Summary of Automatic Quantification Of Serial Pet/ct Images For Pediatric Hodgkin Lymphoma Patients Using a Longitudinally-aware Segmentation Network, by Xin Tie et al.
Automatic Quantification of Serial PET/CT Images for Pediatric Hodgkin Lymphoma Patients Using a Longitudinally-Aware Segmentation Network
by Xin Tie, Muheon Shin, Changhee Lee, Scott B. Perlman, Zachary Huemann, Amy J. Weisman, Sharon M. Castellino, Kara M. Kelly, Kathleen M. McCarten, Adina L. Alazraki, Junjie Hu, Steve Y. Cho, Tyler J. Bradshaw
First submitted to arxiv on: 12 Apr 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Artificial Intelligence (cs.AI); Medical Physics (physics.med-ph)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The paper presents a novel deep learning model called LAS-Net that enables automatic quantification of longitudinal changes in positron emission tomography (PET) scans for pediatric Hodgkin lymphoma patients. The goal is to develop a network that can accurately detect residual disease in interim-therapy scans, which are often subtle and difficult to detect. The proposed method incorporates longitudinal cross-attention, allowing relevant features from baseline scans to inform the analysis of subsequent scans. The model performance was evaluated using Dice coefficients for baseline segmentation and detection F1 scores for interim scans. Additionally, the paper extracts and compares quantitative PET metrics such as metabolic tumor volume (MTV) and total lesion glycolysis (TLG) against physician measurements. The results show that LAS-Net outperforms other methods in detecting residual lymphoma in interim scans, with an F1 score of 0.606. The model also demonstrates high accuracy in quantifying PET metrics such as qPET, ΔSUVmax, MTV, and TLG. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper develops a new way to analyze PET scans for children with a type of cancer called Hodgkin lymphoma. The goal is to make it easier to detect if the cancer has come back after treatment. The researchers created a special kind of artificial intelligence that looks at both the original scan and follow-up scans to see if there are any changes. They tested this AI on 297 scans and found that it was much better than other methods at detecting small amounts of cancer. This is important because doctors want to be able to catch any remaining cancer early so they can start treatment again. |
Keywords
» Artificial intelligence » Cross attention » Deep learning » F1 score
