Health Discovery of Savannah, Georgia, has received US Patent No. 7,318,051, “Methods for feature selection in a learning machine.” The patent claims a method of pre-processing data that reduces the quantity of features using feature-selection methods that include recursive feature elimination; minimizing the number of non-zero parameters of the system; evaluation of cost function to identify a subset of features that are compatible with constraints imposed by the learning set; an unbalanced correlation score; and transductive feature selection. According to the patent, the method can be applied to gene expression data obtained from DNA microarrays and the output can comprise identities of a group of genes for detecting a disease or condition.
Intel has received US Patent No. 7,319,323, “Device and method using magnetic pattern on disk.” The patent claims a detection unit capable of creating nuclear magnetic resonance or electron spin resonance within a liquid sample and detecting signals from the NMR or ESR. According to the patent, the detection unit may comprise an integrated circuit, a MEMS device, a microarray, a macroarray, a multi-well plate, a microfluidic device, or a combination of devices.
Stanford University of Palo Alto, Calif., has received US Patent No. 7,318,908, “Integrated nanotube sensor.” The patent claims integrated nanotube sensors for detecting various chemical and biological molecules. According to the patent, nanotube sensor arrays may be formed using the technology and used in medical, military and biological applications.