Modeling of Electronic Health Records for Time-Variant Event Learning Beyond Bio-Markers—A Case Study in Prostate Cancer

Electronic health records (EHR) of large populations constitute a vast untapped resource for data-driven diagnosis and disease progression. We develop a model capable of predicting future steps in a patient’s journey for prostate cancer (PC) and its metastases without relying on direct biomarker-measurements on a set of $18\,529$ EHR. To this end, we 1) harmonise EHR without presumptions–events are sorted and grouped by fundamental a priori principles; 2) develop a new Long-Short-Term Memory (LSTM) recurrent neural network node for learning temporal relations, on which we build an autoencoder based model; 3) derive a graph representation based on unsupervised $k$ -means clustering of events related to PC in the autoencoder’s latent layer. We report $88 {\%}$ predicting accuracy for the targeted metastasis-related events, and lower accuracies for more general events. The model gains interpretability with a graph representation illustrating the patient journey. Most importantly, we predict that $20 {\%}$ of all PC diagnosed patients will progress into metastatic disease one visit ahead of time. For the remaining patients we can predict the next step in their journey. We conclude that the model based on the new LSTM node provides a valuable tool for earlier diagnosis of life threatening metastases and quality assurance of the procedure.