In simple terms the effect of ionising radiation on life may be described by a plot of damage against dose. Such characteristic curves may be made for single cells examined in vitro, for individual organs and for complete creatures. In the absence of perfect data some assumptions have to be made. The mathematically simplest solution is that the characteristic is linear, a straight line through the origin with slope alpha
damage =alpha x dose
This is called the Linear No Threshold (LNT) assumption and the consequence of adopting it is made clear in this quotation from Sir Richard Doll
Whether there is a threshold dose below which no effect is produced is still open to doubt, but on present knowledge it seems unlikely that any such threshold exists. It must, therefore, be assumed that even very small doses produce some small risk.
However, mathematical simplicity is unusual in the natural world and such an assumption cannot be justified unless uniquely required by observational data.
You can read the full article here.