Expert system classifier for adaptive radiation therapy in prostate cancer.
Gabriele Guidi1,2, Nicola Maffei3,4, Claudio Vecchi4
1Medical Physics Department, Az. Ospedaliero-Universitaria of Modena-Policlinico, Via del Pozzo, 71, 41124, Modena, Italy. guidi.gabriele@policlinico.mo.it.
Australasian Physical & Engineering Sciences in Medicine
|March 15, 2017
Summary
A new expert system identifies prostate cancer patients needing adaptive radiation therapy by analyzing daily imaging. This machine learning tool helps detect setup errors and anatomical changes, potentially reducing toxicity and improving treatment efficacy.
Area of Science:
- Radiation Oncology
- Medical Physics
- Machine Learning in Healthcare
Background:
- Adaptive radiation therapy (ART) aims to improve treatment accuracy by adjusting plans based on observed anatomical or dosimetric changes during treatment.
- Prostate cancer radiotherapy is susceptible to dosimetric uncertainties due to organ motion (rectum, bladder) and setup variations.
- Current methods for identifying patients who would benefit from ART are often manual and time-consuming.
Purpose of the Study:
- To develop and validate a classifier-based expert system for automated identification of prostate cancer patients suitable for adaptive radiation therapy.
- To assess the system's ability to detect dosimetric uncertainties and anatomical variations using daily imaging data.
- To support clinical decision-making for individualized treatment strategies in radiation therapy.
Main Methods:
- Development of an unsupervised predictive tool using Support Vector Machine (SVM), K-means clustering, and similarity index analysis.
- Analysis of 1200 MVCT images from 38 prostate cancer patients, including automatic re-contouring, registration, and dose warping.
- Training the system on a stable dataset and applying it to a test cohort to identify dosimetrically unstable patients.
Main Results:
- The system identified two distinct macro clusters representing dosimetrically stable patient groups.
- 25% of patients in the test cohort fell outside these defined thresholds, indicating potential dosimetric instability.
- Patients identified as unstable showed significant mean volumetric changes (30% rectum, 40% bladder) during treatment.
Conclusions:
- A machine learning approach combining daily IGRT, registration, and dose warping can effectively identify patients benefiting from adaptive radiation therapy.
- The developed system provides valuable information for clinical decision-making regarding adaptive strategies, especially in treatments affected by hollow organ motion.
- This approach has the potential to reduce treatment inadequacies, minimize toxicity, and enhance the overall efficacy of radiation therapy for prostate cancer.
Related Concept Videos
02:53Adaptive Mechanisms in Cancer Cells
7.2K
Cancer cells accumulate genetic changes at an abnormally rapid rate due to the defects in the DNA repair mechanisms. From an evolutionary perspective, such genetic instability is advantageous for cancer development. Mutant cell lines accumulate a series of beneficial mutations that contribute to their progression into cancer.
Some of the advantages that cancer cells have on normal cells include - enhanced ability to divide without terminally differentiating, induce new blood vessel formation,...
Some of the advantages that cancer cells have on normal cells include - enhanced ability to divide without terminally differentiating, induce new blood vessel formation,...
7.2K
01:21Cancer Survival Analysis
807
Cancer survival analysis focuses on quantifying and interpreting the time from a key starting point, such as diagnosis or the initiation of treatment, to a specific endpoint, such as remission or death. This analysis provides critical insights into treatment effectiveness and factors that influence patient outcomes, helping to shape clinical decisions and guide prognostic evaluations. A cornerstone of oncology research, survival analysis tackles the challenges of skewed, non-normally...
807