Automatic detection of tumor zones in magmetic resonance images

In this study, the structure, hardware components and factors affecting image quality of the magnetic resonance imaging device, which is one of the commonly used medical imaging methods, have been examined in detail. The operating principle and application areas of the device are explained and the sub-systems that make up the hardware structure are shown. Image processing was done using a computer-based program, tumor candidate region was determined over the processed images and attribute values were found. Later on, Artificial Neural Network was developed and automatic detection of brain tumor regions was performed in magnetic resonance images from the attribute values. Ferromagnetic magnets used in MRI devices create strong magnetic fields. These external electromagnetic fields have a detrimental effect on the body's electrical implants and biomaterials. In addition to the magnetic field, the gradient magnetic field and electromagnetic pulses interact with the biomaterials within their domains. As a result, patients entering the MRI may experience problems such as tissue warming or the failure of embedded medical devices. In order to avoid these unwanted interferences, MRI devices must be designed and manufactured in accordance with these requirements by performing electromagnetic compatibility and signal interference tests. In addition to the hardware features of MRI devices; by selecting various parameters such as field of view, matrix value, NEX and contrast to noise ratio, images with high signal to noise ratio are obtained. The high image quality offers the possibility of examination for accurate diagnosis of diseases. Making these choices correctly and creating the most suitable MRI device structure is an important engineering application field.