The superposition convolution algorithm offers a fast and accurate method of calculation in clinical radiation oncology. Robert Jacques et al takes this to the next level. In his research he begins by acknowledging the advancement of CPU usage and creation, then takes a different route by using the GPU for calculations.
Using a few novel approaches to the transport model in combination with the GPU, he is able to make calculations 34 to 98 times faster than a conventional superposition convolution algorithm.
There are many factors at play in planning—the dosimetrist's accuracy and the ability to create a plan, the doctor's time and most importantly, the patient. If the time it takes to compute dose is dramatically decreased, the dosimetrist can more easily come up with a great plan, thus alleviating the doctor's stress and ultimately the patient's time at the clinic.
