We investigate the problem of classifying individuals based on estimated density functions for each individual. Given labelled histograms characterizing red blood cells (RBCs) for different individuals, the learning problem is to build a classifier which can classify new unlabelled histograms into normal and iron deficient classes. Thus, the problem is similar to conventional classification in that there is labelled training data, but different in that the underlying measurements are not feature vectors but histograms or density estimates. We describe a general framework based on probabilistic hierarchical models for modelling such data and illustrate how the model lends itself to classification. We contrast this approach with two other alternatives: (1) directly defining distance between densities using a cross-entropy distance measure, and (2) using parameters of the estimated densities as feature vectors for a standard discriminative classification framework. We evaluate all three methods on a real-world data set consisting of 180 subjects. The hierarchical modeling and density-distance approaches are most accurate, yielding cross-validated error rates in the range of 1 to 2%. We conclude by discussing the relative merits of each approach, including the interpretability of each model from a clinical diagnostic viewpoint.

Binning and truncation of data is common in data analysis and machine learning. This paper addresses the problem of fitting mixture densities to multivariate binned and truncated data. The EM approach proposed by McLachlan and Jones (1988) for the univariate case is generalized to multivariate measurements. The multivariate solution requires the evaluation of multidimensional integrals over each bin at each iteration of the EM procedure. Naive implementation of the procedure can lead to computationally inefficient results. To reduce the computational cost a number of straightforward numerical techniques are proposed. Results on simulated data indicate that the proposed methods can achieve significant computational gains with no loss in the accuracy of the final parameter estimates. Furthermore, experimental results suggest that with a sufficient number of bins and data points it is possible to estimate the true underlying density almost as well as if the data had not been binned. The paper concludes with a brief description of an application of this approach to diagnosis of iron deficiency anemia, in the context of binned and truncated bivariate measurements of volume and hemoglobin concentration from an individual's red blood cells.

The polyamine-inhibitory regimen difluoromethylornithine (DFMO)+sulindac has marked efficacy in preventing metachronous colorectal adenomas. Polyamines are synthesised endogenously and obtained from dietary sources. Here we investigate dietary polyamine intake and outcomes in the DFMO+sulindac colorectal adenoma prevention trial.

Dietary polyamine data were available for 188 of 267 patients completing the study. Total dietary polyamine content was derived by the sum of dietary putrescine, spermine and spermidine values and categorised into two groups: highest (>75-100%) vs the lower three quartiles (0-25, 25-50 and 50-75%). Baseline tissue polyamine concentration and ODC1 genotype were determined. Logistic regression models were used for risk estimation.

A significant interaction was detected between dietary polyamine group and treatment with regard to adenoma recurrence (P=0.012). Significant metachronous adenoma risk reduction was observed after DFMO+sulindac treatment in dietary polyamine quartiles 1-3 (risk ratio (RR) 0.19; 95% confidence interval (CI) 0.08-0.42; P<0.0001) but not in quartile 4 (RR 1.51; 95% CI 0.53-4.29; P=0.44). However, a lower number of events in the placebo group within dietary quartile 4 confound the aforementioned risk estimates.

These preliminary findings reveal complex relationships between diet and therapeutic prevention, and they support further clinical trial-based investigations where the dietary intervention itself is controlled.

The Bowman-Birk inhibitor (BBI) is a soybean-derived serine protease inhibitor and a potential cancer chemopreventive agent for humans. In this Phase I clinical trial, BBI concentrate was administered as a single oral dose to 24 subjects with oral leukoplakia. Pharmacokinetics of BBI was analyzed, and subjects were monitored clinically for toxic effects. Subjects received between 25 and 800 chymotrypsin inhibitor units (CIU) of the compound in a dose escalation trial. BBI was taken up rapidly, and a metabolic product of BBI was excreted in the urine within 24-48 h. No clinical or laboratory evidence of toxicity was observed in the study. Protease activity was also measured in buccal cells to evaluate usefulness as a biomarker. Single-dose BBI concentrate administered up to 800 CIU was well tolerated and appeared to be nontoxic. Further investigation in Phase II clinical trials is being done.