Publications by authors named "Matthew Cagley"
Article Synopsis
- Image segmentation of the liver is crucial for liver cancer treatment planning, but manual methods are impractical due to scale, leading to a shift towards deep learning models for automation.
- This study focuses on developing a generalizable deep learning model that segments the liver in T1-weighted MR images using three architectures: nnUNet, PocketNet, and Swin UNETR, with data from six different institutions totaling 819 images.
- The results show that nnUNet and PocketNet achieved high similarity scores in liver segmentation, suggesting they can effectively perform segmentation on a diverse dataset, benefiting both intra- and inter-institutional applications.
Article Synopsis
- Image segmentation is crucial for treating liver cancer, but doing it manually on a large scale is impractical, so deep learning models are increasingly used for automation.
- This study focuses on developing a deep learning model using T1w MR images to enhance liver segmentation, evaluating three different architectures with data from 819 images across six datasets.
- The nnUNet model showed strong performance, achieving over 90% in Dice-Sorensen similarity coefficients, indicating that it can effectively segment livers even with variations in imaging techniques and disease types.
Background And Aims: Limited methods exist to accurately characterize the risk of malignant progression of liver lesions. Enhancement pattern mapping (EPM) measures voxel-based root mean square deviation (RMSD) of parenchyma and the contrast-to-noise (CNR) ratio enhances in malignant lesions. This study investigates the utilization of EPM to differentiate between HCC versus cirrhotic parenchyma with and without benign lesions.
View Article and Find Full Text PDFBackground: Outcomes after definitive chemoradiation for squamous cell carcinoma are generally favorable. However, biomarkers to further yield prognostic information are desired. Treatment-related lymphopenia as well as an elevated baseline neutrophil-lymphocyte ratio have been associated with worse survival in several cancer types.
View Article and Find Full Text PDF