Machine Learning to Predict Quality of Life in Colorectal Cancer Patients
Author Information
Author(s): Rhanoui Maryem, Mikram Mounia, Amazian Kamelia, Ait-Abderrahim Abderrahim, Yousfi Siham, Toughrai Imane
Primary Institution: University Claude Bernard Lyon 1, University of Lyon, France
Hypothesis
Can a multimodal machine learning framework predict quality of life indicators in colorectal cancer patients at various stages?
Conclusion
The study demonstrates that integrating clinical data and CT scan images significantly improves the accuracy of predicting quality of life indicators in colorectal cancer patients.
Supporting Evidence
- The integration of clinical data and imaging data improved the accuracy of predictions for quality of life indicators.
- Results showed a notable increase in accuracy for the Wexner score and Anorectal Ultrasound score after data integration.
- The study highlights the importance of multimodal approaches in enhancing patient care.
- Machine learning models were trained on a longitudinal dataset with multiple assessments over a year.
Takeaway
This study uses computers to help doctors understand how well colorectal cancer patients are doing after surgery by looking at different types of information together.
Methodology
The study collected clinical data and CT scan images from 100 colorectal cancer patients at multiple time points and used machine learning models to predict quality of life indicators.
Potential Biases
Potential biases may arise from the retrospective nature of the study and data completeness.
Limitations
The study's sample size is relatively small, and all patients were recruited from a single institution, which may limit the generalizability of the findings.
Participant Demographics
Patients included both genders and varied in age, with a focus on those undergoing treatment for colorectal cancer.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website