Abstract: A system comprises a toolset to render a protocol, such as an industry clinical treatment guideline, into a process map. The process map includes a workflow having a time-based series of steps determined from the protocol. The system performs recursive matching to match events in electronic medical records (EMRs) to nodes in threads in the protocol map to determine protocol compliance. The EMRs include unstructured data, and specialized query objects may be generated to retrieve relevant data for the process map from the unstructured EMR data.

Abstract: A patient anonymizing system includes a plurality of hashing appliances and data sources. Each appliance receives medical records containing at least confidential protected health information (PHI). A salt value is appended to each confidential PHI, and a hash is generated, which replaces the confidential PHI to generate an anonymized record. A master patient index server aggregates the anonymized records. A vector and cluster matching engine determines if the anonymized record matches a unique patient identifier corresponding to a second anonymized record. A comparison vector is generated by comparing hash values of the confidential PHI with hash values in the second anonymized record, and is crossed with a confidence vector having weights based on match conditions. This produces a match confidence level, which is compared to a threshold. If the threshold is met, the anonymized record is mapped to the unique patient identifier associated with the second record.

Abstract: A system comprises a toolset to render a protocol, such as an industry clinical treatment guideline, into a process map. The process map includes a workflow having a time-based series of steps determined from the protocol. The system performs recursive matching to match events in electronic medical records (EMRs) to nodes in threads in the protocol map to determine protocol compliance. The EMRs include unstructured data, and specialized query objects may be generated to retrieve relevant data for the process map from the unstructured EMR data.

Abstract: A clinical quality analytics system may include a data storage to store electronic medical record (EMR) data. The system may map events from the EMR data to a process map through a recursive matching process. The mapping may include recursively matching the events to nodes in threads in a map based on event times and thread times. One of the recursions may be selected as a best fit based on metrics determined for the recursions.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for processing a data model, a data file and user input to provide a populated data model, the populated data model including: a data content table having one or more navigation columns and one or more content columns, at least one navigation column including one or more key values of a key node, and at least one content column including data values of a plurality of data values, and a foreign key table having one or more navigation columns and one or more relationship columns, at least one navigation column including one or more key values of a foreign key node, and the one or more relationship columns associating at least one key node of the data content table to a respective foreign key node.

Abstract: Methods, systems, and apparatuses, including computer programs encoded on a computer storage medium, can be implemented to perform actions including receiving input data defining a predictive model, the predictive model including multiple features. The actions further include weighting the predictive model iteratively for each feature, using actual data including values for each feature for multiple entities within a population, by iteratively adjusting a current weight for the feature by a momentum until the momentum equals zero, the momentum being iteratively adjusted by a momentum factor based on whether a model score improves, the model score being calculated based on the actual data. The actions further include calculating a value score for the entity using the weighted predictive model.

Abstract: A patient anonymizing system includes a plurality of hashing appliances and data sources. Each appliance receives medical records containing at least confidential protected health information (PHI). A salt value is appended to each confidential PHI, and a hash is generated, which replaces the confidential PHI to generate an anonymized record. A master patient index server aggregates the anonymized records. A vector and cluster matching engine determines if the anonymized record matches a unique patient identifier corresponding to a second anonymized record. A comparison vector is generated by comparing hash values of the confidential PHI with hash values in the second anonymized record, and is crossed with a confidence vector having weights based on match conditions. This produces a match confidence level, which is compared to a threshold. If the threshold is met, the anonymized record is mapped to the unique patient identifier associated with the second record.

Abstract: A clinical quality analytics system may include a data storage to store electronic medical record (EMR) data. The system may map events from the EMR data to a process map through a recursive matching process. The mapping may include recursively matching the events to nodes in threads in a map based on event times and thread times. One of the recursions may be selected as a best fit based on metrics determined for the recursions.

Abstract: A computer program product and associated algorithm for sorting S sequences of binary bits. The S sequences may be integers, floating point numbers, or character strings. The algorithm is executed by a processor of a computer system. Each sequence includes contiguous fields of bits. The algorithm executes program code at nodes of a linked execution structure in a sequential order with respect to the nodes. The algorithm executes a masking of the contiguous fields of the S sequences in accordance with a mask whose content is keyed to the field being masked. The sequential order of execution of the nodes is a function of an ordering of masking results of the masking. Each sequence, or a pointer to each sequence, is outputted to an array in the memory device whenever the masking places the sequence in a leaf node of the nodal linked execution structure.

Abstract: A method and associated algorithm for sorting S sequences of binary bits. The S sequences may be integers, floating point numbers, or character strings. The algorithm is executed by a processor of a computer system. Each sequence includes contiguous fields of bits. The algorithm executes program code at nodes of a linked execution structure in a sequential order with respect to the nodes. The algorithm executes a masking of the contiguous fields of the S sequences in accordance with a mask whose content is keyed to the field being masked. The sequential order of execution of the nodes is a function of an ordering of masking results of the masking. Each sequence, or a pointer to each sequence, is outputted to an array in the memory device whenever the masking places the sequence in a leaf node of the nodal linked execution structure.

Abstract: A method and associated algorithm for in-place sorting S sequences of binary bits stored contiguously in an array within a memory device of a computer system prior to the sorting. Each sequence includes contiguous fields of bits. The algorithm is executed by a processor of a computer system. The in-place sorting executes program code at each node of a linked execution structure. Each node includes a segment of the array. The program code is executed in a hierarchical sequence with respect to the nodes. Executing program code at each node includes: dividing the segment of the node into groups of sequences based on a mask field having a mask width, wherein each group has a unique mask value of the mask field; and in-place rearranging the sequences in the segment, wherein the rearranging results in each group including only those sequences having the unique mask value of the group.

Abstract: A method and associated algorithm for sorting S sequences of binary bits. The S sequences may be integers, floating point numbers, or character strings. The algorithm is executed by a processor of a computer system. Each sequence includes contiguous fields of bits. The algorithm executes program code at nodes of a linked execution structure in a sequential order with respect to the nodes. The algorithm executes a masking of the contiguous fields of the S sequences in accordance with a mask whose content is keyed to the field being masked. The sequential order of execution of the nodes is a function of an ordering of masking results of the masking. Each sequence, or a pointer to each sequence, is outputted to an array in the memory device whenever the masking places the sequence in a leaf node of the nodal linked execution structure.

Abstract: A state machine representation is provided that includes a plurality of nodes that are assigned a unique node identifier and an output value, and that are grouped into color segments. The nodes are interconnected by transitional vectors, with each transitional vector being assigned a unique value. Further, at least one of the nodes has a termination vector that leads to a termination point. To encrypt the set of data values, a start node is identified and any path of transitional vectors is be traversed from node to node. As each transitional vector is traversed, its corresponding unique value will be recorded in sequence. When a node is reached that has an output value that matches one of the set of data values, an invalid unique value is designated and recorded in sequence with the unique values.