Cancer Screening Using Biomimetic Pattern Recognition with Hyper-Dimensional Structures

Abstract

Cancer treatments have been shown to be more effective if the cancer is detected and treated at an early stage. Current detection methods include imaging a tissue and blood sample testing. These methods are expensive and invasive for patients, thus scientists have been driven to develop new alternatives to detect cancer. Biomimetic Pattern Recognition (BPR) is a technique that constructs a hyper-dimensional (HD) geometric body by mimicking a biological system and uses it for classification. BPR is derived from the Principle of Homology-Continuity, which assumes elements of the same class are biologically evolved and continuously connected. In other words, between any two elements of the same class, there is a gradual connection. These connecting branches form HD line segments or hyper-surfaces. The resulting topological structure, known as a biomimetic structure, mimics a biological class. In recent years, BPR has been successfully used in voice, facial, and iris recognition software. Here, we developed new BPR algorithms and classification schemes to detect specific cancers using DNA microarray data. We investigated the performance of the proposed BPR methods based on bladder, colon, leukemia, liver, and prostate cancers. Results indicate that the proposed BPR has an increase in recognition rate when compared to previous techniques. BPR has shown to be a promising approach for cancer detection using DNA microarray data.