Abstract: In some examples, a system may generate a plurality of care path patient profile models based on a plurality of care path patterns for a plurality of past patient admissions. For example, each care path patient profile model may include a trained classifier. Further, the system may receive information related to a new patient admission, and may generate features from the received information related to the new patient admission. The system may input the features generated from the received information related to the new patient admission into the plurality of care path patient profile models to obtain a respective probability of being classified in a respective care path based on an amount of similarity to the patients who have gone through each care path. In addition, the system may present, on a display, information related to at least one care path pattern in a graphical user interface.

Abstract: In some examples, a computing device may receive hierarchical data having a hierarchical structure including a plurality of levels. The computing device may determine a plurality of features based at least in part on the hierarchical data, and may select a subset of the features at a first level as candidates for consolidating to a next higher level in the hierarchical structure. The computing device may determine that a predicted loss of information from consolidating the subset of features is less than a threshold, and may revise the hierarchical structure to include a consolidated feature at the next higher level, rather than the subset of features. In some examples, a statistical model may be trained based on the revised hierarchical structure and used at least partially to make a determination, send a notification, and/or control a device.

Abstract: In some examples, a system may generate a plurality of care path patient profile models based on a plurality of care path patterns for a plurality of past patient admissions. For example, each care path patient profile model may include a trained classifier. Further, the system may receive information related to a new patient admission, and may generate features from the received information related to the new patient admission. The system may input the features generated from the received information related to the new patient admission into the plurality of care path patient profile models to obtain a respective probability of being classified in a respective care path based on an amount of similarity to the patients who have gone through each care path. In addition, the system may present, on a display, information related to at least one care path pattern in a graphical user interface.

Abstract: In some examples, a system may train a false positive reduction machine learning model (MLM) for nodule detection. The system may receive training data images including negative images and positive images, along with an indication of nodule locations in the positive images. The system may determine elliptical approximations for nodules in the positive images, and may determine respective binarized contours from the elliptical approximations. Further, the system may determine an elliptical approximation space for the binarized contours, and may determine a subspace angle between individual image samples in the positive images and the elliptical approximation space as at least one feature of the MLM. Subsequently, when applying the MLM during nodule detection, one or more images may be input to the MLM to determine whether an indication of a nodule is correct, and if so, a visualization of a location of the nodule may be provided.

Abstract: In some examples, a computing device may receive sensor data associated with a plurality of patient devices that are associated with a plurality of patients. The computing device may further receive caregiver records corresponding at least partially to the sensor data. At least two groups of indicators may be determined from the caregiver records, such a based on a selected subject. Further, the computing device may determine a plurality of clusters from the sensor data. Based on the plurality of clusters and the at least two groups, the computing device may determine an indication of a discrepancy in care for a patient of the plurality of patients. Based on the indication of the discrepancy, the computing device may send at least one of a notification to a caregiver computing device, a notification to a monitoring computing device, or a control signal to one of the patient devices.

Abstract: In some examples, a computing device may receive sensor data for a target and at least one of: log data for the target, or historical log data and historical sensor data for a plurality of other targets. The computing device may determine at least one event classified as a non-uniform event in at least one of the log data or the historical log data, and may determine combined features, such as a feature vector, based on the sensor data and the non-uniform event(s). The computing device may determine an analysis result from the combined features. Further, based on the analysis result, the computing device may send a control signal to a device associated with the target for controlling the device, and/or may send a communication related to the target to another computing device.

Abstract: In some examples, a system may train a false positive reduction machine learning model (MLM) for nodule detection. The system may receive training data images including negative images and positive images, along with an indication of nodule locations in the positive images. The system may determine elliptical approximations for nodules in the positive images, and may determine respective binarized contours from the elliptical approximations. Further, the system may determine an elliptical approximation space for the binarized contours, and may determine a subspace angle between individual image samples in the positive images and the elliptical approximation space as at least one feature of the MLM. Subsequently, when applying the MLM during nodule detection, one or more images may be input to the MLM to determine whether an indication of a nodule is correct, and if so, a visualization of a location of the nodule may be provided.

Abstract: In some examples, a computing device may receive sensor data for a target and at least one of: log data for the target, or historical log data and historical sensor data for a plurality of other targets. The computing device may determine at least one event classified as a non-uniform event in at least one of the log data or the historical log data, and may determine combined features, such as a feature vector, based on the sensor data and the non-uniform event(s). The computing device may determine an analysis result from the combined features. Further, based on the analysis result, the computing device may send a control signal to a device associated with the target for controlling the device, and/or may send a communication related to the target to another computing device.

