population level radiotherapy (RT) outcomes modeling requires topology preserving inter-patient deformable image registration (DIR) that preserves tumors on moving images while avoiding unrealistic deformations due to tumors occurring on fixed images.
Purpose: We developed a tumor-aware recurrent registration (TRACER) deep learning (DL) method and evaluated its suitability for VBA.
Methods: TRACER consists of encoder layers implemented with stacked 3D convolutional long short term memory network (3D-CLSTM) followed by decoder and spatial transform layers to compute dense deformation vector field (DVF).
Predicting intraoperative 5-ALA-induced tumor fluorescence via MRI and deep learning in gliomas with radiographic lower-grade characteristics.
J Neurooncol
February 2025
Computational NeuroSurgery (CNS) Lab, Macquarie Medical School, Macquarie University, 75 Talavera Road, Sydney, NSW, 2109, Australia.
Purpose: Lower-grade gliomas typically exhibit 5-aminolevulinic acid (5-ALA)-induced fluorescence in only 20-30% of cases, a rate that can be increased by doubling the administered dose of 5-ALA. Fluorescence can depict anaplastic foci, which can be precisely sampled to avoid undergrading. We aimed to analyze whether a deep learning model could predict intraoperative fluorescence based on preoperative magnetic resonance imaging (MRI).
View Article and Find Full Text PDFWiTUnet: A U-shaped architecture integrating CNN and Transformer for improved feature alignment and local information fusion.
Sci Rep
October 2024
College of Mechanical and Vehicle Engineering, Changchun University, Changchun, 629100, Jilin, China.
Article Synopsis
- * The study introduces WiTUnet, a new LDCT denoising method that replaces traditional skip connections with a nested, dense skip pathway to better integrate features, addressing previous limitations in the Unet architecture.
- * WiTUnet also incorporates a windowed Transformer structure to lower computational demands by processing images in smaller segments and includes a Local Image Perception Enhancement (LiPe) module that improves local feature representation, achieving better performance than existing methods in key evaluation metrics.
Transformer-Integrated Hybrid Convolutional Neural Network for Dose Prediction in Nasopharyngeal Carcinoma Radiotherapy.
J Imaging Inform Med
October 2024
College of Mechanical and Electrical Engineering, Chengdu University of Technology, Chengdu, 610059, China.
Article Synopsis
- Existing deep learning methods, particularly CNNs, often fail to capture long-distance information in data, while Transformers lack CNNs' spatial processing capabilities.
- The proposed hybrid model combines CNN and Transformer features with advanced architectural elements to improve dose prediction, showing superior results in experimental tests compared to traditional methods, which could significantly aid radiotherapy planning in clinical settings.
Want AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!