Abstract
Objective: (1) Demonstrate how machine learning can be used for
prediction modeling by predicting the treatment patients with T1-2,
N0-N1 Oropharyngeal Squamous Cell Carcinoma receive. (2) Assess
disparities in the treatment of this population. Design: Retrospective
cohort. The data was split into 80/20 distribution for training and
testing, respectively. Machine learning algorithms were explored for
development. Area Under the Curve, accuracy, precision, and recall were
calculated for the final model. The permutation feature scores highlight
significant variables within the model. Setting: National Cancer
Database. Participants: Adults diagnosed with T1-2, N0-N1 Oropharyngeal
Squamous Cell Carcinoma from 2004 to 2013 were eligible Main Outcome
Measure: Primary treatment modality Results: Among the 19,111 patients
in the study, the mean (standard deviation) age was 61.3 (10.8) years,
14,034 (73%) were male, and 17,292 (91%) were white. Surgery was the
primary treatment in 9,533 (50%) cases, and radiation in 9,578 (50%)
cases. The final model yielded an Area Under the Curve of 78% (95% CI,
77% to 79%), accuracy of 71%, precision of 72%, and recall of 71%.
The T-stage, primary site, N-stage, grade, and type of treatment
facility were impactful variables included in the model. Conclusion:
Machine learning was used to predict primary treatment modality for
T1-2, N0-N1 Oropharyngeal Squamous Cell Carcinoma. This study
demonstrates how machine learning can be used for prediction modeling.
The results also suggest treatment is influenced by clinical staging and
type of treatment facility.