TREATMENT OF MISSING DATA VALUES IN A NEURAL-NETWORK-BASED DECISION-SUPPORT SYSTEM FOR ACUTE ABDOMINAL-PAIN

In this study different substitution methods for the replacement of mi ssing data values were inspected for the use of these cases in a neura l network based decision support system for acute appendicitis. The le ucocyte count had the greatest number of missing values and was used i n the analyses. Four different methods were compared: substituting mea ns, random values, nearest neighbour and a neural network. There were great differences in the substituted leucocyte count values between di fferent methods and only nearest neighbour and neural network agreed a bout most of the cases. The importance of the substitution method for the final diagnostic classification of the patients by the neural netw ork based decision support system was found to be small. (C) 1998 Else vier Science B.V. All rights reserved.