Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for generating a predictive system that obtains and processes data describing terms for different medical concepts to generate commands from a user query. An entity module of the system determines whether a term describes a medical entity associated with a healthcare condition affecting an individual. When the term describes the medical entity an encoding module links the medical entity with a specified category based on an encoding scheme. The system receives the user query. A parsing engine of the system uses the received query to generate a machine-readable command by parsing the query against terms that describe the medical entity and based on the encoding scheme for linking the medical entity to the specified category. The system uses the command to query different databases to obtain data for generating a response to the received query.

Abstract: A computer-assisted method including obtaining healthcare records from multiple different data sources that each provide information regarding a corresponding type of healthcare events, identifying healthcare records from the multiple different data sources that are for a healthcare event associated with a particular individual and that occurred during a particular period of time, and generating a composite record for the particular individual for the particular period of time, and storing the composite record in a database of composite records. The composite record include an identifier for the particular individual, a pharmaceutical transactions array, where each entry in the pharmaceutical transactions array represents a pharmaceutical transaction that occurred during the particular period of time, and a medical visit array, where each entry in the medical visit array represents a medical visit that occurred during the particular period of time.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for responding to a query. In some implementations, a computer obtains a query. The computer determines a meaning for each term in the query. The computer determines user data for the user that submitted the query. The computer identifies one or more ontologies based on the meanings for at least some of the terms. The computer identifies a knowledge graph based on the identified ontologies and the user data. The computer generates a response to the query by traversing a path of the identified knowledge graph to identify items in the knowledge graph based on the determined meaning for each of the terms. The computer generates path data that represents the path taken by the computer through the identified knowledge graph. The computer provides the generated response and the path data to the client device.

Abstract: A classification code parser and method can include: reading a classification code having a description; reading a required keyword, and a total number of keywords associated with the classification code; reading text of a note; tokenizing the text of the note to create a note token stream, the note token stream having a note token and a position of the note token within the note token stream; creating a keyword map including a total number of matched keywords; determining a match ratio from the total number of the matched keywords and the total number of the keywords; determining a proximity factor based on a shortest span of tokens within the note token stream containing all the matched keywords; and determining a strength of a match between the classification code and the note based on the match ratio being multiplied by the proximity factor.

Abstract: Cohort definition and selection system for a computer having a memory, a central processing unit and a display, the system including: a cohort definition module to configure the memory according to a phenotype vector. The phenotype vector includes a patient ID to uniquely associate the phenotype vector to a patient, a plurality of demographic dimension fields, each demographic dimension field to describe a respective demographic aspect of the patient, a calculated dimension field to describe a calculated information related to the patient, a plurality of phenotype-based dimension fields, each phenotype-based dimension field to indicate relevance of the respective phenotype-based dimension field to the patient, and a child phenotype vector to recursively define a phenotype-based dimension field, and a cohort selection module to select a set of phenotype vectors that are within a predetermined error from a cohort selection criteria.

Abstract: A computer-implemented tool for automated classification and interpretation of documents, such as life science documents supporting clinical trials, is configured to perform a combination of raw text, document construct, and image analyses to enhance classification accuracy by enabling a more comprehensive machine-based understanding of document content. The combination of analyses provides context for classification by leveraging relative spatial relationships among text and image elements, identifying characteristics and formatting of elements, and extracting additional metadata from the documents as compared to conventional automated classification tools, wherein natural language processing (NLP) is applied to associate text with tokens, and relevant differences and similarities between protocols are identified.

Abstract: A parser is deployed early in a machine learning pipeline to read raw data and collect useful statistics about the raw data's content to determine which items of raw data exhibit a proxy for feature importance for the machine learning model. The parser operates at high speeds that approach the disk's absolute throughput while utilizing a small memory footprint. Utilization of the parser enables the machine learning pipeline to receive a fraction of the total raw data that would otherwise be available. Several scans through the data are performed, by which proxies for feature importance are indicated and irrelevant features may be discarded and thereby not forwarded to the machine learning pipeline. This reduces the amount of memory and other hardware resources used at the server and also expedites the machine learning process.

Abstract: Implementations provide a method to consolidate data records of regimens for treating oncology conditions. The method includes: accessing data records each encoding multi-tier data characteristics of a regimen for treating a particular oncology condition; receiving a first data record encoding a first regimen specific to a first healthcare provider institution; parsing the first data record according to a hierarchy of the encoded multi-tier data characteristics; distributing a respective weight to each of the encoded data characteristics to account for the potentially missing data characteristic; comparing data characteristics of the first data record with data characteristics from the data records by applying the respective weight to each data characteristic at a particular tier of the hierarchy such that a respective compound score is generated for each data record; and based on the compound scores for all data records, determining a prevailing data record of regimen as matching the first data record.

Abstract: Aspects of the subject matter described in this specification are embodied in systems and methods that utilize machine-learning techniques to evaluate clinical trial data using one or more learning models trained to identify anomalies representing adverse events associated with a clinical trial investigation. In some implementations, investigation data collected at a clinical trial site is obtained. A set of models corresponding to the clinical trial site is selected. Each model included in the set of models is trained to identify, based on historical investigation data collected at the clinical trial site, a distinct set of one or more indicators that indicate a compliance risk associated with the investigation data. A score for the clinical trial site is determined based on the investigation data relative to the historical investigation data. The score represents a likelihood that the investigation data is associated with at least one indicator representing the compliance risk.

Abstract: An indication that an occupant has entered a vehicle is detected on a computing device. The occupant is identified based on the indication that the occupant has entered the vehicle. A health record that is associated with the occupant is accessed by the computing device. The computing device enables the accessed health record associated with the occupant to be available from a secure digital storage container. The health record associated with the occupant is provided to a second computing device registered to a first responder from the secure digital storage container based on detecting an indication that the vehicle has had a collision.

