St.

Francis Institute of Technology, Mumbai 2011-12

1.0 INTRODUCTION

Clinical decision support (CDS) systems provide clinicians, staff, patients, and other
individuals with knowledge and person-specific information, intelligently filtered and
presented at appropriate times, to enhance health and health care. CDS can provide
support to clinicians at various stages in the care process, from preventive care
through diagnosis and treatment to monitoring and follow-up. CDS as implemented
today can include, for example, order sets tailored for particular conditions or types of
patients (ideally based on evidence-based guidelines and customized to reflect
individual clinicians’ preferences), access to guidelines and other external databases
that can provide information relevant to particular patients, reminders for preventive
care, and alerts about potentially dangerous situations that need to be addressed.
To illustrate this point, imagine the following scenario:

While his doctor is out-of-town, an elderly asthma patient who has developed severe
knee pain sees another physician in his doctor’s office. An EMR (Electronic Medical
Record) provided documentation of the last visit, including recent laboratory results
and a list of the patient’s medications. This information easily brought the doctor up
to date on the patient’s condition. The doctor entered an order for medicine for the
knee pain into the system, printed out a (legible) prescription for the patient, and sent
him on his way. Unfortunately, within 2 months, the patient wound up in the
emergency room with a bleeding ulcer caused by interaction of the pain medicine
with the patient’s asthma medicine.

Problems of this kind occur frequently, as documented in reports from the Institute of
Medicine, USA. Any of several types of CDS tools could have prevented this
patient’s drug interaction. Examples include a pop-up alert to the potential drug
interaction when the doctor prescribed the new medicine; clinical prediction rules to
assess the risks of the pain medication for this patient; clinical guidelines for
treatment of asthma; or reminders for timely follow-up. This scenario illustrates that

-1-
St. Francis Institute of Technology, Mumbai 2011-12

EMRs are the foundation for patient safety and health care quality improvement, but
CDS is an essential element in fully realizing these goals.
Most CDSS consist of three parts, the knowledge base, inference engine, and
mechanism to communicate. The knowledge base contains the rules and associations
of compiled data which most often take the form of IF-THEN rules. If this was a
system for determining drug interactions, then a rule might be that IF drug X is taken
AND drug Y is taken THEN alert user. Using another interface, an advanced user
could edit the knowledge base to keep it up to date with new drugs. The inference
engine combines the rules from the knowledge base with the patient’s data. The
communication mechanism will allow the system to show the results to the user as
well as have input into the system.

The most common use of CDS is for addressing clinical needs, such as ensuring
accurate diagnoses, screening in a timely manner for preventable diseases, or averting
adverse drug events. However, CDS can also potentially lower costs, improve
efficiency, and reduce patient inconvenience. In fact, CDS can sometimes address all
three of these areas simultaneously—for example, by alerting clinicians to potentially
duplicative testing. For more complex cognitive tasks, such as diagnostic decision-
making, the aim of CDS is to assist, rather than to replace, the clinician, whereas for
other tasks (such as presentation of a predefined order set) the CDS may relieve the
clinician of the burden of reconstructing orders for each encounter. The CDS may
offer suggestions, but the clinician must filter the information, review the suggestions,
and decide whether to take action or what action to take.

-2-
St. Francis Institute of Technology, Mumbai 2011-12

1.1 AIM AND OBJECTIVES

Aim:
Knowledge based Clinical Decision Support Systems (CDSS) aims to directly assist
physicians and other health professionals with decision making tasks.

Objectives:
-Knowledge Based CDSS are intended to support the clinician, rather than replace
them.
-CDSS assist the clinician to interact with the CDSS utilizing both the clinician
knowledge and the CDSS to make a better analysis of patient data.
-CDSS makes suggestion of output or a set of outputs for the clinician.
-Clinician can take useful information and removes redundant CDSS suggestion.
-To capture the knowledge of master clinician
-To assist clinician to point of care
-CDSS can notify health care professionals of drug-drug interactions, contraindicated
drugs, alternative medications, cost of interventions and abnormal laboratory results
-CDSS provide an opportunity to improve care
-Alerting the care provider to situations of concern, critiquing previous decisions,
suggesting interventions at the direct request of a care a care provider and
retrospective quality assurance reviews

-3-
St. Francis Institute of Technology, Mumbai 2011-12

2.0 LITERATURE SURVEY

[1]. David Robbins, Varadraj Gurupur, John Tanik, “Information architecture of a

clinical decision support system”, Southeastcon, 2011 Proceedings of IEEE, pp 374 –
378, April 2011.
ABSTRACT:
Recent healthcare legislation, the move to electronic health records, and the
accelerating pace of medical knowledge discovery combine to create an environment
in which clinical decision support systems are a necessary part of most medical
processes. Based on a review of current best practices, we present the information
architecture of a clinical decision support system comprised of four primary
components: patient model, treatment library, intelligent agents, and an authenticated
knowledge base. The patient model contains all patient information in the form of
measurable parameters, treatment history, and health goals. Information regarding
treatment procedures, dosages, and effects is stored in treatment library. Intelligent
agents are used to optimize courses of treatments to best achieve desired health goals.
Finally, an authenticated knowledge base containing the most recent medical research
is used to keep the patient model and treatment library up to date.

