While there has been much negative press regarding the patentability of developments that rely upon software for implementation, the decision by the Commissioner of Patents that allowed Patent application number 2,333,184 (“Application ‘184”) demonstrates that software based patents are alive and well in Canada.
An artificial seed includes an encapsulated plant embryo. These embryos are produced in the laboratory. As these embryos have unknown germination potential, it is necessary to select those with suitable germination potential prior to encapsulation and formation of artificial seeds. Weyerhaeuser NR Company developed an automated process to classify these plant embryos of unknown germanation potential before they are encapsulated and ultimately formed into artificial seeds.
The claimed method used a plant embryo classification process that relied on automated analyses of plant embryos while in culture. Using digital imaging and spectral analysis, information is obtained regarding each embryo. This information is then analyzed by applying classification algorithms in order to develop a classification model which, in turn, is used for more accurate identification and sorting of plant embryos with suitable germination potential.
Manually selecting suitable embryos is an inherently subjective, labour intensive, tedious, time consuming and expensive, and it poses a major obstacle to the production of seeds on a mass scale. Problems inherent in the manual selection process are a major production bottleneck when embryos are classified visually by technicians; subjectively attempting to distinguish between subtle morphological differences that exist between plant embryos within a given plant species; and the subjective pre-judgement of features considered important during embryo selection. These problems highlight that an argument could be made that the claimed process merely used a computer to expedite a process that could otherwise be performed manually, albeit slowly.
A patent cannot be granted for an abstract idea. It is implicit in the definition of 'invention' that patentable subject matter must be something with physical existence, or something that manifests a discernible effect or change. During examination of Application ‘184, the Examiner objected to the method claims under s. 2 of the Patent Act as lacking physicality.
In reversing the patent Examiner’s conclusion and holding that the claims did meet the physicality requirement, the Commissioner noted that the embryo classification process generally needs to be conducted in an expedient manner before embryos lose viability. The Commissioner concluded that claim 1 included a number of physical essential elements that exists in combination with other claim elements and therefore, comply with s. 2 of the Patent Act.
The various physical elements that the Commissioner identified included a classification model, raw digital image data, and a digital scanner for acquiring the raw digital image data. The Commissioner stated that, in light of the identified motivations for the invention, these physical elements cannot be substituted or omitted without affecting the working of the invention – that is, the invention cannot be performed without physically acquiring digital image data and, in order to classify the embryos according to the claimed methods, as much data should be acquired as possible.
The Commissioner also indicated that the expression “classifying plant embryos” in the preamble of claim 1 involves, at least, physical separation of the embryos into distinct sub-populations, and not merely a purely abstract recognition of “good” versus “bad” embryos.
In the end, the Commissioner stated that “computer implemented data analyses are not the sole focus of the claimed invention”.
In this case, the description of the problems that the claims addressed appeared to be critical to the Commissioner’s decision. Accordingly, computer implemented inventions should include steps that have a physical existence in the real world, even if those steps are routine and easily developed once the software and the problem is conceived and the software is developed. When preparing patent applications with method claims that involve various degrees of data analysis, it is helpful to include a detailed description of the motivation and deficiencies of existing systems that led to the invention.