Browse Prior Art Database

Automated Patient Care Planner

IP.com Disclosure Number: IPCOM000224416D
Original Publication Date: 2012-Dec-18
Included in the Prior Art Database: 2012-Dec-18
Document File: 4 page(s) / 266K

Publishing Venue

Linux Defenders

Related People

ABHIRAMI BASKARAN: AUTHOR

Abstract

Documentation and monitoring of health records is one of the primary tasks being undertaken by most health centers. A significant percentage of patient deaths is attributed to inaccurate record keeping and planning. The current work focuses on coupling a system tracking patient history with an automatic care plan designer. This can device care plans for patients based on the history of medications, vitals and lab results. It helps reduce medical errors and also can help primary health care workers, nurses. The decision making system can be a trained neural network or a support vector machine used in data mining. Medical algorithms can be used to base the rules on which the decision making system is implemented.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 53% of the total text.

Page 01 of 4

AUTOMATED PATIENT CARE PLANNER

ABHIRAMI BASKARAN


1. Problem Description

  Documentation and monitoring of health records is one of the primary tasks being un- dertaken by most health centers. A significant percentage of patient deaths is attributed to inaccurate record keeping and planning. The current work focuses on coupling a system tracking patient history with an automatic care plan designer. This can device care plans for patients based on the history of medications, vitals and lab results. It helps reduce medical errors and also can help primary health care workers, nurses. The decision making system can be a trained neural network or a support vector machine used in data mining. Medical algorithms can be used to base the rules on which the decision making system is implemented.


2. Implementation


Datasource is contained in a container on a web server and this data is queried to obtain the records for analysis of a patients condition.It contains data like glucose, blood pressure etc.


The client application is an application that accesses the data and forms the visu- alization component of the system. This is displayed in Figure 1.


The medication, vitals and lab data are displayed on a common temporal time axis. This is beneficial to analyze the patient's progress over the time.

Finally a feature vector containing the patients condition is built. This is a data matrix containing all the data of the patients medical history. The features could be glucose level, urea level etc. This is input to a support vector machine. The support vector machine or the analyzer classifies the patients condition.


The support vector machine as seen in Figure 2 is essentially a fitting function. The function is formulated based on sample data. The function should have minimal error so that it does not add any risk to the system. Based on standard medical algorithms the output of the support vector machine is used to output decisions and classifications for the pat...