The document describes a presentation on robust prediction of cancer disease using pattern classification of microarray gene-expression data. It discusses using robust classifiers with gene-expression data to more accurately classify samples as cancerous or normal, even in the presence of outliers in the data. The presentation outlines include introducing gene-expression data, describing robust classifiers, investigating their performance on simulated data with and without outliers, applying the proposed robust classifier to simulated and real gene-expression cancer data involving two and three classes, and comparing results.