Related Experiment Video
Updated: Jul 2, 2025
Quantifying Cognitive Decrements Caused by Cranial Radiotherapy
Published on: October 18, 2011
Quantitative US Delta Radiomics to Predict Radiation Response in Individuals with Head and Neck Squamous Cell
Laurentius Oscar Osapoetra1, Archya Dasgupta1, Daniel DiCenzo1
1From the Departments of Radiation Oncology (L.O.O., A.D., I.K., I.P., Z.H., W.T.T., G.J.C.), Medical Oncology (W.T.T.), and Medicine (W.T.T.), Sunnybrook Health Sciences Centre, 2075 Bayview Ave, Toronto, ON, Canada M4N 3M5; Departments of Radiation Oncology (L.O.O., A.D., I.K., I.P., Z.H., W.T.T., G.J.C.) and Medical Biophysics (G.J.C.), University of Toronto, Toronto, Canada; and Departments of Physical Sciences (L.O.O., A.D., D.D., K.F., K.Q., M.S., L.S., G.J.C.) and Evaluative Clinical Sciences (W.T.T.), Sunnybrook Research Institute, Toronto, Canada.
Quantitative US radiomics after one week of radiation therapy can predict head and neck cancer treatment response. This early prediction using machine learning may improve patient outcomes for head and neck squamous cell carcinoma.
Area of Science:
- Oncology
- Medical Imaging
- Radiomics
Background:
- Head and neck squamous cell carcinoma (HNSCC) presents a significant treatment challenge.
- Predicting treatment response early in radiation therapy (RT) is crucial for optimizing patient management.
- Quantitative ultrasound (QUS) radiomics offers a non-invasive method for analyzing imaging data.
Purpose of the Study:
- To evaluate the predictive capability of QUS radiomics parameters derived after the first week of RT for treatment response in HNSCC.
- To assess the performance of machine learning classifiers in predicting response based on early QUS radiomics data.
Main Methods:
- A prospective study involving 55 participants with bulky node-positive HNSCC undergoing curative-intent RT.
- QUS radiofrequency data from metastatic lymph nodes were acquired before and after one week of RT.
- Texture analysis using a gray-level co-occurrence matrix generated radiomics features; delta (Δ) parameters were calculated.
Main Results:
- Five Δ-parameters showed statistically significant differences between responders and non-responders (P < .05).
- A support vector machines classifier achieved 71% sensitivity, 76% specificity, and 0.77 AUC for response prediction.
- Classifier performance improved using data acquired after the first week of RT.
Conclusions:
- A QUS Δ-radiomics model utilizing early post-RT data can reasonably predict treatment response in HNSCC.
- Early QUS radiomics shows promise as a tool for computer-aided diagnosis in radiation oncology.
- This approach may facilitate timely treatment adjustments for improved HNSCC outcomes.
More Related Videos
07:29Intramucosal Inoculation of Squamous Cell Carcinoma Cells in Mice for Tumor Immune Profiling and Treatment Response Assessment
Published on: April 22, 2019
07:57Positron Emission Tomography-based Dose Painting Radiation Therapy in a Glioblastoma Rat Model using the Small Animal Radiation Research Platform
Published on: March 24, 2022