US Patent 8,050,870. Identifying associations using graphical models. Inventors: David Heckerman, Jonathan Carlson, Carl Kadie. Assignee: Microsoft
Describes a statistical method for identifying associations between variables in biological data. It includes a model builder that uses information about the variables to generate a set of models; and an association identifier that assesses the strength of the association between the variables in question.
US Patent 8,046,174. System and method for identifying networks of ternary relationships in complex data systems. Inventors: Ker-Chau Li. Assignee: The Regents of the University of California
Describes a method for identifying "high order associations" between two variables in complex systems, particularly where there is no correlation or the correlation is weakened by the influence of a third variable. The method can be applied to gene expression data to study how the activities of pairs of correlated genes are impacted by the influence of a third gene or genes.
US Patent 8,041,090. Method of, and apparatus and computer software for, performing image processing. Inventors: Yuriy Alexandrov, Iain Bye, Geert Kalusche. Assignee: GE Healthcare
Describes a method for processing images of biological objects. According to the abstract, the approach accepts image data from a reference and a target population of objects; obtains a set of measurements of "descriptors" that are contained in both images, and compares both sets of measurements to select a combination of descriptors that serve as a standard for defining the characteristics of the target population.