Apixio, Inc.

John O Schneider

Vishnuvyas Sethumadhavan

Robert Derward Rogers

Darren Matthew Schulte

Shahram Shawn Dastmalchi

Richard Joseph Belcinski

Haoning Fu

Imran N Chaudhri

Jose Cruz Toledo

Erin Michelle Faverty

Austin Robert Rogers

Systems and methods for data warehouse management for medical information is provided. The system receives a set of medical record documents. These documents include evidence for one or more findings which are identified using natural language processing of evidence. The data set is probabilistically transformed into a structured data set (often as a table). This structured data set includes embedded links that reference the source document where the evidence is located. If the finding is supported by multiple articles of evidence, the link will direct the user to the source document with the highest confidence ranking. Evidence in the source document is highlighted and may include associated annotations. Evidence, findings and annotations may be updated by the user.

Systems and methods for medical information data warehouse management

Systems and methods to improve the optical character recognition of records, and in particular health records, are provided. An image of a medical record is received, and an initial optical image recognition (OCR) on the image is performed to identify text information. The OCR signal quality may be measured, and areas of insufficient OCR signal quality may be isolated. The signal quality is determined by a weighted average of semantic analysis of the resulting text, and/or OCR accuracy measures. The OCR process may be repeated on the isolated regions of lower signal quality, each time using a different OCR transform, until all regions are completed with a desired degree of signal quality (accuracy). All the regions of the document may then be recompiled into a single document for outputting.

Systems and methods for improved optical character recognition of health records

Systems and methods for managing audit risks utilizing the true state of the patient are provided. A number of medical records for a patient are subjected to predictive modeling for various conditions (known as patient 'true state'). The true state is then cross referenced by the eligible Medicare documentation, and any findings that are being submitted to MediCare for reimbursement. The result of this cross referencing is the ability to classify each finding and/or true state into a 'green, “yellow', or “red zone”. The green zone is where the finding, documentation and true state are in good alignment. A red zone is where the finding and the true state are entirely at odds. The yellow zone is where the findings and the true state are in agreement, but where there is still audit risk that may be resolved through one or more “opportunities”.

Systems and methods for determination of patient true state for risk management

To achieve the foregoing and in accordance with the present invention, systems and methods for workflow management are provided. In some embodiments, a set of data elements, which are extracted from medical information, are received. The elements are bundled according to one or more similar attributes to form work items. Work items are bundles of data elements for which value is to be extracted for a particular user objective. Next the workflows are configured according to event history for the work items, current action, and user context. The work items may then be routed through the workflow based upon the work items' 'energy'. Work item energy is the probability and degree to which a next action taken in a workflow to that particular work item will further a user objective.

Systems and methods for enhancing workflow efficiency in a healthcare management system

Systems and methods for generating customized annotations of a medical record are provided. The system receives a medical record and processes it using a predictive model to identify evidence of a finding. The system then determines whether to have a recall enhancement or validation of a specific finding. Recall enhancement is used to tune or develop the predictive model, while validation is used to rapidly validate the evidence. The source document is provided to the user and feedback is requested. When asking for validation, the system also highlights the evidence already identified and requests the user to indicate if the evidence is valid for a particular finding. If recall enhancement is utilized, the source document is provided and the user is asked to find evidence in the document for a particular finding. The user may then highlight the evidence that supports the finding. The user may also annotate the evidence using free form text.

Systems and methods for customized annotation of medical information

Systems and methods for personalizing medicine utilizing the true state of the patient are provided. A number of medical records for a patient are subjected to predictive modeling for various conditions (known as patient 'true state'). The patient personal information, previous care, and true state may be provided into a state machine in order to determine the resources needed for the patient. The medical resources may be any of laboratory services, diagnostics, therapies and medications. Using the true state information, and number of activities may be performed for the patient based upon the patient's needs. These activities include scheduling lab or diagnostic procedures in advance of an appointment, filling in documentation gaps, identifying items that require additional documentation using the true state, and tracking follow-up. It may also be beneficial to validate the true state.

Systems and methods for determination of patient true state for personalized medicine

Systems and methods to automatically generate classifiers are provided. A labeled dataset is initially received. The dataset may be for a positive class, or may be a negative for a class, or a false positive class. N features that are predictive for the class (or false positive or the negative class) are identified. These features are combined within a classifier dictionary. Medical records received may be processed in order to be machine readable. Features within the medical records are identified and are compared against the dictionary of classifiers. Matches indicate classes within the medical record. The classifier dictionary may be periodically updated in response to insufficient classification accuracy, or when new data becomes available.

Systems and methods for automated generation classifiers

Systems and methods for managing audit risks utilizing the true state of the patient are provided. A number of medical records for a patient are subjected to predictive modeling for various conditions (known as patient 'true state'). The true state is then cross referenced by the eligible Medicare documentation, and any findings that are being submitted to MediCare for reimbursement. The result of this cross referencing is the ability to classify each finding and/or true state into a 'green, 'yellow', or &#x201c;red zone'. The green zone is where the finding, documentation and true state are in good alignment. A red zone is where the finding and the true state are entirely at odds. The yellow zone is where the findings and the true state are in agreement, but where there is still audit risk that may be resolved through one or more &#x201c;opportunities&#x201d;. Examples of opportunities include bolstering the documentation for the reimbursement, getting additional evidence to improve the confidence of a true state inference, or including additional documentation for a finding that exists in the true state, but hasn't been previously identified.

A sorter of medical findings for assessment by a medical coder is provided. In some embodiments, the sorter receives information about a user (coder), including identification, a role, and historical activity. The sorter determines whether to run in exploration or exploitation modes. Exploration is used to explore the scope of the findings and also identify variables that impact a finding. Exploitation is designed to maximize a goal (such as throughput or profitability). Lastly a finding is selected and provided to the user. The selection is based upon computing internal parameters when in exploration, or based upon optimizing for criteria when in exploitation. The sorter may also determine competency for the user, and cut them off from performing additional coding if they are found incompetent.

Growing community of inventors

Los Gatos, CA, United States of America

John O Schneider