A novel method for mass spectrometry data representation and analysis

Abstract

In this paper a novel representation/analysis method on high throughput SELDI-TOF mass-spectroscopy data is developed. To avoid complexity of conventional methods, mass spectrum is converted to an intensity image and then image processing techniques is implemented to solve the cancer classification problem. The proposed system benefits a thoroughly novel and efficient idea to design an image-based pattern recognition system for cancer classification. The system is successfully validated using a well-known ovarian cancer proteomic dataset. Results of applying the method are comparable to state of the art methods in proteomic pattern recognition.