[2]. Kim, J.A.; InSook Cho; Yoon Kim, “CDSS (Clinical Decision Support
System) Architecture in Korea”, IEEE International Conference on Convergence and
Hybrid Information Technology, 2008, pp 700 – 703, September 2008
ABSTRACT:
We present architecture for implementing independent, extensible, and interoperable
clinical decision support service in perspective of EHR. In this architecture,
components for implementing independent knowledge service and interface
mechanism with EHR service or existing hospital information system are identified.
In our architecture knowledge engine is critical component for implementing
knowledge service. In this paper, we suggested interoperable CDSS (clinical decision
support system) architecture and knowledge engine architecture.

-4-
St. Francis Institute of Technology, Mumbai 2011-12

[3]. Ceccarelli M.; Donatiello A.; Vitale D., “KON^3: A Clinical Decision Support
System, in Oncology Environment, Based on Knowledge Management “, 20th IEEE
International Conference on Tools with Artificial Intelligence, 2008. ICTAI '08, pp
206 – 210, November 2008.
ABSTRACT:
The application of scientific methodology to clinical practice is typically realized
through recommendations, policies and protocols represented as Clinical Practice
Guidelines (CPG). CPG help the clinician in his choices, improving the patient care
process. The representation of Guidelines and their introduction in medical
information system can lead to efficient Clinical Decision Support Systems (CDSS);
however this poses several interesting challenges as problems of knowledge
representation, inference, workflow definition, and access to unstructured data, etc. In
this paper we analyze approaches and methods in computer-based CPG realization
and illustrate the choices made (tools, architecture and clinical domain) to realize the
KON3 system to achieve a CDSS and semantic information representation, in
oncology environment.

[4]. Wu, F.; Williams M.; Kazanzides P.; Brady K.; Fackler J, “A modular Clinical
Decision Support System Clinical prototype extensible into multiple clinical settings”,
3rd IEEE International Conference on Pervasive Computing Technologies for
Healthcare, 2009. Pervasive Health 2009, pp 1 – 4, August 2009.
ABSTRACT:
Traditionally, Clinical Decision Support Systems (CDSS) collect patient data from
physiological monitors and other sources, providing clinicians with derived
instructions and information to aid treatment planning. With advancements in
telecommunication networks, CDSS functionality can be extended over distances, and
accessed remotely (e.g. by appropriate healthcare providers not available in the
patient's immediate surroundings). This paper discusses a modular CDSS that features
real-time continuous patient monitoring, high-fidelity analysis and incorporation of

-5-
St. Francis Institute of Technology, Mumbai 2011-12

clinical guidelines for decision support. A modular CDSS prototype was designed,
implemented and tested in a pediatric intensive care environment, by incorporating a
guideline for pediatric traumatic brain injury (TBI). System outputs show successful
aggregation and analysis of continuous and periodic data and automation of
guidelines by recognizing deviation of patient's condition from normal states. The
modular design will allow extension into pre-hospital treatment environments by
taking advantages of advances in pervasive monitoring.

[5]. Courtright, C.G.; Crawford, R.S.; Klubert, D.M., “Criteria for developing
clinical decision support systems”, 14th IEEE Symposium on Computer-Based
Medical Systems, 2001, pp 270 – 275, August 2002.
ABSTRACT:
The use of archived information and knowledge derived from data-driven system,
both at the point of care and retrospectively, is critical to improving the balance
between healthcare expenditure and healthcare quality. Data-driven clinical decision
support, augmented by performance feedback and education, is a logical addition to
consensus- and evidence-based approaches on the path to widespread use of
intelligent search agents, expert recognition and warning systems. We believe that
these initial applications should (a) capture and archive, with identifiable end-points,
complete episode-of-care information for high-complexity, high-cost illnesses, and (b)
utilize large numbers of these cases to drive risk-adjusted “individualized”
probabilities for patients requiring care at the time of intervention

-6-
St. Francis Institute of Technology, Mumbai 2011-12

3.0 EXISTING SYSTEM

MYCIN was a rule-based expert system designed to diagnose and recommend

treatment for certain blood infections (antimicrobial selection for patients with
bacteremia or meningitis). It was later extended to handle other infectious diseases.
Clinical knowledge in MYCIN is represented as a set of IF-THEN rules with certainty
factors attached to diagnoses. It was a goal-directed system, using a basic backward
chaining reasoning strategy (Resulting in exhaustive depth-first search of the rules
base for relevant rules though with additional heuristic support to control the search
for a proposed solution).
MYCIN was developed in the mid-1970s by Ted Shortliffe and colleagues at Stanford
University.
It is probably the most famous early expert system, described by Mark Musen as
being "the first convincing demonstration of the power
Of the rule-based approach in the development of robust clinical decision-support
systems" [Musen, 1999].

