A Reliable brain tumor detection and classiication frame work in soft computing

Abstract

Generally, cancer is a disease, in which brain tumor is one of the newlinesevere syndromes that hold large mass and variant in size. It may be either newlinebenign or malignant. The variation from being normal and cancerous is newlineidentified by some real-time equipment. In the medical field, the past decade newlinefocusing more on developing and improving the tumor detection modules to newlineprovide an excellent diagnosis. An image processing framework in addition to newlinesoft computing plays a major role in detecting various images, collective newlineclinical database and so on. It allows differentiating the cancerous and newlinenoncancerous part from the given sample. However, one of the problems that newlineare faced by image processing equipment is backward compatibility and less newlineaccuracy. newlineTo overcome such limitations, the proposed research work focuses newlineon implementing soft computing approximations techniques to solve complex newlinecomputational problems. In such functions, the segmentation and classification newlinemechanisms are improved to deliver proper diagnosing with the rapid process. newlineHence, the processing time is greatly reduced and accuracy is improved greatly. newlineHence in the initial phase of this research work, brain tumors regions newlineare extracted by using image processing and soft computing approaches. The newlineSoft Computing based Brain Tumor Detection and Classification Technique newline(SC-BTDC) is introduced for providing efficient classification. It collects the newlineMRI based brain image database and processed using the median filter with edge newlinedetection. Here, the noise factor is eliminated. newline