A
system and method for
joint arthroplasty includes capturing patient-specific three-dimensional joint data during a first
arthroplasty stage and utilizing a joint
phenotype classification
algorithm to predict a likelihood of a
phenotype match or transition between first and second
arthroplasty stages. The
algorithm selects at least one joint-specific metric from the pre-
cut data, transforms the metric into a
feature vector in an N-dimensional feature space, assigns the
feature vector to at least one patient-specific pre-
cut cluster trained on data from other patients, determines a pre-
cut joint
phenotype, predicts at least one patient-specific post-cut cluster and a post-cut joint phenotype. The
system compares pre-cut and post-cut phenotypes to compute a likelihood of a match or transition, and outputs the likelihood on a
user interface intraoperatively to facilitate
verification that a planned bone cut will achieve a desired clinical target, or modification of
implant parameters, surgical parameters, or soft-tissue balance, for continuing the procedure.