Abstract: Techniques are described herein for segmenting source code into syntactically coherent sequences of tokens that satisfy constraints inherent in sequence-to-sequence networks. In various implementations, source code may be processed to generate one or more graphs representing the source code. One or more of the graphs may then be traversed to identify one or more sequences of tokens within the source code that satisfy an input constraint of a sequence-to-sequence network. The source code may be segmented into the identified one or more sequences of tokens. The one or more sequences of tokens may then be processed using the sequence-to-sequence network.

Abstract: Techniques are described herein for segmenting source code into syntactically coherent sequences of tokens that satisfy constraints inherent in sequence-to-sequence networks. In various implementations, source code may be processed to generate one or more graphs representing the source code. One or more of the graphs may then be traversed to identify one or more sequences of tokens within the source code that satisfy an input constraint of a sequence-to-sequence network. The source code may be segmented into the identified one or more sequences of tokens. The one or more sequences of tokens may then be processed using the sequence-to-sequence network.

Abstract: Implementations are described herein for leveraging machine learning to automate source code refactoring and/or rearchitecting. In various implementations, one or more ground truth boundaries may be removed from one or more boundaried source code files to produce one or more boundary-less source code files. One or more of the boundary-less source code files may be processed using a machine learning model to predict one or more candidate boundaries for reintroduction into the one or more boundary-less source code files. The one or more ground truth boundaries may be compared with the one or more predicted candidate boundaries. The machine learning model may be trained based on the comparing.

Abstract: Implementations are described herein for leveraging machine learning to automate source code refactoring and/or rearchitecting. In various implementations, one or more ground truth boundaries may be removed from one or more boundaried source code files to produce one or more boundary-less source code files. One or more of the boundary-less source code files may be processed using a machine learning model to predict one or more candidate boundaries for reintroduction into the one or more boundary-less source code files. The one or more ground truth boundaries may be compared with the one or more predicted candidate boundaries. The machine learning model may be trained based on the comparing.

Abstract: Automated diagnosis archetypes based on a complete list of library of medical illnesses are used dynamically to help prioritize health data that is relevant to a physician's medical diagnosis for a patient. A super plurality of diagnoses patient archetypes are narrowed to match a patient profile based on information collected from different sources. Based on dynamic physician's diagnosis, the patient profile is mapped to a subset of plurality of patient archetypes. An automated agent is employed to customize a user interface to emphasize order selections to support eliminating a false positive in the proper subset of the plurality of patient persons, a false negative in the plurality of patient archetypes, or to increase a predicted applicability of a set of patient archetypes to the patient profile. The subset of patient archetypes ranking is updated based on received order results. A confirmed dynamic physician diagnosis is made available.

Abstract: Automated diagnosis archetypes based on a complete list of library of medical illnesses are used dynamically to help prioritize health data that is relevant to a physician's medical diagnosis for a patient. A super plurality of diagnoses patient archetypes are narrowed to match a patient profile based on information collected from different sources. Based on dynamic physician's diagnosis, the patient profile is mapped to a subset of plurality of patient archetypes. An automated agent is employed to customize a user interface to emphasize order selections to support eliminating a false positive in the proper subset of the plurality of patient persons, a false negative in the plurality of patient archetypes, or to increase a predicted applicability of a set of patient archetypes to the patient profile. The subset of patient archetypes ranking is updated based on received order results. A confirmed dynamic physician diagnosis is made available.

Abstract: A prescription outcome system comprising: a prescription outcome engine server connected via a network to a pharmacy computer, the prescription outcome engine server having an engine memory, and an engine processor communicably coupled to the engine memory, and the engine processor configured to execute instructions to: receive claim from a pharmacy, the claim including a patient identifier for a patient and a pharmaceutical identifier for a pharmaceutical desired to be filled; determine if the patent is included on an enrolled patient list, if the pharmaceutical is included on a pharmaceutical allowed list, and if the patient may refill the pharmaceutical at a present time; when the patient is included on the enrolled patient list, the pharmaceutical is included on a pharmaceutical allowed list, and the patient may refill the pharmaceutical at the present time, price the claim, debit employer funds, report adjudication to an employer, report adjudication to a pharmaceutical manufacturer, report adjudication

Abstract: Automated diagnosis archetypes based on a complete list of library of medical illnesses are used dynamically to help prioritize health data that is relevant to a physician's medical diagnosis for a patient. A super plurality of diagnoses patient archetypes are narrowed to match a patient profile based on information collected from different sources. Based on dynamic physician's diagnosis, the patient profile is mapped to a subset of plurality of patient archetypes. An automated agent is employed to customize a user interface to emphasize order selections to support eliminating a false positive in the proper subset of the plurality of patient persons, a false negative in the plurality of patient archetypes, or to increase a predicted applicability of a set of patient archetypes to the patient profile. The subset of patient archetypes ranking is updated based on received order results. A confirmed dynamic physician diagnosis is made available.

Abstract: A method of confirming a medical diagnosis using a graphical user interface includes receiving an input selecting a first medical diagnosis as a cause of a patient complaint, and displaying a template based on the first medical diagnosis. The template includes a first query associated with the first medical diagnosis, a suggested response to the first query, and a first input field configured to receive the suggested response to the first query. The method includes using provider-configurable diagnostic based templates facilitating automated data entry to current and future electronic medical records systems. Systems and machine-readable media are also described.