Method and system for medical suggestion search
The present teaching relates to medical suggestion searching. In one example, data related to a medication drug are obtained. One or more candidate prescription strings are identified from the obtained data. Each of the candidate prescription strings is associated with a plurality of attributes. Each of the one or more candidate prescription strings is automatically processed based on at least one model to generate one or more prescription strings each with an associated ranking. At least some of the generated one or more prescription strings and the associated rankings are stored for future use.
1. A method comprising:
receiving a request for a medical suggestion, wherein the request results from a single action of a user and is associated with at least two dimensions, a first dimension of the at least two dimensions relating to an attribute of a patient and a second dimension of the at least two dimensions relating to an attribute of a physician;
providing the at least two dimensions to a machine-learned medical suggestion model to obtain a plurality of medical suggestion strings from a database of medical suggestions, the plurality of medical suggestion strings being determined, by the machine-learned medical suggestion model, to belong in a set of highest-scoring medical suggestion strings for the at least two dimensions, a score of a medical suggestion string in the set represents an aggregated degree of match between the medical suggestion string and the at least two dimensions;
providing the plurality of medical suggestion strings in response to the request;
receiving selection of a medical suggestion string of the plurality of medical suggestion strings; and
generating a medical transaction for the patient using the medical suggestion string.
2. The method of claim 1 , further comprising:
adding the medical transaction to a medical transaction database; and
updating statistics in the database of medical suggestions.
3. The method of claim 2 , wherein updating the statistics in the database includes:
updating a frequency of occurrence of the medical suggestion string; and
updating a recency of occurrence of the medical suggestion string.
4. The method of claim 3 , wherein the recency is weighted by a time decay factor in the aggregated degree of match.
5. The method of claim 1 , wherein the aggregated degree of match is weighted by weights learned during training of the machine-learned medical suggestion model.
6. The method of claim 1 , the at least two dimensions including a plurality of dimensions describing the patient.
7. The method of claim 6 , the plurality of dimensions including dimensions related to a diagnosis for the patient and dimensions related to a profile of the patient.
8. The method of claim 1 , wherein the aggregated degree of match is based on total occurrence, frequency of occurrence, and recency of occurrence of the medical suggestion string with respect to the at least two dimensions.
9. The method of claim 8 , wherein medical suggestion strings with recent occurrences are boosted.
10. The method of claim 1 , wherein the medical suggestion is one of a prescription, a physical therapy order, a diet recommendation, a lab test order, or a radiology test order.
11. A system comprising:
a processor; and
memory storing instructions that, when executed by the processor, cause the system to perform operations including:
receiving a request for a medical suggestion, wherein the request results from a single action of a user and is associated with at least two dimensions, a first dimension of the at least two dimensions relating to an attribute of a patient and a second dimension of the at least two dimensions relating to an attribute of a physician,
providing the at least two dimensions to a machine-learned medical suggestion model to obtain a plurality of medical suggestion strings from a database of medical suggestions, the plurality of medical suggestion strings being determined, by the machine-learned medical suggestion model, to belong in a set of highest-scoring medical suggestion strings for the at least two dimensions, a score of a medical suggestion string in the set represents an aggregated degree of match between the medical suggestion string and the at least two dimensions,
providing the plurality of medical suggestion strings in response to the request,
receiving selection of a medical suggestion string of the plurality of medical suggestion strings, and
generating a medical transaction for the patient using the medical suggestion string.
12. The system of claim 11 , the operations further including:
adding the medical transaction to a medical transaction database; and
updating statistics in the database of medical suggestions.
13. The system of claim 12 , wherein updating the statistics in the database includes:
updating a frequency of occurrence of the medical suggestion string; and
updating a recency of occurrence of the medical suggestion string.
14. The system of claim 13 , wherein the recency is weighted by a time decay factor in the aggregated degree of match.
15. The system of claim 11 , wherein the aggregated degree of match is weighted by weights learned during training of the machine-learned medical suggestion model.
16. The system of claim 11 , the at least two dimensions including a plurality of dimensions describing the patient.
17. The system of claim 16 , the plurality of dimensions including dimensions related to a diagnosis for the patient and dimensions related to a profile of the patient.
18. The system of claim 11 , wherein the aggregated degree of match is based on total occurrence, frequency of occurrence, and recency of occurrence of the medical suggestion string with respect to the at least two dimensions.
19. The system of claim 18 , wherein medical suggestion strings with recent occurrences are boosted.
20. The system of claim 11 , wherein the medical suggestion is one of a prescription, a physical therapy order, a diet recommendation, a lab test order, or a radiology test order.