Conventional spin-warp MRI is a relatively slow technique. This is because to collect an N x N image we need to repeat the sequence N time to fill up every line of k-space. In this case the total scan time will be N x TR where TR is the sequence repetition time.
In order to bring down scan time one must either reduce N or reduce TR (or both). There are a number of ways this is accomplished but typically N is reduced by collecting more than one line of k-space per repetition. TR can be reduced by using a low flip angle to preserve longtitudinal magnetisation, and by either dephasing or recycling transverse magnetisation after each repetition.
There are an enormous number of variations on rapid imaging sequences available in modern MRI. The most common types are described on their own pages. More sequences will be added as I implement them on our system.