Biodosimetry is a key diagnostic tool for radiation exposure, risk assessment and treatment planning of acute radiation sickness. To effectively respond to a large-scale radiological incident, there is a need for the development of biodosimetric methods with fast, portable, and convenient operating advantages. We employed the recombinase polymerase amplification specific high-sensitivity enzymatic reporter unlocking (RPA-SHERLOCK) technology to establish a method for fast radiation dose assessment by measuring the expression level of radiation-inducible genes. Moreover, we proposed for the first time the principle of quantitative detection of curve slopes based on this method. Using this new method, changes in mRNA expression were confirmed in a number of radiation-sensitive genes (XPC, CDKN1A, and ATM) in human lymphocytes after irradiation. The standard curve of the dose-effect relationship was established, which can be used to quickly determine the exposed dose of the irradiated samples. Compared with traditional detection methods such as RT-qPCR, this method was found to be more convenient, fast and easy to operate. With the same amount of template input as RT-qPCR, the detection time of this method can be shortened to less than 20 min. The detection instrument required by this method is also more portable than a qPCR system.