An intelligent ventilation and oxygenation management system in neonatal intensive care using fuzzy trend template fitting

This paper describes an open loop feedback intelligent system for neonatal intensive care management. The system provides a tool enabling the user to make the final decision to accept or reject the advice given. The system collects 18 parameters from the bedside monitor and ventilator using a Medical Information Bus (MIB) system. Comparison between the system's recommendations and seven clinical users (three doctors and four nurses) actions was made during monitoring of seven neonates with gestation age of 27-31 weeks for 124.13 h (mu=17.7329, sigma=5.3843 h, range=10.40-23.85 h). The validation process compared the recommendations triggered by the system with the user feedback (agree, disagree, wait). The system made 191 recommendations in total, 33 of which (17%) were for ventilation and 158 (83%) for oxygenation. The clinician agreed with the system ventilation decisions in 30 occasions (91%) and in 148 occasions for the system oxygenation decisions (94%). The overall percentage of the agreement between the system and the clinician was 93%.

[1]  M. Noshiro,et al.  Fuzzy and conventional control of high-frequency ventilation , 1994, Medical and Biological Engineering and Computing.

[2]  François Pachet,et al.  NéoGanesh: a working system for the automated control of assisted ventilation in ICUs , 1997, Artif. Intell. Medicine.

[3]  P Niederer,et al.  Improved monitoring of preterm infants by Fuzzy Logic. , 1996, Technology and health care : official journal of the European Society for Engineering and Medicine.

[4]  Antonio Giannella-Neto,et al.  Decision support system to assist mechanical ventilation in the adult respiratory distress syndrome , 1997, International journal of clinical monitoring and computing.

[5]  Cungen Cao,et al.  Artifact Detection in the PO2 and PCO2 Time Series Monitoring Data from Preterm Infants , 2004, Journal of Clinical Monitoring and Computing.

[6]  D F Sittig,et al.  Fuzzy classification of hemodynamic trends and artifacts: experiments with the heart rate , 1992, International journal of clinical monitoring and computing.

[7]  M. Dunn,et al.  Approaches to the initial respiratory management of preterm neonates. , 2003, Paediatric respiratory reviews.

[8]  Silvia Miksch,et al.  Utilizing temporal data abstraction for data validation and therapy planning for artificially ventilated newborn infants , 1996, Artif. Intell. Medicine.

[9]  N. McIntosh,et al.  Comparison of two different pulse oximeters in monitoring preterm infants. , 1997, Critical care medicine.

[10]  F Paky,et al.  Pulse oximetry in ventilated preterm newborns: reliability of detection of hyperoxaemia and hypoxaemia, and feasibility of alarm settings , 1995, Acta paediatrica.

[11]  J. Bates,et al.  Automatic control of pressure support mechanical ventilation using fuzzy logic. , 1999, American journal of respiratory and critical care medicine.

[12]  Y Sun,et al.  Fuzzy logic assisted control of inspired oxygen in ventilated newborn infants. , 1994, Proceedings. Symposium on Computer Applications in Medical Care.

[13]  S Y Belal,et al.  A computerized data acquisition system for infusion devices--a clinical support tool, or a risk management tool? , 2001, Journal of medical engineering & technology.

[14]  Silvia Miksch,et al.  Therapy Planning Using Qualtitative Trend Descriptions , 1995, AIME.

[15]  David Roden,et al.  Automatic detection of distorted plethysmogram pulses in neonates and paediatric patients using an adaptive-network-based fuzzy inference system , 2002, Artif. Intell. Medicine.

[16]  J. Højstrup.  A statistical data screening procedure , 1993 .

[17]  W. Northway,et al.  Pulmonary disease following respirator therapy of hyaline-membrane disease. Bronchopulmonary dysplasia. , 1967, The New England journal of medicine.

[18]  R J Kennelly,et al.  Improving acute care through use of medical device data. , 1998, International journal of medical informatics.

[19]  S Snowden,et al.  An expert system to assist neonatal intensive care. , 1997, Journal of medical engineering & technology.

[20]  S Miksch,et al.  Effective data validation of high-frequency data: Time-point-, time-interval-, and trend-based methods , 1997, Comput. Biol. Medicine.

[21]  M Dojat,et al.  A knowledge-based system for assisted ventilation of patients in intensive care units , 1992, International journal of clinical monitoring and computing.

[22]  R M Gardner,et al.  Representative charting of vital signs in an intensive care unit. , 1994, Proceedings. Symposium on Computer Applications in Medical Care.

[23]  R M Gardner,et al.  Perspectives on Development of IEEE 1073: The Medical Information Bus (MIB) Standard , 1997, International journal of clinical monitoring and computing.

[24]  D. Vidyasagar,et al.  Pulse oximetry in neonatal intensive care. , 1989, Clinics in perinatology.

[25]  D Shaw,et al.  An intelligent control system for ventilators. , 1998, Medical engineering & physics.

[26]  A. Mäkivirta,et al.  A knowledge-based alarm system for monitoring cardiac operated patients — technical construction and evaluation , 1993, International journal of clinical monitoring and computing.

[27]  V. Moret-Bonillo,et al.  ESTER: an expert system for management of respiratory weaning therapy , 1989, IEEE Transactions on Biomedical Engineering.

[28]  O Wigertz,et al.  VentEx: an on-line knowledge-based system to support ventilator management. , 1994, Technology and health care : official journal of the European Society for Engineering and Medicine.

[29]  S. Lawless.  Crying wolf: False alarms in a pediatric intensive care unit , 1994, Critical care medicine.