DefinePK

Abstract

Early cancer diagnosis using inexpensive and fast techniques could save many lives. Late-stage cancer treatment is difficult. Invasive or non-invasive brain cancer diagnosis is possible. ANN architecture must be carefully chosen for the application, as biopsy is intrusive. This research proposes a functional ANN model to improve diagnosis procedures. ANN research is hot in radiology, cardiology, and oncology. This study used ANNs in medicine. The study also attempted brain tumor diagnosis application. The MR picture requires the neural network to assess brain health. In the hands of human clinicians, this process has a high computational complexity. Experienced radiologists are necessary for accurate classification and segmentation of the brain tumour. We suggest a computer-aided diagnosis-based strategy for segmenting brain tumours as a means of getting around these problems. Below are the steps that make up the proposed artificial neural network and random forest algorithm, support vector machine method. An asymmetrical gradient filter is applied during preprocessing to reduce background noise and even out intensity variations. The artificial neural network is then used to determine if the input brain image is normal or pathological. The result is a model in the vector space. A greater success rate of 2.31% is achieved using the proposed strategy. When compared to other techniques, this one has a sensitivity of 5.81% and a specificity of 2.28%, respectively.