The invention relates to the technical field of tumor radiotherapy, and discloses a radiotherapy parameter intelligent recommendation method based on
disease species, target area volume and
dose characteristics, and the method comprises the steps: firstly obtaining historical
helical tomography radiotherapy data containing different
disease species and
dose schemes, and carrying out the
standardization; then, non-iterative forward
dose calculation based on full-factor experimental design is executed, and treatment
simulation plans are generated in batches; then,
dosimetry and efficiency indexes are extracted, and an optimization rule base containing a response surface model and a
Pareto optimal solution set is constructed; and finally,
processing new case features by using a
decision tree model, and outputting a recommended
collimator mode, a screw
pitch, a
modulation factor value and an expected index range. According to the method, data, a model and a decision
closed loop are constructed, so that the defects that the traditional plan design depends on artificial experience, the
trial and error cost is high and multiple targets are difficult to balance are overcome, and the efficiency and the quality consistency of the radiotherapy plan design are remarkably improved.