International Journal of Engineering and Technical Research (IJETR)

ISSN: 2321-0869, Volume-1, Issue-8, October 2013

Research Review of Expert Systems for Newborns

Mrs. M. K. Patkar, Dr. R.V. Kulkarni

 relationships within a dataset can be used in the diagnosis,

Abstract— One of the most popular technology of AI is nothing treatment and prediction in many clinical scenarios.
but Expert system. It is rapidly growing technology. An expert
system is a computer system that emulates, or acts in all respects, Today, it can be noticed that most modern hospitals are
with the decision-making capabilities of a human expert. Expert becoming smart hospitals, digitalizing their administrative
systems have been developed in different areas like medicine processes, incorporating latest technology equipment, storing
,engineering and business. Author wants to review the expert their images in PACS (Picture Archiving and Communication
systems for newborns.
System) and RIS (Radiology Information System), using
diagnostic expert systems, smart labels for tracking patients
Index Terms— Artificial Intelligence, Expert system,
and mobile medical equipment, and making use of smart
rule-based system
clothes to monitor pregnant women, children, elderly and
patients with particular physical disabilities.
I. INTRODUCTION
II. LITERATURE REVIEW
Expert systems are an offspring of the more general area of
study known as artificial intelligence (AI). The spectrum of The Children's Hospital in Ottawa is using artificial
applications of expert systems technology to industrial and intelligence to gather information on newborns with critical
commercial problems is so wide as to defy easy illnesses. The data collected is used to suggest treatment
characterization. The applications find their way into most approaches and also to help predict and improve health
areas of knowledge work. They are as varied as helping outcomes. Until now, these methods have only been used in
salespersons sell modular factory-built homes to helping adult medicine and this is one of the few approaches used on
NASA plan the maintenance of a space shuttle in preparation newborns. They use monitoring systems which are hooked up
for its next flight. to each baby in the unit to collect and store data such as
respiratory rates and heart beat rhythms. From this data, the
Applications tend to cluster into seven major classes. technology can predict outcomes like chance for survival and
the length of the hospital stay. This technology has an
1) Diagnosis and Troubleshooting of Devices and Systems of accuracy rate of over 95% in predicting survivors. Doctors are
All Kinds hoping to be able to use this to predict rates of complication in
common health problems in infants. Since the data is
2) Planning and Scheduling collected continuously by the monitoring systems, when
babies develop complications the doctors receive warning
3) Configuration of Manufactured Objects from signs right away that allows them to treat the problem. It is
Subassemblies like using instant communication to let the doctors know
when there is a crisis to be averted. So far, this system has
4) Financial Decision Making been very consistent and accurate. It is much more efficient
than human observation 24 hours a day. Over the next couple
5) Knowledge Publishing
of years, hundreds of new cases (babies) will be added to the
6) Process Monitoring and Control baby database, where more clinical trials using this
technology will be used.
7) Design and Manufacturing
Jirapaet V.from Chulalongkorn University, Bangkok
10330, Thailand developed a computer expert system
prototype for mechanically ventilated neonates and impact on
Medical diagnosis was one of the first knowledge areas to clinical judgment and information access capability of
which ES technology was applied. Artificial intelligence is a nurses.The system is a multimedia interactive
branch of computer science capable of analyzing complex consultation-based program that contains 2 major parts: the
medical data. Its techniques can be applied in almost every nursing diagnosis and the knowledge base on nursing care of
field of medicine. Their potential to exploit meaningful mechanically ventilated neonates. A rule-based (Boolean
frame) was chosen for the nursing diagnosis decision model.
According to developer computer expert system is an
Manuscript received September 30, 2013. alternative method of training and providing real-time clinical
Mrs. M. K. Patkar, Research student, CSIBER,Kolhapur, India.
decision support for nurses to advance their practices from a
Dr. R.V. Kulkarni, Professor & H.O.D.Dept.Computer Science, novice to a proficient level. The purpose of this study was to
CSIBER,Kolhapur, India. develop a prototype of a computer expert system for

