Classification methodology for mitigating the effects of unreliable class assignments in high dimensional biomedical data

点击翻译按钮获取中文摘要

Classifying biomedical data involves finding a mapping (relationship) from patterns (e.g., data relating to some type of tissue) to a set of classes (e.g., disease states). Patterns are represented by features (e.g., concentrations of biological compounds) and class labels are assigned using a reference test (RT) (e.g., a medical expert's analysis of tissue being "normal" or "abnormal"). This process often suffers from three significant challenges: the RT may be unreliable; the number of patterns is low; and the number of features in a pattern is high. While RTs may be well-established benchmarks, they are seldom perfectly accurate and sometimes improperly applied. Nevertheless, any strategy that compensates for this imprecision must ensure that the mapping is correctly validated against the benchmark. The last two challenges, known collectively as the "curse of dimensionality", cause an inability to find robust, general solutions. This is often addressed by transformations that select the most discriminatory subset of features while not impeding the medical expert's ability to make informed judgments about the mapping's predictive power.