The EMYCIN (Essential MYCIN) expert system shell, employing MYCIN's control
structures was developed at Stanford in 1980.This domain-independent framework
was used to build diagnostic rule-based expert systems such as PUFF, a system
designed to interpret pulmonary function tests for patients with lung disease.

INTERNIST-I was a rule-based expert system designed at the University of

Pittsburgh in 1974 for the diagnosis of complex diagnosis of complex problems in
general internal medicine. It uses patient observations to deduce a list of compatible
disease states (based on a tree-structured database that links diseases with symptoms).
By the early 1980s, it was recognized that the most valuable product of the system
was its medical knowledge base. This was used as a basis for successor systems
including CADUCEUS and Quick Medical Reference (QMR), a commercialized
diagnostic DSS for internists.

-7-
St. Francis Institute of Technology, Mumbai 2011-12

4.0 PROBLEM STATEMENT

Time is one of the most critical aspects of success in today’s modern world.
There are new discoveries and inventions being made by the minute. The field of
medicine itself has seen the rise of previously unknown diseases, some thought to be
affecting only animals such as the swine flu (pigs) and the avian flu (birds).
Today’s physician is overwhelmed with the amount of data with respect to
diseases as well as the insufficient time for diagnosis and treatment. Moreover, the
error rate in the diagnosis as well as prognosis of a disease is alarmingly high.
Giving a computer based solution to such problems is the best way to ease the
difficulty of doctors as well as improve the quality of medical care. A computer based
system can help to yield knowledge in any dimension by simple methods.
Medical artificial intelligence is primarily concerned with the construction of
AI programs that perform diagnosis and make therapy recommendations. Unlike
medical applications based on other programming methods, such as purely statistical
and probabilistic methods, medical AI programs are based on symbolic models of
disease entities and their relationship to patient factors and clinical manifestations.
These systems can also provide additional support in the form of alerts (e.g. If a
patient has been prescribed two medicines having potentially harmful side effects if
used together), reminders (e.g. If a patient has not come for his scheduled
appointment or if prescribed drug is not available or is banned), diagnostic assistance
etc.
With this objective in mind, we proceed to build a system to assist clinicians
with diagnosis of diseases and their treatments.

-8-
St. Francis Institute of Technology, Mumbai 2011-12

5.0 SCOPE OF THE PROJECT

Business Goals:
 To make high-quality medical care affordable.
 To provide assistance to the clinician in making a good and precise diagnosis.
 To provide alerts and reminders to the clinician during drug prescription.

Technical Goals:
 To harness the utility of artificial intelligence for improving the quality of
medical care.
 To increase the speed and quality of clinical process.

-9-
St. Francis Institute of Technology, Mumbai 2011-12

6.0 CONSTRAINTS OF THE PROJECT

Business Constraints:
 Net connection is required to update the knowledge base.
 Keyboard is required.
 The system lacks human instinct.
Technical Constraints:
 AIML interpreter is required to run the software
 A doctor must have some technical knowledge about the CDSS software

-10-
St. Francis Institute of Technology, Mumbai 2011-12

7.0 PROPOSED SYSTEM

The following applications can be provided by the CDSS:

 Alerts and Reminders: An expert system connected to patient monitoring
devices can provide real time alerts and warnings to medical staff in the case
of a change in the patient's condition. They can also be used for less critical
situations to provide tasks lists for clinicians, such as post-op checks.
 Diagnostic Assistance: Patient data can be compared to the system knowledge
base in order to present possible diagnoses. This is beneficial when the
clinician is inexperienced (i.e. a GP trying to assess a neurological problem) or
when the patient's symptoms are complex or seemingly unrelated.
 Prescription Decision Support: A specialized CDSS known as a Prescribing
Decision Support System (PDSS) can check for drug-to-drug interactions,
dosage errors, and drug contraindications such as patient allergies. Many
systems also enable automated script generation and electronic transmission to
the pharmacy. PDSS is one of the most widely accepted and used CDSS.
 Information Retrieval: CDSS can assist in the location and retrieval of
appropriate and accurate clinical data which may be used for diagnosis or
treatment planning. Depending on the complexity of the system, the CDSS
may simply perform as a filter to search queries or be able to assess the
importance, applicability, or utility of the information it retrieves.

