TY - JOUR
T1 - Application of clustering analyses to the diagnosis of Huntington disease in mice and other diseases with well-defined group boundaries
AU - Nikas, Jason B.
AU - Low, Walter C.
PY - 2011/12
Y1 - 2011/12
N2 - Nuclear magnetic resonance (NMR) spectroscopy has emerged as a technology that can provide metabolite information within organ systems in vivo. In this study, we introduced a new method of employing a clustering algorithm to develop a diagnostic model that can differentially diagnose a single unknown subject in a disease with well-defined group boundaries. We used three tests to assess the suitability and the accuracy required for diagnostic purposes of the four clustering algorithms we investigated (K-means, Fuzzy, Hierarchical, and Medoid Partitioning). To accomplish this goal, we studied the striatal metabolomic profile of R6/2 Huntington disease (HD) transgenic mice and that of wild type (WT) mice using high field in vivo proton NMR spectroscopy (9.4. T). We tested all four clustering algorithms (1) with the original R6/2 HD mice and WT mice, (2) with unknown mice, whose status had been determined via genotyping, and (3) with the ability to separate the original R6/2 mice into the two age subgroups (8 and 12 weeks old). Only our diagnostic models that employed ROC-supervised Fuzzy, unsupervised Fuzzy, and ROC-supervised K-means Clustering passed all three stringent tests with 100% accuracy, indicating that they may be used for diagnostic purposes.
AB - Nuclear magnetic resonance (NMR) spectroscopy has emerged as a technology that can provide metabolite information within organ systems in vivo. In this study, we introduced a new method of employing a clustering algorithm to develop a diagnostic model that can differentially diagnose a single unknown subject in a disease with well-defined group boundaries. We used three tests to assess the suitability and the accuracy required for diagnostic purposes of the four clustering algorithms we investigated (K-means, Fuzzy, Hierarchical, and Medoid Partitioning). To accomplish this goal, we studied the striatal metabolomic profile of R6/2 Huntington disease (HD) transgenic mice and that of wild type (WT) mice using high field in vivo proton NMR spectroscopy (9.4. T). We tested all four clustering algorithms (1) with the original R6/2 HD mice and WT mice, (2) with unknown mice, whose status had been determined via genotyping, and (3) with the ability to separate the original R6/2 mice into the two age subgroups (8 and 12 weeks old). Only our diagnostic models that employed ROC-supervised Fuzzy, unsupervised Fuzzy, and ROC-supervised K-means Clustering passed all three stringent tests with 100% accuracy, indicating that they may be used for diagnostic purposes.
KW - Clustering analyses
KW - Diagnostic methods
KW - Fuzzy Clustering
KW - Hierarchical Clustering
KW - Huntington disease
KW - K-means Clustering
KW - Medoid Partitioning Clustering
KW - Metabolomics
KW - Nuclear magnetic resonance spectroscopy
KW - Receiver operating characteristic (ROC) curve analysis
UR - http://www.scopus.com/inward/record.url?scp=80655127860&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=80655127860&partnerID=8YFLogxK
U2 - 10.1016/j.cmpb.2011.03.004
DO - 10.1016/j.cmpb.2011.03.004
M3 - Article
C2 - 21529982
AN - SCOPUS:80655127860
VL - 104
SP - e133-e147
JO - Computer Methods and Programs in Biomedicine
JF - Computer Methods and Programs in Biomedicine
SN - 0169-2607
IS - 3
ER -