Dealing with missing values in a clinical case-based reasoning system

In clinical case-based reasoning systems, missing values in the case-base pose a common but serious problem that impairs the performance of the system. Imputation of missing values is popular in many knowledge-based systems. However, applications need to be adjusted to take into account that imputed values are only estimates of original values. In this paper, we introduce an imputation method that exploits the correlation between clinical data by filtering the case-base according to attributes correlated to the missing attribute. We also present a framework that shows how any imputation method can be used in a case-based reasoning system to account for the inherent uncertainty of imputation and to reflect the quality of the imputation method in the similarity calculation. The filter imputation method and the imputation framework are evaluated using a case-based reasoning decision support system developed for radiotherapy treatment planning for prostate cancer.

[1]  Judi Scheffer,et al.  Dealing with Missing Data , 2020, The Big R‐Book.

[2]  Bodo B. Schlegelmilch,et al.  Taking the Fear out of Data Analysis , 1998, J. Oper. Res. Soc..

[3]  SongQinbao,et al.  A new imputation method for small software project data sets , 2007 .

[4]  Ito Wasito,et al.  Nearest neighbours in least-squares data imputation algorithms with different missing patterns , 2006, Comput. Stat. Data Anal..

[5]  Francesco Ricci,et al.  Learning a Local Similarity Metric for Case-Based Reasoning , 1995, ICCBR.

[6]  Sanja Petrovic,et al.  A Case-Based Reasoning Approach to Dose Planning in Radiotherapy , 2007 .

[7]  Witold Pedrycz,et al.  A Novel Framework for Imputation of Missing Values in Databases , 2007, IEEE Transactions on Systems, Man, and Cybernetics - Part A: Systems and Humans.

[8]  David McSherry.  Precision and Recall in Interactive Case-Based Reasoning , 2001, ICCBR.

[9]  Xiao-Li Meng,et al.  Applications of multiple imputation in medical studies: from AIDS to NHANES , 1999, Statistical methods in medical research.

[10]  Chengqi Zhang,et al.  POP algorithm: Kernel-based imputation to treat missing values in knowledge discovery from databases , 2009, Expert Syst. Appl..

[11]  D. Rubin.  INFERENCE AND MISSING DATA , 1975 .

[12]  Rainer Schmidt,et al.  A CBR Solution for Missing Medical Data , 2007 .

[13]  M. Saeed,et al.  Estimation of missing values in clinical laboratory measurements of ICU patients using a weighted K-nearest neighbors algorithm , 2004, Computers in Cardiology, 2004.

[14]  Qinbao Song,et al.  A new imputation method for small software project data sets , 2007, J. Syst. Softw..

[15]  M. Frize,et al.  Discrimination of Inconsistencies in Medical Data , 2008, 2008 IEEE International Workshop on Medical Measurements and Applications.