Method localises molecules 200 times faster than existing techniques
9 Aug 2012
Isis Innovation, a company owned by Oxford University, has developed a method of rapidly and accurately estimating the location of a molecule.
Modern microscopy methods depend on precise estimation of the location of molecules to render super-resolution images.
Computation speed is also important because more than a million estimations are required for a single image. Existing estimation methods are either precise or fast, but not both.
The technique uses information ignored by other methods to be both maximally precise and computationally fast.
It is up to 200-times faster than other methods and is suitable for images with a low signal-to-noise ratio.
According to the company, such a ’closed-form’ maximum precision estimation has long been sought and will synergistically improve most other localisation technology, rendering it faster and/or more precise.
Unlike other high-precision location estimation methods, it does not rely on a pre-determined mathematical model.
Such models can be unreliable in images where signal is low relative to background noise, which is true for biological samples deep within a cell.
The underlying technology is currently the subject of a UK patent application.