-11-
St. Francis Institute of Technology, Mumbai 2011-12

8.0 METHODOLOGY

-12-
St. Francis Institute of Technology, Mumbai 2011-12

9.0 REQUIREMENTS GATHERINGAND PLANNING

9.1 REQUIREMENTS ELICITATION
9.1.1. USE CASE DIAGRAM AND DESCRIPTION

 Administrator:
The administrator is the one who maintains the database by
updating, adding or deleting the data. The administrator has a user
Id and password to log in.

Fig: 9.1(a) Use Case of Admin

 System:
The system validates the user using the user ID and password. The
system also provides various diagnoses to the entered symptoms. It
also provides reminders and alerts.

Fig: 9.1(b) Use Case of System

 doctor:
The doctor can either be a registered doctor i.e. registered with the
system or a non registered doctor. Non registered doctors have to
first register with the system, whereas a registered doctor can
access the system for making a sound decision by entering
symptoms and finding the possible diagnosis. A registered doctor
will also get reminders about any possible side effects due to
prescribed medicine and alerts for the same.

-13-
St. Francis Institute of Technology, Mumbai 2011-12

9.2 FEASIBILITY STUDY

9.2.1 TECHNICAL FEASIBILITY

Hardware requirements:
 Intel Pentium Core2 Duo Processor
 80GB HDD
 Minimum of 1GB RAM
Software requirements:
 Microsoft Windows XP or higher
 AIML,SQL

9.2.2 ECONOMIC FEASIBILTY

Higher level of automation requires more funds. Hence based on hardware and
software specification a desirable alternative is selected by performing cost/benefit
analysis i.e. the process of comparing cost and benefits to see if the investments made
in creating or developing a new system is costlier or beneficial.
By implementing the application at the clinician’s premises, the system will
cut down its own cost of operations by not doing a lot of manual work. This will help
them to carry more important and revenue generating activities at their premises and
unnecessarily not wasting their limited budgeted amount on less important activities.

-14-
St. Francis Institute of Technology, Mumbai 2011-12

9.3 REQUIREMENTS ANALYSIS

9.3.1 FLOW CHART

Fig: 9.3.1 Flow Chart

-15-
St. Francis Institute of Technology, Mumbai 2011-12

9.4 TIMELINE CHARTS

Fig: 9.4 Time Line Chart

-16-
St. Francis Institute of Technology, Mumbai 2011-12

9.5 W.B.S. CHART

Fig: 9.5 WBS Chart

-17-
St. Francis Institute of Technology, Mumbai 2011-12

10. ANALYSIS
10.1 ER DIAGRAM

Fig: 10.1 ER Diagram

-18-
St. Francis Institute of Technology, Mumbai 2011-12

DATA DICTIONARY

-19-
St. Francis Institute of Technology, Mumbai 2011-12

10.2 ACTIVITY DIAGRAM

Fig: 10.2(a) Activity Diagram Administrator

-20-
St. Francis Institute of Technology, Mumbai 2011-12

Non-Registered Users

Fig: 10.2(b) Activity Diagram None registered User

Fig: 10.2(c) Activity Diagram Registered

-21-
St. Francis Institute of Technology, Mumbai 2011-12

10.3 DATA FLOW DIAGRAM

Fig: 10.3(a) DFD Level 0

Fig: 10.3(b) DFD Level 1

Fig: 10.3(c) DFD Level 2

Fig: 10.3(d) DFD Level 2

Fig: 10.3(e) DFD Level 3

Fig: 10.3(f) DFD Level 3

-22-
St. Francis Institute of Technology, Mumbai 2011-12

11.0 DESIGN
11.1 UI DESIGN (LAYOUT)/SNAPSHOTS

-23-
St. Francis Institute of Technology, Mumbai 2011-12

-24-
St. Francis Institute of Technology, Mumbai 2011-12

12.0 APPENDIX

 AI Artificial Intelligence.
 ALS Adaptive Learning System.
 Decentralized The process of dispersing decision making closer to the
people
 Hypermedia The use of graphics, audio, video, plain text and
hyperlinks intertwine to create a generally non-linear medium of
information.
 Paradigm An example or pattern, distinct concept.
 Pervasive Extensive, spread throughout

-25-
St. Francis Institute of Technology, Mumbai 2011-12

13.0 BIBLIOGRAPHY AND REFERENCES

A) Paper references:

[1]. Y.S. Abu-Mostafu, A.F.Atiya, M. magdon-Ismail, and H.White, “Introduction to the

special issue on neural networks in financial engineering”, IEEE Transaction Neural
Networks, vol.12, no.4, pp.653-656, July 2001
Websites
[1] www.openclinical.org/dss.html
[2] www.wikipedia.com
[3] www.java2s.com
[4] www.adaptivelearning.com

-26-