26 www.erpublication.org
 Research Review of Expert Systems for Newborns

mechanically ventilated neonates (ES-MVN) and to assess the than in a group of patients who died. The overall agreement of
impact of the ES-MVN on the clinical judgment and the AVES-N advice and real therapeutic actions leads to the
information access capability of nurses. Lance Everett clinical evaluation of the expert system. The differences can
Rodewald from University of Illinois Urbanna,Champaign be attributed to a) different therapeutic strategies at 2 NICU's,
developed BABY an expert system that monitors new-born, b) missing data regarding complications in the data base
intensive care unit, on-line patient data. The system monitors which were not taken into account by the expert system.
the data, looks for significant patterns and suggests further
evaluation. BABY also tracks the clinical status of the A number of expert systems to assist nurses have been
new-borns and can answer questions about each patient.The developed in recent years. One such system is called Expert
system will help clinicians and computer scientists. The Nurse., This system allows nurses to rapidly input patient data
clinicians will most likely be interested in BABY for its and obtain all known possible diagnoses which can be
approach to the common problem or finding significant reached from known patient data, specific patient data which
information and patterns in potentially large amounts or support each diagnosis, and suggested patient goals which
medical data. Patient monitoring with the use of a logical may be modified for individual patients. Evaluation data
model or a patient within the context of a specific medical collected in the use of this system shows nurses spent less time
knowledge base should also be of interest. Of more relevance arriving at a nursing diagnosis, identified a greater number of
to the computer scientists will be the knowledge engineering diagnostic possibilities and increased the quality of patient
techniques used in the implementation. The knowledge records.
incorporates control information to reduce compu-tational
complexity and knowledge storage costs so that the Another successful expert system in the nursing domain is
knowledge base will not become unmanage-able as its size CANDI (Computer Aided Nursing Diagnosis and
increases. The design of BABY was guided by the role Intervention) It is used to assist in assigning nursing diagnoses
envisioned for it in the nursery. based on clinical assessment data. CANDI uses a rule-based
BABY's task is to find clinically important patterns in the system that processes the data entered by the nurse on the
medical and demographic data about NICU patients. It is computer during an assessment. The inference engine takes
targeted specifically on the NICU for two reasons-there is a into account the possibility of the presence of multiple
need for a system like diagnoses.
BABY, and the chance for success is good due to peculiarities Parental Nutrition Solutions (PNS)project is a joint
or neonatology. There are few areas in medicine where the cooperation of the Department of Medical Cybernetics and
amount of data, especially numeric data, is so great, and Artificial Intelligence the Austrian Research Institute for
where the vast majority or it can be made available on-line for Artificial Intelligence and the Neonatal Intensive Care Unit
a computer system machines already for many physiologic (NICU) of the Department of Pediatrics of the University of
parameters and have been used for extraction or significant Vienna.
events from the stream or signals coming from monitoring The aim of the project was to develop an expert system
equipment. In contrast with adult medicine, the relative representing the clinical and theoretical knowledge about the
importance or the monitoring data to physical exam findings composition of parenteral nutrition solutions for newborn
is greater because the babies often do not demonstrate infants treated at neonatal intensive care units (NICUs).
obvious physical signs with serious disease. Additionally, the
past medical history or a newborn is much more concise than Planning of an adequate nutritional support for maintaining
that or an adult. the metabolic needs of sick newborn infants is time
Although the variety or clinical problems encountered in the consuming, needs experts' knowledge and involves the risk of
nursery is large, it is practically limited to a few common introducing possibly fatal errors. Recent systems used for
diseases and complications. composing parenteral nutrition solutions mainly support the
calculation and the documentation process and cannot easily
be adapted for neonates. Computerized expert system
Michinikowski M.and team ,Institute of Biocybernetics technology may help to develop time saving solutions to a
and Biomedical Engineering PAS, Warsaw, Poland evaluated given problem and to avoid errors within certain limits. It is an
the artificial ventilation expert system for neonates (AVES-N) interactive expert system for calculating the composition of
using archival data. parenteral nutrition solutions (PNS) for newborn infants.
The recommendations of the system were compared to the
decisions made by the expert-physician in the same clinical The knowledge base of the expert system consists of the rules
situation (patient condition, respirator settings). In this study for composing the PNS according to heuristic rules used at the
they used data of 320 newborns which were ventilated in the cooperating NICU. Applying these rules, the daily fluids,
Neonatal Intensive Care Unit of the Vanderbilt University electrolytes, vitamins, and nutritional requirements were
Hospital in Nashville (USA). Best agreement between the calculated according to the estimated needs, the patient's body
recommendations of the system and the decisions of the weight, the age, and the individual tolerance. The
experts was found for positive end expiratory pressure requirements were also corrected according to the daily
(PEEP), inspired oxygen fraction (FiO2) and peak inspiratory measurements of serum electrolytes, triglycerides and protein
pressure (PIP)--about 70%. Worse agreement was found for if available. Glucose supply was adjusted depending on the
time related parameters: respiratory frequency (f) - 54%, time type of venous access used (peripheral or central venous line),
of inspiration (ti) - 46%, time of next blood gas analysis - on the glucose tolerance and on the total fluid allowances.
15%. The expert system advised lower FiO2 PEEP and f. The Finally, the PNS was reduced according to the proportion of
differences were smaller in a group of patients who survived oral feedings. The program works interactively asking for

