Browse Prior Art Database

Method and System for Learning Medical Attributes of a Patient based on Preferences of Practitioner and Potential Diagnosis

IP.com Disclosure Number: IPCOM000247683D
Publication Date: 2016-Sep-27
Document File: 2 page(s) / 95K

Publishing Venue

The IP.com Prior Art Database

Abstract

A method and system is disclosed for learning medical attributes of a patient based on preferences of practitioner and potential diagnosis. The method and system provides most relevant medical attributes of the patient such as, but need not be limited to, medical conditions, procedures, lab results, allergies and medications to the practitioner to make right diagnosis. The method and system provides the most relevant medical attributes to the practitioner based on the preferences of the practitioner, the potential diagnosis and characteristics of the patient.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 51% of the total text.

Page 01 of 2

Method and System for Learning Medical Attributes of a Patient based on Preferences of Practitioner and Potential Diagnosis

Disclosed is a method and system for learning medical attributes of a patient based on preferences of practitioner and potential diagnosis. The method and system provides most relevant medical attributes of the patient such as, but need not be limited to, medical conditions, procedures, lab results, allergies and medications to the practitioner to make right diagnosis. The method and system provides the most relevant medical attributes to the practitioner based on the preferences of the practitioner, the potential diagnosis and characteristics of the patient.

Following figure is a flow diagram for recommending a set of medical attributes as most relevant medical attributes of a patient to a practitioner.

Figure

As illustrated in figure, the method and system extracts medical characteristics and diagnosis of patients from medical records using medical analytics pipelines and diagnosis provided by the practitioner. The method and system also uses feature vectors such as, but need not be limited to, age, gender and diagnosis to determine characteristics of the patients.

The method and system, then, allows the practitioner to select medical attributes. The medical attributes include, but need not be limited to, medical conditions, allergies, procedures, lab results and medications. The medical attributes are used as labels to form a training set of data. The method and system employs supervises machine

1


Page 02 of 2

learning algorithms such as, but need not be limited to, collaborative filtering, linear regression, K Nearest Neighbors (kNN) and neural networks to create a hypothesis for the training set of data.

Once the hypothesis is created, the method and system validates medical characteristics and diagnosis of a patient against the hypothesis to recommend the set of medical attributes to the practitioner. The set of attributes recommended are the most relevant medical attributes of the patient.

Consider a scenario, where a patient walks to an eme...