No abstract

An important application of gene expression microarray data is classification of biological samples or prediction of clinical and other outcomes. One necessary part of multivariate statistical analysis in such applications is dimension reduction. This paper provides a comparison study of three dimension reduction techniques, namely partial least squares (PLS), sliced inverse regression (SIR) and principal component analysis (PCA), and evaluates the relative performance of classification procedures incorporating those methods. A five-step assessment procedure is designed for the purpose. Predictive accuracy and computational efficiency of the methods are examined. Two gene expression data sets for tumor classification are used in the study.

Background and Purpose-Cerebral vasospasm continues to be a major cause of poor outcome in patients with ruptured aneurysms. Prophylactic Transluminal Balloon Angioplasty (pTBA) appeared to prevent delayed ischemic neurological deficit in a pilot study. A phase II multicenter randomized clinical trial was subsequently designed. Methods-One hundred and seventy patients with Fisher Grade III subarachnoid hemorrhage were enrolled in the study.Of these, 85 patients were randomized to the treatment group and underwent pTBA within 96 hours after subarachnoid hemorrhage. Main end points of the study included the 3-month dichotomized Glasgow Outcome Score (GOS), development of delayed ischemic neurological deficit (DIND), occurrence of Transcranial Doppler (TCD) vasospasm, and length of stay in the ICU and hospital. Results-The incidence of DIND was lower in the pTBA group (Pϭ0.30) and fewer patients required therapeutic angioplasty to treat DIND (Pϭ0.03). Overall pTBA resulted in an absolute risk reduction of 5.9% and a relative risk reduction of 10.4% unfavorable outcome (Pϭ0.54). Good grade patients had absolute and relative risk reductions of respectively 9.5 and 29.4% (Pϭ0.73). Length of stay in ICU and hospital was similar in both groups. Four patients had a procedure-related vessel perforation, of which three patients died. Conclusions-While the trial is unsuccessful as defined by the primary end point (GOS), proof of concept is confirmed by these results. Fewer patients tend to develop vasospasm after treatment with pTBA and there is a statistically significantly decreased need for therapeutic angioplasty. pTBA does not improve the poor outcome of patients with Fisher grade III subarachnoid hemorrhage. (Stroke. 2008;39:1759-1765.)