Purely medical cancer screening methods are often costly, time-consuming, and weakly applicable on a large scale. AdvancedArtificial Intelligence (AI) methods greatly help cancer detection but require specific or deep medical data. These aspects affectthe mass implementation of cancer screening methods. For these reasons, it is a disruptive change for healthcare to apply AImethods for mass personalized assessment of the cancer risk among patients based on the existing Electronic Health Records(EHR) volume. This paper presents a novel method for mass cancer risk prediction using EHR data. Among other methods, ourone stands out by the minimum data greedy policy, requiring only a history of medical service codes and diagnoses from EHR.We formulate the problem as a binary classification. This dataset contains 175 441 de-identified patients (2 861 diagnosed withcancer). As a baseline, we implement a solution based on a recurrent neural network (RNN) and pre-trained BERT. We proposea method that combines machine learning and survival analysis since these approaches are less computationally heavy, canbe combined into an ensemble (the Survival Ensemble), and can be reproduced in most medical institutions. We test theSurvival Ensemble in some studies. Firstly, we obtain a significant difference between values of the primary metric (AveragePrecision) with 22.8%±2.7% (ROC AUC 83.7%±1.7%, F1 17.8%±2.8%) for the Survival Ensemble versus 15.1%±2.6% (ROCAUC 84.9%±0.8%, F1 21.4%±3.1%) for the Baseline method. Secondly, the performance of the Survival Ensemble is alsoconfirmed during the ablation study. Thirdly, our method exceeds age baselines by a significant margin. Fourthly, in the blindretrospective out-of-time experiment, the proposed method is reliable in cancer patient detection (9 out of 100 selected). Suchresults exceed the estimates of medical screenings, e.g., the best Number Needed to Screen (9 out of 1000 screenings).