Abstract: A GxP (good practice) platform is implemented to enable artificial intelligence (AI) algorithms to be tracked from creation through training and into production. Deployed algorithms are assigned a GxP chain ID that enables identification of production details associated with respective algorithms. Trained algorithms, each of which are respectively associated with a GxP chain ID, are containerized and can be utilized through an application programing interface (API) to provide a service. The GxP chain ID is linked to production details stored within a database, in which the production details can include information such as data used to train the algorithm, a history version, a date/time stamp when the algorithm was validated, software and hardware on which the algorithm was developed and trained, among other details. Changes to the algorithm can be tracked using an immutable ledger facilitated by the implementation of blockchain.

Abstract: The invention generally relates to computer-based systems to evaluate and market clinical trial research centers. In certain aspects, the invention provides computer-based systems to collect information about clinical research centers. Systems include a tangible, non-transitory memory coupled to a processor operable to retrieve, based on a user's input, an identity of a clinical research center and prompt the user for information relating generally to the center. The system can collect disease-specific information by prompting the user for a selection of a disease and then collecting from the user information identifying an ability of the center to perform one or more tests relating to the disease.

Abstract: A computer-implemented tool for automated classification and interpretation of documents, such as life science documents supporting clinical trials, is configured to perform a combination of raw text, document construct, and image analyses to enhance classification accuracy by enabling a more comprehensive machine-based understanding of document content. The combination of analyses provides context for classification by leveraging relative spatial relationships among text and image elements, identifying characteristics and formatting of elements, and extracting additional metadata from the documents as compared to conventional automated classification tools.

Abstract: Systems and techniques are disclosed for using machine-learning to identify potential opportunity patients that are more likely to adjust his/her preference for a healthcare provider or service. In some implementations, integrated patient data is obtained. A patient sequence feature vector, a provider sequence feature vector, and a set of entity-specific feature vectors are generated. A set of opportunity patients is identified. A notification is transmitted to the set of opportunity patients about a second treatment plan.

Abstract: The disclosure generally describes computer-implemented methods, software, and systems for accessing volumes of data records structured to include sets dimensions, each dimension labelled in a manner specific to respective entities; identifying candidates data records keyed by managed keys that span a subset of dimensions even though at least one dimension from the subset of dimensions is labelled differently between the different volumes; comparing the candidate data records from the different volumes to determine whether a particular managed key is valid based on contents of the candidate data records from the different volumes; in response to determining that the particular managed key is valid, combining the candidate data records keyed by the valid managed key to be merged and accessible as one continuous entry; and in response to determining that the particular managed key is invalid, combining the candidate data records from the different volumes as separate entries.

Abstract: A method of managing consent using a computing device, the consent is given by a subject to one or more events in one or more studies, wherein the consent and the plurality of activities are changeable, the method including: authoring one or more first data forms describing the one or more events and one or more selections responsive to the one or more events; authoring, for each of the plurality of subjects, one or more second data forms including description of a plurality of preferences; forming, for a first of the plurality of subjects, an Informed Consent Forms document by combining the one or more first data forms of a first of the one or more studies and one or more second data forms for the first subject; and generating a manifest indicating the one or more events in the first study to which the first subject has granted consent.

Abstract: System and method to determine a reduced cohort criteria, the method including: defining N selection criteria to select a cohort from among a universe of patient data; querying a patient database, by use of a processor, and by use of the N selection criteria, in order to define a full patient population; selecting a subset of size M of the N selection criteria, to produce a subset criteria; selecting a permutation of the subset criteria, to produce a permuted subset criteria in a predetermined order; for each member of the permuted subset criteria: querying the patient database by use of the member of the permuted subset criteria to produce a respective interim patient population; combining all respective interim patient populations to produce a partial patient population; and calculating, by a processor, a coverage figure of merit that compares the partial patient population to the full patient population.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for generating a predictive system that obtains and processes data describing terms for different medical concepts to generate commands from a user query. An entity module of the system determines whether a term describes a medical entity associated with a healthcare condition affecting an individual. When the term describes the medical entity an encoding module links the medical entity with a specified category based on an encoding scheme. The system receives the user query. A parsing engine of the system uses the received query to generate a machine-readable command by parsing the query against terms that describe the medical entity and based on the encoding scheme for linking the medical entity to the specified category. The system uses the command to query different databases to obtain data for generating a response to the received query.

Abstract: A computer-implemented method includes a machine learning system receiving distinct types of data associated with multiple individual entities. For each of the individual entities, the machine learning system determines a first attribute that indicates a predicted attribute of the entity based on analysis of the data. The machine learning system also determines a second attribute that indicates a predicted quality attribute of the entity, based on analysis of the data. An attribute weighting module of the machine learning system generates weight values for each of the first attribute and the second attribute of the entity. The machine learning system generates a data structure that identifies a set of entities from among the multiple individual entities, where entities of the set are ranked based on a tier indicator that corresponds to either the first attribute, the second attribute, or both.

Abstract: Some implementations may provide a computer-assisted method for master data management, the method including: receiving configuration information defining a model of entities, each entity encoding attributes of a prescriber of one or more healthcare products; receiving specification information defining mapping logic, searching logic, and matching logic, and merging logic for processing base entities and related entities of the model; receiving data from more than one source customer databases, the customer database including data encoding prescribers of healthcare products and being maintained by more than one organizations; translating the received data into staging data according to the mapping logic in the received specification information; generating master data by processing the staging data according to the searching logic, matching logic, and merging logic in the received specification information; and synchronizing at least a portion of the master data to at least one of the source customer database