Abstract: A method includes discarding a current knowledge corpus; selecting a new knowledge corpus; performing operations as follows using an Artificial Intelligence (Al) retriever engine: dividing the new knowledge corpus into a plurality of sub-documents; encoding a query for the plurality of sub-documents using a query encoding model; encoding each of the plurality of sub-documents using a document encoding model; and determining at least one matching sub-document of the plurality of sub-documents that is a match for containing an answer to the query based on the encoded query and each of the plurality of encoded sub-documents; performing operations as follows using an AI reader engine: generating an inference about the answer to the query based on a concatenation of each of the at least one matching sub-document with the query, each of the at least one matching sub-document having an associated reader loss function result for the inference; identifying one of the at least one matching sub-document having a lowest

Abstract: A method includes receiving a record including at least one page and containing clinical information associated with a first patient; receiving respective first patient identification values for one or more patient identification parameters corresponding to the first patient; automatically processing the record to identify first example instances referencing the patient identification parameters including values therefor; automatically processing the record to identify second example instances of the first patient identification values; automatically processing the record to determine whether any of the at least one page contained therein cannot be semantically linked to another one of the at least one page in the record; and assigning a grade to each of the at least one page indicating a degree of confidence that the record does not include clinical information associated with a second patient based on the first example instances, the second example instances, and the determination whether any of the at leas

Abstract: A method, apparatus and computer program product refine an automated coding model, such as for a medical chart. For each respective candidate code from a set of candidate codes, the method predicts a probability of the respective code being contained in a medical chart. The method also selects one of the candidate codes as being contained in the medical chart based upon the probability and removes the selected candidate code from the set of candidate codes. The method then repeatedly predicts the probability of a respective code being contained in the medical chart, selects one of the candidate codes based upon the predicted probability and removes the selected candidate code from the set of candidate codes. The method further determines a categorical crossentropy loss as to permit adjustment of one or more parameters of the automated coding model.

Abstract: A method, apparatus and computer program product are provided for implementing an actor-critic model for processing a large document. The document is processed in smaller portions, and the information is accumulated or aggregated in a vector representation of the document. The vector representation may be reconfigured with each time step so that the vector representation includes an estimate of the most relevant information from the document processed thus far. The vector representation may be used to classify the document, such as by determining a primary and/or secondary diagnosis in a medical record. Feedback from the critic is provided in the form of a reward signal to the actor, which fine tunes its parameters and processes the different portions of the document. The actor-critic model may be trained with training data including documents and their known associated classifications.

Abstract: A method, apparatus and computer program product are provided for classifying pages of a document with a linear regression model, and a deep learning (non-linear) model utilizing a neural network. The classification of each page is determined by determining which of the linearly predicted category or the non-linearly predicted category to use, such as for transmission to an auditor. Pages of medical records generated by concatenating reports from distinct sources are classified according to both models, and embodiments determine which classification should be used. The classifications may be optionally smoothed for continuity. The classifications may be sent to auditors and used to review and audit the medical records.