LSTM-based adaptive whale optimization model for classification of fused multimodality medical image

Multimodality medical image fusion is the important area in the medical imaging field which enhances the reliability of medical diagnosis. Medical image fusion as well as their classification is employed to achieve significant multimodality of medical image data. The single modality image does not provide the adequate information needed for an accurate diagnosis. An adaptive whale optimization algorithm (AWOA) with long short-term memory (LSTM) based efficient multimodal medical image fusion classification is proposed to enhance diagnostic accuracy. To obtain the fused images, discrete wavelet transform with an arithmetic optimization algorithm is used for the fusion process by taking the multimodal medical images. In this AWOA algorithm, the classification accuracy is enhanced, and also the weight of the LSTM is optimized. The three dataset images used in evaluating the experimental set with the representation of several diseases like mild Alzheimer’s encephalopathy, hypertensive encephalopathy and glioma to validate the proposed method are demonstrated. The classification accuracy obtained for each respective dataset is 98.25%, 98.54% and 98.75%. The proposed classifier has achieved better accuracy as compared to other classifiers.