Today’s accelerated world has caused significant changes in our lifestyle. The permanent stress has changed the habits of everyday people fighting against it. The daily “drugs”, such as alcohol or cigarette increased the frequency of diseases attacking the vocal organs. That is why it is important to recognize and to diagnose diseases in time before they can have a dramatic ending. Therefore, I decided to design and develop a medical decision support system, which simply uses the voice of the patients and it is able to diagnose the cause of the pathological voice disorders and determine the stage of the diseases.
In the recent decades a lot of research focused on the computer based human speech processing and speech recognition with varying success. There are two core elements of such systems. The first one is the feature vector of the given voice containing its most relevant parameters. The other one is an artificial intelligence (AI) system, which is able to distinguish between different cases and to make decisions in specific situations.
During my research my first aim was to find parameters, which can be used to determine the disease using only the patient’s voice and which is good enough for teaching an AI system. The system of speaking and hearing is too complex for using only the traditional linear signal processing parameters, so I turned to use non-linear signal processing parameters. I approached the complexity of the speech producing, as a chaotic system and I used one of the most characteristic parameters of chaos, the fractal dimension (Df). The measurements proved, that depending on the origin of the disease the Df of the voice varied in a scalable interval and showed significant differences between the diseases. I also managed to point out, that it is necessary to examine more than one voice sample from a patient in order to perform an accurate and successful diagnosis.
Besides these I was looking for a mathematic model, which is good enough to handle such a chaotic and dynamic system easily and quickly and to make decisions independently. These conditions led me to the soft computing methods. Mainly the strong opportunities in fuzzy inference systems and evolutionary algorithms drove me. I described the phases of the design and development of a medical decision support system based on fuzzy logic. The system also uses the Df as an input parameter and focuses on determining the origin of the disease.