Abstract: In some examples, a computing device may receive sensor data associated with a plurality of patient devices that are associated with a plurality of patients. The computing device may further receive caregiver records corresponding at least partially to the sensor data. At least two groups of indicators may be determined from the caregiver records, such a based on a selected subject. Further, the computing device may determine a plurality of clusters from the sensor data. Based on the plurality of clusters and the at least two groups, the computing device may determine an indication of a discrepancy in care for a patient of the plurality of patients. Based on the indication of the discrepancy, the computing device may send at least one of a notification to a caregiver computing device, a notification to a monitoring computing device, or a control signal to one of the patient devices.

Abstract: In some examples, a computing device may receive hierarchical data having a hierarchical structure including a plurality of levels. The computing device may determine a plurality of features based at least in part on the hierarchical data, and may select a subset of the features at a first level as candidates for consolidating to a next higher level in the hierarchical structure. The computing device may determine that a predicted loss of information from consolidating the subset of features is less than a threshold, and may revise the hierarchical structure to include a consolidated feature at the next higher level, rather than the subset of features. In some examples, a statistical model may be trained based on the revised hierarchical structure and used at least partially to make a determination, send a notification, and/or control a device.

Abstract: One embodiment is a method that builds a model of multi-dimensional sequence data in real-time with cuboids that aggregate the multi-dimensional sequence data over both patterns and dimensions. The model provides search results for a query.

Abstract: A method of processing a stream of raw data from a plurality of distributed data producing devices includes reducing the raw data to a plurality of representative synopsis coefficients, organizing the synopsis coefficients into a data structure with at least three dimensions, including a time window dimension and an accuracy dimension. Responsive to a detected anomaly in the data structure, at least one of a predetermined autonomous action and an action directed by a user is performed.

Abstract: There is provided a computer-implemented method of determining an execution ordering. An exemplary method comprises generating a directed graph based on a hierarchy. The hierarchy includes a plurality of pattern queries. The method also includes determining a minimum spanning tree of the directed graph. The method further includes determining an execution order of the pattern queries based on the minimum spanning tree.

Abstract: A query scheduler orders queries in a queue. Each query is executed based on its position in the queue. When a new query is received, the new query is inserted in the queue. A position in the queue for inserting the new query is determined based on a stretch metric for each query in the queue.

Abstract: There is provided a computer-implemented method of determining an execution ordering. An exemplary method comprises generating a directed graph based on a hierarchy. The hierarchy includes a plurality of pattern queries. The method also includes determining a minimum spanning tree of the directed graph. The method further includes determining an execution order of the pattern queries based on the minimum spanning tree.

Abstract: A method of evaluating nested complex sequence pattern queries includes obtaining events from an event stream and evaluating the events within a first window using an outer query to produce outer partial results. The method also includes determining a more stringent window constraint, the more stringent window constraint comprising a subset of the window constraint corresponding to events that produces the outer partial results and passing the more stringent window constraint to an inner query nested within the outer query. A complex event processing system is also provided.

Abstract: An exemplary embodiment of the present techniques may detect and correlate events from moving object sensor data by receiving data from a sensor. The data received from the sensor may be mapped, and events may be detected based on the mapped sensor data. Events from the mapped sensor data may be correlated online.

Abstract: New data points are added to a streaming window of data points and existing data points are removed from the window over time. Each data point has a value for each of one or more dimensions. Each time a given new data point is added to the window or a given existing data point is removed from the window, one or more outlier detection data structures are updated. Each outlier detection data structure encompasses the data points within the streaming window for a corresponding dimension. The outlier detection data structures are used to detect outlier data points within the window over selected one or more dimensions.

Abstract: Processing batch database workload while avoiding overload. A method for efficiently processing a database workload in a computer system comprises receiving the workload, which comprises a batch of queries directed toward the database. Each query within the batch of queries is assigned a priority. Resources of the computer system are assigned in accordance with the priority. The batch of queries is executed in unison within the computer system in accordance with the priority of each query thereby resolving a conflict within the batch of queries for the resources of the computer system, hence efficiently processing the database workload and avoiding overload of the computer system.

Abstract: A mixed workload management system and associated operating method modify a shortest job first (SJF) by service levels. The workload management system comprises a scheduler configured for scheduling mixed workloads. The scheduler comprises an analyzer that determines query execution time, assigns scheduling priority to a query in order inverse to the query execution time, weights the assigned scheduling priority by service level of the query, and sorts a list of queries in order of weighted scheduling priority. A schedule controller selects a query for execution from head of the sorted list of queries.