Week 7

 Lecture 

Communication is is a very important part in Healthcare and in this weeks lecture we were infomed about what is communicatiuon in Healthcare and how it works.

Communication Networks In Healthcare

Who needs to comminicate in Healthcare?

• Hospitals
• GPs 
• Parmacy
• Patients

Electronic communication system

 Electronic communication systems are a collection of communications networks, transmissions systems, relay stations, tributary stations and data terminal equipment capable of interconnection and interoperation.

Computer network types

• LAN (Local Area Network)
• WAN (Wide Area Network)

Local Are Network is a computer network covering a small physical area, like a home, office, or small group of buildings, such as a school or an airport. Wide Area Network is a computer network that covers a broad area (i.e., any network whose communications links cross metropolitan, regional or national boundaries.

Data Interchange Standards

• EDI (Electronic Data Interchange is the transfer of data between different companies using networks such as VANs Or the Internet.)
• HL7 ( Specifies standards, guidelines and methodoligies by which various healthcare systems can communicate with each other)

This is the HL7 website and has alot of information about them. http://www.hl7.org/

In this weeks lecture i leaned alot more information about communication and especially communication in Healthcare.

Tutorial 

In this weeks tutorial we discussed more about health in communication and got information about the week 7 practical.

 Practical

Week 7 practical we learned what is data mining and we got to use the Data mining tool. Data mining is searching for patterns in large ammounts of data examples of this are Health Data and Marketing. The Data Mining tool we used was WEKA.

Supervised algorithm is “taught” what patterns/classes to expect in the data during a training phase; after training algorithm classifies without further help.

Unsupervised is an attempt to find patterns without the use of a prior knowledge.

In the practical we were given a few taks to complete using the WEKA tool and the following is some of the completed tasks i completed.

1. 1.      What is the Total Number of Instances? 150
2. 2.      What is the percentage of Correctly Classified Instances? 96%
3. 3.      What is the percentage of Incorrectly Classified Instances? 4%

Cross Validation an approach by which the sets of scientific data generated using two or more methods are critically assessed.

Confusion matrix  is a visualization tool typically used in supervised learning. Each row of the matrix represents the instances in a predicted class, while each column represents the instances in an actual class. One benefit of a confusion matrix is that it is easy to see if the system is confusing two classes.

What would the matrix for a perfect model (100% success) look like?

The Matrix of a  perfect model will show the data going up in a diagonal row.

 

 

Advertisement