Ransomwares have become a growing threat since 2012, and the situation continues to worsen until now. In this paper, a new framework called 2entFOX is proposed in order to detect high survivable ransomwares (HSR). Because of little research in ransomware detection, this framework can be considered as one of the first frameworks in this field. We analyze Windows ransomwares’ behaviour and we tried to find appropriate features that are particular useful in detecting this type of malwares with high detection accuracy and low false positive rate. The outcome of this part of the work, was extraction of 20 features which due to two highly efficient features in this set that according to our assessments are identified and have been used for the first time in this field we could achieve an appropriate set for HSRs detection. In final stage, after proposing architecture based on Bayesian decision network, the final evaluation is done on some known ransomware samples and unknown ones based on six different scenarios. The result of this evaluations shows the high accuracy of 2entFox in detection of HSRs.