Dahhay Lee

Deep Learning-Based Dynamic Risk Prediction of Venous Thromboembolism for Patients With Ovarian Cancer in Real-World Settings From Electronic Health Records

PURPOSE Patients with epithelial ovarian cancer (EOC) have an elevated risk for venous thromboembolism (VTE). To assess the risk of VTE, models were developed by statistical or machine learning algorithms. However, few models have accommodated deep learning (DL) algorithms in realistic clinical settings. We aimed to develop a predictive DL model, exploiting rich information from electronic health records (EHRs), including dynamic clinical features and the presence of competing risks. METHODS We extracted EHRs of 1,268 patients diagnosed with EOC from January 2007 through December 2017 at the National Cancer Center, Korea. DL survival networks using fully connected layers, temporal attention, and recurrent neural networks were adopted and compared with multi-perceptron–based classification models. Prediction accuracy was independently validated in the data set of 423 patients newly diagnosed with EOC from January 2018 to December 2019. Personalized risk plots displaying the individual interval risk were developed. RESULTS DL-based survival networks achieved a superior area under the receiver operating characteristic curve (AUROC) between 0.95 and 0.98 while the AUROC of classification models was between 0.85 and 0.90. As clinical information benefits the prediction accuracy, the proposed dynamic survival network outperformed other survival networks for the test and validation data set with the highest time-dependent concordance index (0.974, 0.975) and lowest Brier score (0.051, 0.049) at 6 months after a cancer diagnosis. Our visualization showed that the interval risk fluctuating along with the changes in longitudinal clinical features. CONCLUSION Adaption of dynamic patient clinical features and accounting for competing risks from EHRs into the DL algorithms demonstrated VTE risk prediction with high accuracy. Our results show that this novel dynamic survival network can provide personalized risk prediction with the potential to assist risk-based clinical intervention to prevent VTE among patients with EOC.

Cause-specific mortality rate of ovarian cancer in the presence of competing risks of death: a nationwide population-based cohort study

This nationwide cohort study aimed to evaluate the cause-specific mortality (probability of death by ovarian cancer, probability of death by other causes) under the competing risks of death in women with ovarian cancer. The Korea Central Cancer Registry was searched to identify women with primary ovarian cancer diagnosed between 2006 and 2016. Epithelial ovarian cancer cases were identified using the International Classification of Diseases for Oncology 3rd edition. We estimated the cause-specific mortality according to age (<65 years, ≥65 years), stage (local, regional, and distant), and histology (serous, mucinous, endometrioid, clear cell, and others) under the competing risks framework; moreover, cumulative incidences were estimated. We included 21,446 cases. Cause-specific mortality continuously increased throughout 10 year follow-up. Compared with women aged <65 years, ovarian cancer-specific mortality (5-year, 28.9% vs. 61.9%; 10-year, 39.0% vs. 68.6%, p<0.001) and other cause mortality (5-year, 1.7% vs. 4.8%; 10-year, 2.8% vs. 8.2%, p<0.001) increased in women aged ≥65 years. This trend was consistent across all the stages and histological types. There was a substantial increase in competing risks from 1.1% in women aged <65 years to 8.0% in women aged ≥65 years in patients with early-stage (p<0.001) non-serous ovarian cancer (p<0.001). Older age at diagnosis is associated with increasing ovarian cancer-specific mortality and competing risks. Given the substantial effect of competing risks on elderly patients, there is a need for assessment tools to balance the beneficial and harmful effects to provide optimal treatment.