Hybrid Spinal-Fuzzy-DKN approach for breast cancer detection using magnetic resonance images

With the growing modern globe, breast cancer (BC) has become the foremost kind of cancer in women. The recognition of BC in the beginning stage is more significant; hence, the patients take the required treatment for extending their survival rate. Therefore, this work designs the hybrid SpinalNet-Fuzzy-Deep Kronecker Network (Spinal-Fuzzy-DKN) for BC detection.

Magnetic resonance imaging (MRI) plays a crucial role in BC identification. In the preliminary stage of this process, the MRI is applied for image preprocessing. With the aid of a Gaussian filter, the noise level is diminished. The cancer area segmentation is significant for isolating the tumor. The Bayesian Fuzzy Clustering (BFC) model effectively segments the cancer part. Furthermore, the dimensions of the image are enhanced in the image augmenting stage. For extracting the effectual features, feature extraction is employed, in which the texture and the statistical features are extracted. In BC detection, the Spinal-Fuzzy-DKN is utilized.

The accuracy, specificity, and sensitivity metrics are used to validate the Spinal-Fuzzy-DKN and yielded the optimal outcomes of 0.907, 0.914, and 0.924, respectively.

The proposed method is effective for the early detection of BC, which enhances patient survival rates.