Ilker Ali Ozkan

Prognosis of prostate cancer by artificial neural networks

In this study, an artificial neural network has been devised that yields a prognostic result indicating whether patients have cancer or not using their free prostate-specific antigen, total prostate-specific antigen and age data. Though this system does not diagnose cancer conclusively, it helps the doctor in deciding whether a biopsy is necessary by providing information about whether the patient has prostate cancer or not. Data from 121 patients who were definitively diagnosed with cancer after biopsy were used in devising the system. The results of the definitive diagnoses of the patients and the results of the ANN that was performed were analysed using confusion matrix and ROC analyses. As a result of ANN, which was implemented on the basis of these analyses, success rates of 94.11% and 94.44% were achieved for prognosis of disease and validity, respectively. The ANN, which yielded these high rates of reliability, will help doctors make quick and reliable diagnoses without any risks and make it a better option to monitor patients with low prostate cancer risk on whom biopsies must not be carried out through a policy of wait and see.

Determination of the drug dose by fuzzy expert system in treatment of chronic intestine inflammation

In this study, chronic intestine illness symptoms such as sedimentation and prostate specific antigen are used for the design of fuzzy expert system to determine the drug (salazopyrine) dose. Suitable drug dose for patients is obtained by using data of ten patients. The results of some patients are compared with the doses recommended to them by the physician. As a result, it has been seen that proposed system is helped to shorten the treatment duration and minimize or remove the negative effects of determination of drug dose for helping physicians.

A comparative study of ANN and FES for predicting of cutting forces and tool tip temperature in turning

In this study, fuzzy expert system (FES) and artificial neural network (ANN) models are designed for the estimation of cutting forces in turning operations. On designed models, cutting forces and experimental temperature data obtained from different cutting conditions were used in process of turning. Cutting forces at different cutting conditions and temperature values can be estimated with the help of developed models. The results obtained with these models, compared with the experimental data. The regression values were found as 0.99505 between the Experiment-FES and, 0.9888 between Experiment-ANN in the analysis. As a result, the both artificial intelligence (AI) methods have made successful modeling, but its seen that, realized FES model has more successful results than the ANN model in the process of estimation of cutting forces.

The effects of fuzzy control of magnetic flux on magnetic filter performance and energy consumption

Magnetic filters are used effectively in many industrial areas to clean up technological liquids and gases from micron and submicron size magnetic particles. Performance of the magnetic filter is affected by technological parameters like flow rate of the industrial liquid and concentration and flux magnitude of the magnetic filter. These parameters exhibit differences depending on the field of work. Controlling of magnetic filters without regard to these parameters has disadvantages such as low filter performance, ineffectiveness in parameter changes and high energy consumption. To remove these disadvantages, an adaptive fuzzy control system which considers these technological parameters was designed and realized. When the realized filter is compared to the filter that ignores technological parameters, it is observed that energy can be saved at a rate of 68% annually.

Diagnosis of urinary tract infection based on artificial intelligence methods

BACKGROUND AND OBJECTIVE Urinary tract infection (UTI) is a common disease affecting the vast majority of people. UTI involves a simple infection caused by urinary tract inflammation as well as a complicated infection that may be caused by an inflammation of other urinary tract organs. Since all of these infections have similar symptoms, it is difficult to identify the cause of primary infection. Therefore, it is not easy to diagnose a UTI with routine examination procedures. Invasive methods that require surgery could be necessary. This study aims to develop an artificial intelligence model to support the diagnosis of UTI with complex symptoms. METHODS Firstly, routine examination data and definitive diagnosis results for 59 UTI patients gathered and composed as a UTI dataset. Three classification models namely; decision tree (DT), support vector machine (SVM), random forest (RF) and artificial neural network (ANN), which are widely used in medical diagnosis systems, were created to model the definitive diagnosis results using the composed UTI dataset. Accuracy, specificity and sensitivity statistical measurements were used to determine the performance of created models. RESULTS DT, SVM, RF and ANN models have 93.22%, 96.61%, 96.61%, 98.30% accuracy, 95.55%, 97.77%, 95.55%, 97.77% sensitivity and 85.71%, 92.85%, 100%, 100% specificy results, respectively. CONCLUSIONS ANN has the highest accuracy result of 98.3% for UTI diagnosis within the proposed models. Although several symptoms, laboratory findings, and ultrasound results are needed for clinical UTI diagnosis, this ANN model only needs pollacuria, suprapubic pain symptoms and erythrocyte to get the same diagnosis with such accuracy. This proposed model is a successful medical decision support system for UTI with complex symptoms. Usage of this artificial intelligence method has its advantages of lower diagnosis cost, lower diagnosis time and there is no need for invasive methods.