Abstract
Incidence and mortality rates of endometrial cancer are increasing, leading to increased interest in endometrial cancer risk prediction and stratification to help in screening and prevention. Previous risk models have had moderate success with the area under the curve (AUC) ranging from 0.68 to 0.77. Here we demonstrate a population-based machine learning model for endometrial cancer screening that achieves a testing AUC of 0.96.We train seven machine learning algorithms based solely on personal health data, without any genomic, imaging, biomarkers, or invasive procedures. The data come from the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial (PLCO). We further compare our machine learning model with 15 gynecologic oncologists and primary care physicians in the stratification of endometrial cancer risk for 100 women.We find a random forest model that achieves a testing AUC of 0.96 and a neural network model that achieves a testing AUC of 0.91. We test both models in risk stratification against 15 practicing physicians. Our random forest model is 2.5 times better at identifying above-average risk women with a 2-fold reduction in the false positive rate. Our neural network model is 2 times better at identifying above-average risk women with a 3-fold reduction in the false positive rate.Our machine learning models provide a non-invasive and cost-effective way to identify high-risk sub-populations who may benefit from early screening of endometrial cancer, prior to disease onset. Through statistical biopsy of personal health data, we have identified a new and effective approach for early cancer detection and prevention for individual patients.
Highlights
Endometrial cancer is the fourth most common cancer among women (Howlader et al, 2017)
We evaluated seven different algorithms: logistic regression (LR), neural network (NN), support vector machine (SVM), decision tree (DT), random forest (RF), linear discriminant analysis (LDA), and naïve Bayes (NB)
There is no significant difference in the training and testing performance for four of the algorithms (LR, NN, LDA and naive Bayes (NB)), but SVM, DT, and RF have a significant drop in performance going from training to testing
Summary
Endometrial cancer is the fourth most common cancer among women (Howlader et al, 2017). Symptoms such as bleeding or spotting often manifest early in the disease, resulting in the early detection of most cancers and a relatively high 5-years survival rate of 82% (American Cancer Society, 2017). Screening recommendations from the American Cancer Society (ACS) have remained constant since 2001 (Smith et al, 2018). For very high-risk women such as those with Lynch syndrome, a high likelihood of being a mutation carrier, or families with suspected autosomal-dominant predisposition to colon cancer, ACS recommends annual screening (Smith et al, 2001)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have