Method and apparatus for assay analysis

A method of analysing a sample assay, the method comprising the steps of:

• (A) locating the multiple sample assay in an inspection station,
• (B) imaging light emitted from samples onto a plural array of addressable photosensitive detector elements so that light emitted from a sample impinges on a unique group of the photosensitive elements to produce a change in the electrical characteristics thereof so that a change attributable to the group is associated with that sample, which change is permanent at least until the elements have been addressed and any change in their characteristics has been read out as an item of information,
• (C) addressing the elements at regularly occurring intervals of time to generate the said items of information,
• (D) converting each item of information into a numerical value,
• (D) storing the numerical values in such a manner that all the values which relate to any one of the samples are linked,
• (F) generating and storing a sample identifier for each set of linked values,
• (G) reading out the linked values relating to a sample and computing the arithmetic mean of the linked numerical values, and
• (H) supplying as output information the numerical value of the computed arithmetic mean of the linked values, together with the sample identifier therefor, in which a first calibration step is performed in a series of tests on the groups of detector elements using different samples of known activities, the observed numerical output information from the groups of detector elements in response to the different known activities being logged against the known activity in a first calibration memory so as to provide a look-up table for converting future numerical output information to sample activity;
• and in which a second calibration step is performed to allow correction for colour quench, the second calibration step comprising performing a large number of exposures of at least part of the array of photosensitive detector elements to samples of constant known activity and having different known dye concentrations, computing from the exposure results the mean and one or more characteristics of the statistical distribution of the amount of light detection per exposure, and computing from said mean and one or more characteristics an index for each known sample, so as to enable look-up table for correction for colour quench correction to be compiled.