Signal and Image Representations Based Hybrid Intelligent Diagnosis Approach for a Biomedicine Application

Hybrid Intelligent Diagnosis Systems

“…This task is not trivial, because the time (or frequency) is not always the variable that points up the studied phenomena's features leading then to a necessity of multiple knowledge representations (signal, image, …). Thus, an interesting combinations: -combination of the classification of the signal knowledge representation and the one of the global image knowledge representation, in this combination several variants can be exploited, e.g., using the same classifier (e.g., MultiLayer feedforward Perceptron (MLP) network classifier or Radial Basis Function (RBF) network classifier) for the two classifications ; or using different classifiers (e.g., MLP and RBF classifiers) see [12], … -combination of the classification of the global image knowledge representation and the one of the subdivided image knowledge representation, also in this combination several variants can be exploited, e.g., using the same classifier (MLP classifier or RBF classifier) for the two classifications see [10] ; or using different classifiers (MLP and RBF classifiers), … -combination of the classification of the global image using two different classifiers (MLP and RBF) see [13], -combination of the classification of the subdivided image using two different classifiers (MLP and RBF) see [11], [13], -… In fact, in such combination of different classifiers the following properties are exploited: MLP are neural global approximators, whereas RBF network are neural local approximators [16]. The idea is to classify global images using global approximators and to classify signals and subdivided images using local approximators.…”

“…Given the observed symptoms, the diagnosis system has the tasks of ascribing symptoms to abnormal or normal functioning. Based on this basis principle, several biomedicine diagnosis applications have been developed [6], [7], [8], [9], [10], [11], [12], [13]. For instance, the biomedicine diagnosis application illustrated in Fig.…”

“…Currently, one of the most used approaches to feature identification, classification, and decision-making problems inherent to fault detection and diagnosis, is multiple model approaches with soft computing implying mainly neural networks and fuzzy logic [1], [3], [4], [5], [6], [9], [10], [11], [12], [13], [14], [15].…”

“…Recently, several decision support systems and intelligent systems have been developed [1], [2], [3] and the diagnosis approaches based on such intelligent systems have been developed for industrial applications [1], [4], [5], and biomedicine applications [6], [7], [8], [9], [10], [11], [12], [13]. Currently, one of the most used approaches to feature identification, classification, and decision-making problems inherent to fault detection and diagnosis, is multiple model approaches with soft computing implying mainly neural networks and fuzzy logic [1], [3], [4], [5], [6], [9], [10], [11], [12], [13], [14], [15].…”