Abstract Data Analysis is difficult to patent unless some practical applications are included

A recent decision at the UKIPO (BL 0/381/17) demonstrates that applicants should be wary in the way in which they describe and claim inventions that involve data analysis.  In this decision, Agilent Technologies’ patent application was facing a “full-house” of excluded subject matter objections.  The examiner had asserted that their application was non-technical as it related to a mathematical method, presentation of information, or a computer program, which in certain circumstances are all excluded from patentability in the UK.

The application related to analysing data obtained from measurements carried out on a fluidic sample which has been separated in a separation system.  This had applications for use in the field of pharmaceuticals for quality control, for instance to identify suspicious features in a sample.

Applying the Aerotel/Macrossan four step test, the hearing officer identified the contribution of the invention to be:

“analysing measurement data of a plurality of measurements on a fluidic sample to be separated in a separation system;

determining identified and unidentified feature clusters from a plurality of  measurements, each measurement having multiple features; and

displaying the identified and unidentified feature clusters with a graphical indication of the corresponding spread of the feature clusters.”

The hearing officer then looked to the ‘AT&T signposts’ (subsequently updated in HTC v Apple) in which the Courts have provided additional guidance on what constitutes a “technical contribution”.  According to one of these signposts a patentable invention may be present if it can be shown that “the claimed technical effect has a technical effect on a process which is carried out outside a computer”.  This was clearly not the case for the above steps since they were directed merely to identifying and displaying clusters and their spread.

This application might have been patentable if it had been described differently.  It would have been particularly helpful if the applicant had described how the analysed data was actually used.  The applicant argued that “displaying the unidentified peaks to the user allowed the user, probably a chemist, to check whether all components have been identified and whether additional compounds have been detected … and subsequent action by the chemist would have a technical effect on the manufacturing process carried on outside the computer.”  However, these steps were not part of the claimed invention and, as such, the argument was futile.  In the absence of these additional steps the invention simply related to a mathematical method and a program for a computer, ultimately in the field of excluded matter.

This demonstrates a key consideration in drafting patent applications for the UK that involve data analysis.  For these cases, it is important that any steps of using the analysed data outside of the computer should also be described and claimed.  If these steps are missing then, as with this case, you may find that your patent application is destined for failure.