Patient-centric clinical knowledge discovery system
A medical information retrieval system comprises a natural language processing system that processes a vocal user query to identify key words and phrases. These key words and phrases are provided to an inferencing engine that provides a set of knowledge-based inferences from medical knowledge sources, based on these key words and phrases. Thereafter, these knowledge-based inferences are provided to an information retrieval engine that retrieves a corresponding plurality of medical articles based on these knowledge-based inferences, and ranks each with respect to the knowledge-based inferences. A summary engine receives the ranked articles and creates a model based on the topical keywords and candidate sentences found in the highly ranked articles. A paraphrase engine processes the candidate sentences to provide a summary response based on a knowledge-based paraphrase model. An audio output device renders the summary report as the response to the user's original vocal query.
1. A medical information retrieval system comprising:
an input device that receives a user query;
a natural language processing system that processes the user query to identify key words and phrases;
an inferencing engine that receives the key words and phrases and provides a plurality of knowledge-based inferences from medical knowledge sources based on the key words and phrases;
an information retrieval engine that receives the plurality of knowledge-based inferences and searches a plurality of biomedical reference sources to identify a plurality of medical articles based on the plurality of knowledge-based inferences, and ranks each article with respect to the knowledge-based inferences;
a summary engine that receives the plurality of ranked articles, identifies topical keywords and candidate sentences based on the ranked articles, and provides a summary report based on the topical keywords and candidate sentences;
a paraphrase engine that provides paraphrased sentences based on the candidate sentences to the summary report, wherein the paraphrase engine includes a Recurrent Neural Network based encoder/decoder framework; and
an output device that renders the summary report in response to the user query.
2. The system of claim 1 , wherein the input device includes an audio input device, and the output device includes an audio output device.
3. The system of claim 1 , wherein the inferencing engine includes a neural model that is trained using the medical knowledge sources, and provides the knowledge-based inferences upon receipt of the key words and phrases.
4. The system of claim 1 , wherein the medical knowledge sources include one or more medical texts.
5. The system of claim 1 , wherein the plurality of biomedical reference sources are accessed via an Internet connection.
6. The system of claim 1 , wherein the summary engine creates a model based on the topical keywords and candidate sentences, and uses the model to create the summary report.
7. The system of claim 1 , wherein the summary engine includes a Restricted Boltzmann Machine.
8. A non-transitory computer-readable medium that includes a program that, when executed by a processing system, causes the processing system to:
receive a user query;
processes the user query using natural language processes to identify key words and phrases;
provide the key words and phrases to an inferencing engine that provides a plurality of knowledge-based inferences from medical knowledge sources based on the key words and phrases;
provide the plurality of knowledge-based inferences to an information retrieval engine that searches one or more biomedical reference sources to identify a plurality of medical articles based on the plurality of knowledge-based inferences;
rank each article with respect to the knowledge-based inferences to identify one or more target articles;
summarize the one or more target articles by identifying topical keywords and candidate sentences in the target articles;
provide the candidate sentences to a paraphrase engine that provides paraphrased sentences based on the candidate sentences, wherein the paraphrase engine includes a Recurrent Neural Network based encoder/decoder framework;
create a summary report based on the topical keywords and paraphrased sentences; and
provide the summary report to an output device that renders the summary report in response to the user query.
9. The medium of claim 8 , wherein the processing system receives the user query and provides the summary report in audible form.
10. The medium of claim 8 , wherein the inferencing engine includes a neural model that is trained using the medical knowledge sources, and provides the knowledge-based inferences upon receipt of the key words and phrases.
11. The medium of claim 8 , wherein the medical knowledge sources include one or more medical texts, and the information retrieval engine accesses the plurality of biomedical reference sources via an Internet connection.
12. The medium of claim 8 , wherein the summary engine includes a Restricted Boltzmann Machine.