Systems and methods for patient retention in network through referral analytics
A medical information navigation engine (“MINE”) is capable of inferring referral activity not reported into a referral workflow system by utilizing intent-based clustering of medical information. The intent based clustering reconciles received medical data, from a variety of sources, and then clusters the data by applying one or more clustering rules. After the referrals not otherwise reported are inferred, they may be utilized to generate metrics that can be utilized to enhance patient care, and reduce costs. Metrics may be generated for both in-network and out-of-network referrals in order to distinguish differences in reporting activity.
1. A computer-implemented method for electronic referral analysis, the method implemented by a medical information navigation engine (“MINE”) including a front-end server, a back-end server, and an interface server, the method comprising:
receiving, by the interface server and via at least one secure communication link, medical information including reported-referral data and visit-note data, wherein the medical information is received in different formats;
reformatting, by the back-end server, the medical information by converting different formats into a single format that facilitates reconciling and de-duplicating of the medical information;
transmitting, by the interface server, the reformatted medical information to a database in encrypted form, the database configured to store the reformatted and encrypted medical information;
retrieving, by the back-end server from the database, the reformatted and encrypted medical information;
applying, by the back-end server, a set of rules to the retrieved medical information, wherein applying the set of rules comprises:
(i) identifying one or more similarities within the retrieved medical information and defining characteristics of the retrieved medical information by applying at least one first rule of the set of rules to the retrieved medical information, wherein the at least one first rule is configured to learn to detect similarities using first feedback data associated with a plurality of medical information received by the MINE, thereby refining the at least one first rule,
(ii) clustering the retrieved medical information into one or more groups by applying at least one second rule of the set of rules to the identified one or more similarities, wherein the at least one second rule is configured to learn to detect clusters using second feedback data associated with the plurality of medical information received by the MINE, thereby refining the at least one second rule, and
(iii) identifying information within the retrieved medical information that has changed over time by applying at least one third rule of the set of rules to the one or more groups, wherein the at least one third rule is configured to learn to detect changes over time using third feedback data associated with the plurality of medical information received by the MINE, thereby refining the at least one third rule;
inferring, by the back-end server based on the application of the set of rules to the retrieved medical information, unreported referrals, the unreported referrals comprising instances in which a referral does not appear in the reported-referral data;
generating, by the front-end server, metrics by (i) adding reported referral activity included in the reported-referral data to the inferred unreported referrals to generate a combined referral number and (ii) dividing the reported referral activity by the combined referral number; and
displaying, by the front-end server, an analytical dashboard on a user interface of a user computing device, the analytical dashboard displaying the generated metrics.
2. The method of claim 1 , wherein
receiving the medical information comprises receiving, by the interface server via the at least one secure communication link, the medical information from a plurality of medical data sources.
3. The method of claim 1 further comprising:
building, by the back-end server executing a reconciliation engine, an intent-based clustering model by applying, the set of rules to the retrieved medical information, wherein the intent-based clustering model is configured to identify one or more referrals within the retrieved medical information and other retrieved medical information.
4. The method of claim 1 , wherein the back-end server includes:
a reconciliation engine configured to apply the at least one first rule to the retrieved medical information; and
an intent-based presentation engine configured to apply the at least one second rule to the identified one or more similarities.
5. The method of claim 1 further comprising:
computing, by the back-end server, a distance between patient data attributes included in a set of medical information and the clustered medical information.
6. The method of claim 1 , further comprising:
generating, by the front-end server, the metrics for in network referrals and out of network referrals.
7. A medical information navigation engine (“MINE”) comprising:
an interface server, the interface server configured to:
receive, via at least one secure communication link, medical information including reported-referral data and visit-note data, wherein the medical information is received in different formats; and
transmit reformatted medical information to a database in encrypted form, the database configured to store the reformatted and encrypted medical information;
a back-end server configured to:
reformat the medical information by converting different formats into a single format that facilitates reconciling and de-duplicating of the medical information;
retrieve, from the database, the reformatted and encrypted medical information including;
apply a set of rules to the retrieved medical information, wherein applying the set of rules includes:
(i) identifying one or more similarities within the retrieved medical information and defining characteristics of the retrieved medical information by applying at least one first rule of the set of rules to the retrieved medical information, wherein the at least one first rule is configured to learn to detect similarities using first feedback data associated with a plurality of medical information received by the MINE, thereby refining the at least one first rule,
(ii) clustering the retrieved medical information into one or more groups by applying at least one second rule of the set of rules to the identified one or more similarities, wherein the at least one second rule is configured to learn to detect clusters using second feedback data associated with the plurality of medical information received by the MINE, thereby refining the at least one second rule, and
(iii) identifying information within the retrieved medical information that has changed over time by applying at least one third rule of the set of rules to the one or more groups, wherein the at least one third rule is configured to learn to detect changes over time using third feedback data associated with the plurality of medical information received by the MINE, thereby refining the at least one third rule; and
infer, based on the application of the set of rules to the retrieved medical information, unreported referrals, the unreported referrals comprising instances in which a referral does not appear in the reported-referral data; and
a front-end server configured to:
generate metrics by (i) adding reported referral activity included in the reported-referral data to the inferred unreported referrals to generate a combined referral number and (ii) dividing the reported referral activity by the combined referral number; and
display an analytical dashboard on a user interface of a user computing device, the analytical dashboard displaying the generated metrics.
8. The MINE of claim 7 , wherein the interface server is configured to:
receive, via the at least one secure communication link, the medical information from a plurality of medical data sources.
9. The MINE of claim 7 , wherein the back-end server is further configured to:
build, by executing a reconciliation engine, an intent-based clustering model by applying, the set of rules to the retrieved medical information, wherein the intent-based clustering model is configured to identify one or more referrals within the retrieved medical information and other retrieved medical information.
10. The MINE of claim 7 , wherein the back-end server comprises:
a reconciliation engine configured to apply the at least one first rule to the retrieved medical information; and
an intent-based presentation engine configured to apply the at least one second rule to the identified one or more similarities.
11. The MINE of claim 7 , wherein the back-end server is further configured to compute a distance between patient data attributes included in a set of medical information and the clustered medical information.
12. The MINE of claim 7 , wherein the front-end server is further configured to generate the metrics for in network referrals and out of network referrals.