27 www.erpublication.org
 International Journal of Engineering and Technical Research (IJETR)
ISSN: 2321-0869, Volume-1, Issue-8, October 2013

relevant data, calculating the PNS, and displaying the results.

The physician has the choice to adjust calculated values
according to special clinical requirements. The final output is
a PNS schedule that can be used directly in the case history of
neonates. Possible input and dosage errors are eliminated by
methods of data validation using body weight and age
dependent thresholds.

Dr. C.A.Holzman and team developed an expert system for

automated classification of the sleep/waking states in human
infants; i.e. active or rapid-eye-movement sleep (REM), quiet
or non-REM sleep (NREM), including its four stages,
indeterminate sleep (IS) and wakefulness (WA). A model to
identify these states, introducing an objective formalisation in
terms of the state variables characterising the recorded
patterns, is presented. The following digitally recorded
physiological events are taken into account to classify the
sleep/waking states: predominant background activity and the
existence of sleep spindles in the electro-encephalogram;
existence of rapid eye movements in the electro-oculogram;
and chin muscle tone in the electromyogram. Methods to
detect several of these parameters are described. An expert
system based on artificial ganglionar lattices is used to
classify the sleep/waking states, on an off-line
minute-by-minute basis. Algorithms to detect patterns
automatically and an expert system to recognise sleep/waking
states are introduced, and several adjustments and tests using
various real patients are carried out. Results show an overall
performance of 96.4% agreement with the expert on
validation data without artefacts, and 84.9% agreement on
validation data with artefacts. Moreover, results show a
significant improvement in the classification agreement due
to the application of the expert system, and a discussion is
carried out to justify the difficulties of matching the expert's
criteria for the interpretation of characterising patterns.

III. CONCLUSION
Today many new problems or diseases are occuring to babies.
Researcher wants to design and develop a rule-based system
for different diseases or problems faced by baby and possible
causes of it by concerning experts from the field.Junior
doctors or nurses may get help from the proposed system.

REFERENCES
[1] Decision Support Systems-V.S.Jankiraman , K Sarukesi
[2] Peter Jackson-Introduction to Expert Systems
[3] BABY -Expert System for patient monitoring in a newborn ICU-Lance
Everett Rodewald-B.S.University,Illinois,1976
[4] Evaluation of expert system for respiratory therapy of newbornson archival
data-Michinikowski M,Rudowski R,Siugocki P,Graboski J,Rondoo
Z,Lindstrom DP,Institute of Biocybernetics and Biomedical
Engineering,Poland,1997
[5] Parental Nutrition System-Dept. of Medical Cybernetics and AI,the Austrian
Research Institute for Artificial Intelligence and the Neonatal Intensive
Care Unit (NICU) of the Department of Pediatrics of the University of
Vienna.
[6] Expert system for Children in Ottawa-Frize M,Walker R School of It and
Engineering, Canada,2000

28 www.erpublication.org