Abstract
This paper addresses the expert system design methodology in determining the Total Parenteral Nutrition (TPN) for neonatal patients. The TPN is recommended for initiation of nutrition for the patients who are unable to tolerate any external feeds. Nutrition plays an important and crucial role in the management of Neonatal patients. TPN calculation depends on the requirement of the patient's calories, volume of fluid, electrolytes, and other less significant factors. The calculation must be consistent with several factors; some of the determining factors are: age, weight, electrolytic status, and clinical condition of the patient. The knowledge based approach will assist the physicians in their determination of the required parameters. This study explains various rules that must be programmed into the knowledge base. The lipids, protein and carbohydrate intake is calculated based on the desired energy. The oral intake is taken into consideration for these calculations. The 'Level-5' Expert System Shell is used as an inference engine. The user interface is menu driven. The built-in inference engine works on the knowledge base to arrive at the desired output. The system performs at a level generally recognized as equivalent to that of a human expert, in this case a pediatrician and nutritionist. The actual Neonatal TPN order sheet used at a University Medical Center was used as a model for design. The paper also addresses some knowledge engineering issues, i.e., the task of extracting knowledge from human experts, and translating it into a rule based format.
| Original language | English |
|---|---|
| Title of host publication | Southern Biomedical Engineering Conference - Proceedings |
| Editors | J.D. Bumgardner, A.D. Puckett |
| Publisher | IEEE |
| Pages | 50-53 |
| Number of pages | 4 |
| State | Published - 1997 |
| Externally published | Yes |
| Event | Proceedings of the 1997 16th Southern Biomedical Engineering Conference - Biloxi, MS, USA Duration: Apr 4 1997 → Apr 6 1997 |
Other
| Other | Proceedings of the 1997 16th Southern Biomedical Engineering Conference |
|---|---|
| City | Biloxi, MS, USA |
| Period | 4/4/97 → 4/6/97 |
Fingerprint
All Science Journal Classification (ASJC) codes
- Engineering(all)
Cite this
Expert system in neonatal total parenteral nutrition. / Varman, Meera; Niccolai, Marino; Varman, Mahendra.
Southern Biomedical Engineering Conference - Proceedings. ed. / J.D. Bumgardner; A.D. Puckett. IEEE, 1997. p. 50-53.Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - Expert system in neonatal total parenteral nutrition
AU - Varman, Meera
AU - Niccolai, Marino
AU - Varman, Mahendra
PY - 1997
Y1 - 1997
N2 - This paper addresses the expert system design methodology in determining the Total Parenteral Nutrition (TPN) for neonatal patients. The TPN is recommended for initiation of nutrition for the patients who are unable to tolerate any external feeds. Nutrition plays an important and crucial role in the management of Neonatal patients. TPN calculation depends on the requirement of the patient's calories, volume of fluid, electrolytes, and other less significant factors. The calculation must be consistent with several factors; some of the determining factors are: age, weight, electrolytic status, and clinical condition of the patient. The knowledge based approach will assist the physicians in their determination of the required parameters. This study explains various rules that must be programmed into the knowledge base. The lipids, protein and carbohydrate intake is calculated based on the desired energy. The oral intake is taken into consideration for these calculations. The 'Level-5' Expert System Shell is used as an inference engine. The user interface is menu driven. The built-in inference engine works on the knowledge base to arrive at the desired output. The system performs at a level generally recognized as equivalent to that of a human expert, in this case a pediatrician and nutritionist. The actual Neonatal TPN order sheet used at a University Medical Center was used as a model for design. The paper also addresses some knowledge engineering issues, i.e., the task of extracting knowledge from human experts, and translating it into a rule based format.
AB - This paper addresses the expert system design methodology in determining the Total Parenteral Nutrition (TPN) for neonatal patients. The TPN is recommended for initiation of nutrition for the patients who are unable to tolerate any external feeds. Nutrition plays an important and crucial role in the management of Neonatal patients. TPN calculation depends on the requirement of the patient's calories, volume of fluid, electrolytes, and other less significant factors. The calculation must be consistent with several factors; some of the determining factors are: age, weight, electrolytic status, and clinical condition of the patient. The knowledge based approach will assist the physicians in their determination of the required parameters. This study explains various rules that must be programmed into the knowledge base. The lipids, protein and carbohydrate intake is calculated based on the desired energy. The oral intake is taken into consideration for these calculations. The 'Level-5' Expert System Shell is used as an inference engine. The user interface is menu driven. The built-in inference engine works on the knowledge base to arrive at the desired output. The system performs at a level generally recognized as equivalent to that of a human expert, in this case a pediatrician and nutritionist. The actual Neonatal TPN order sheet used at a University Medical Center was used as a model for design. The paper also addresses some knowledge engineering issues, i.e., the task of extracting knowledge from human experts, and translating it into a rule based format.
UR - http://www.scopus.com/inward/record.url?scp=0030643338&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=0030643338&partnerID=8YFLogxK
M3 - Conference contribution
SP - 50
EP - 53
BT - Southern Biomedical Engineering Conference - Proceedings
A2 - Bumgardner, J.D.
A2 - Puckett, A.D.
PB - IEEE
ER -