CSE5810: Intro to Biomedical Informatics

CSE5810: Intro to Biomedical Informatics CSE5810: Intro to Biomedical Informatics - Start

2016-05-13 36K 36 0 0

CSE5810: Intro to Biomedical Informatics - Description

The Role of AI in Clinical Decision  Support  . Saahil Moledina. University of Connecticut. saahil.moledina@uconn.edu. Clinical Decision Support in Biomedical Informatics:. CDS in Biomedical Informatics. ID: 317639 Download Presentation

Download Presentation

CSE5810: Intro to Biomedical Informatics




Download Presentation - The PPT/PDF document "CSE5810: Intro to Biomedical Informatics" is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.



Presentations text content in CSE5810: Intro to Biomedical Informatics

Slide1

CSE5810: Intro to Biomedical Informatics

The Role of AI in Clinical Decision  Support 

Saahil Moledina

University of Connecticut

saahil.moledina@uconn.edu

Slide2

Clinical Decision Support in Biomedical Informatics:

Slide3

CDS in Biomedical Informatics

Introduction:

The Clinical Decision support in biomedical informatics is the knowledge that is provided to assist the clinician and/or patients for assisting them in making decisions regarding choice of treatment.

These decisions are tried to be made easy by giving the knowledge of the outcomes and complications of the treatment chosen.

Now, Clinical Decision Support systems are systems are systems designed to do the clinical decision support and process them using AI and machine learning.

For these systems to work efficiently it needs to combine the efforts of the patients, clinicians, nurses and decision aids.

Slide4

CDS in Biomedical Informatics

Research Areas:

Artificial Intelligence

Machine Learning

User Interfaces

Data Mining

Data warehousing

Medicine

Algorithms

Benefits of CDS:

Increased quality of care and enhanced health outcomes

Avoidance of errors and adverse events

Improved efficiency, cost-benefit, and provider and patient satisfaction

Slide5

CDS in Biomedical Informatics

Stakeholders:

Patients

Clinicians, nurses, Physicians.

Vendors.

Hospitals/clinics.

Standards:

HL7 version 3( representation of patient data for Clinical decision support).

Infobutton ( Context-Aware Retrieval Application)

 

GLIF (knowledge representation)

Arden Syntax

(

knowledge representation

)

GELLO (Common Expression Language

)

Unified Medical Language

System and

component terminologies (e.g.,

SNOMED, LOINC, RxNorm)

Slide6

CDS in Biomedical Informatics

An important aspect for clinical decision making is the patients perspective of their health problems and preferences for the treatment.

One of the biggest problem in clinical decision support systems is integrating the patient perspective in decision making.

For this we need to make the use of Shared Decision making (SDM).

This results into a new problem on how to develop a clinical decision support system for SDM.

Slide7

CDS in Biomedical Informatics

Slide8

CDS in Biomedical Informatics

Features of a Clinical Decision Support with SDM are:

Provide Clinicians with the health problems associated with a patient’s illness.

Treatment Options

Benefits and Risks of the treatment.

Patient Preferences.

Acceptable to the clinicians.

Slide9

CDS in Biomedical Informatics

To make such a system knowledge about the following things are vital:

Clinical Domain

for understanding the decision problem. E.g. coronary artery.

Decision Science and research of SDM

to draw out the patient’s preferences.

Biomedical informatics

Algorithms and technologies.

Organizational knowledge

To adapt the system to the practices and workflows of the clinicians .

To adapt to the settings where these systems are used.

Slide10

CDS in Biomedical Informatics

There are two major types of system used for drawing out the patient’s preferences in clinical decision making:

DA(Decision Aids)

Assists patients in difficult decision making.

It does that by giving the patients information about the various choices about the various treatments available and its outcome.

DA’s need to be working in supplementary of the clinician counseling.

Its seen that the results of the patients working with clinicians and DA’s have been really good.

Slide11

CDS in Biomedical Informatics

DA’s are useful only when a decision is difficult. E.g. more than one treatment recommended, outcomes of the treatment uncertain, complications, tradeoff between outcomes or small chance of a grave outcome.

The drawback with DA’s is that these systems have narrow segment of decisions of choice of treatments hence other systems preferred.

CHOICE(Creating better Health Outcomes by Improving Communication about patients Expectations):

This system is for the clinicians to teach them how to draw out the preferences from a patient.

Slide12

CDS in Biomedical Informatics

It is a system that is mostly used by nurses which collect the preferences from the patient bedside and integrate it in the model.

It has been seen that the congruence of the actual problem patient is having and patient’s self assessment is very high.

It is also easy to use.

Hence, CHOICE is used to develop, implement and evaluate CDSS for SDM.

Slide13

CDS in Biomedical Informatics

Model:

Slide14

CDS in Biomedical Informatics

Slide15

CDS in Biomedical Informatics

Artificial Intelligence is used to process and analyze the data for Clinical Decision support since the framework of system in Biology and Medicine are very complex.

There are a lot of challenges that one has to face to implement these techniques.

These challenges are:

feature selection

Visualization

classification

data warehousing

data mining

analysis of the biological networks

Slide16

CDS in Biomedical Informatics

Key Technical Problems:

Challenges in implementing AI.

Training the Dataset.

Developing generic Algorithm to handle data.

Integrating it with EHR’s

Out of Control Alerts.

Key People Problems:

Using the CDSS.

Poor UI.

Training the people to use the system.

Slide17

CDS in Biomedical Informatics

Conclusion:

Hence, this shows that the use of Clinical decision support system is necessary since it gives an improvement in the patient satisfaction because of his involvement in the decision of the treatment used to cure him/her. It also shows that there are many challenges that we have to face to make it as accurate and efficient as possible. But also one thing is clear that it is still not possible to replace the physician, clinician or nurse it can only assist them and it can never be fully trusted since it can never be 100% accurate.

Slide18

Thank You

Slide19

Questions

Slide20

Slide21

Slide22


About DocSlides
DocSlides allows users to easily upload and share presentations, PDF documents, and images.Share your documents with the world , watch,share and upload any time you want. How can you benefit from using DocSlides? DocSlides consists documents from individuals and organizations on topics ranging from technology and business to travel, health, and education. Find and search for what interests you, and learn from people and more. You can also download DocSlides to read or reference later.