Appeal 2007-2681
Application 10/680,676
can be used to classify embryos as resistant to pathogens, drought, etc.
Therefore, we agree with the Examiner that using the claimed method to
classify plant embryos according to characteristics other than likelihood to
germinate would likely require undue experimentation on the part of those
skilled in the art.
Appellants argue that “the present invention is not directed to
requiring to first identify a particular set of parameters or data that can be
always used as indicative of specific quantifiable characteristics of plant
embryos” (Br. 9). Appellants argue that “the specification clearly describes
how a single metric classification model is developed based on data acquired
from reference embryos of known quantifiable characteristics, and then is
used to classify embryos of unknown quantifiable characteristics according
to their presumed quantifiable characteristics” (id. at 12).
We do not find this argument persuasive. Since the Specification does
not disclose which morphological characteristics are associated with
pathogen resistance, drought resistance, etc., Appellants’ argument,
essentially, is that disclosing a method of training a computer to recognize
images of embryos that are most likely to germinate entitles them to a patent
on training a computer to recognize any subset of plant embryos. Then, if it
turns out that there are visually recognizable characteristics of embryos that
are pathogen resistant, drought resistant, etc., Appellants’ patent would
cover teaching computers to recognize them, too, even though Appellants
haven’t disclosed (presumably because they don’t know) what to train the
computer to look for.
11
Page: Previous 1 2 3 4 5 6 7 8 9 10 11 12 13 Next
Last modified: September 9, 2013