Optimal IoT Based Improved Deep Learning Model for Medical Image Classification

Recently medical image classification plays a vital role in retrieval and computer-aided diagnosis system. Despite deep learning has proved to be superior previous approaches that depend on handcrafted features; it remains difficult implement because of the high intra-class variance inter-class similarity generated by wide range imaging modalities clinical diseases. The Internet Things (IoT) healthcare systems is quickly becoming viable alternative for delivering high-quality treatment today’s e-healthcare systems. In recent years, been identified as one most interesting research subjects field health care, notably processing. For picture analysis, researchers used combination machine techniques well artificial intelligence. These newly discovered are employed determine diseases, which may aid specialists disease at an earlier stage, giving precise, reliable, efficient, timely results, lowering death rates. Based this insight, novel optimal IoT-based improved model named optimization-driven belief neural network (ODBNN) proposed article. context, primarily quality enhancement procedures like noise removal contrast normalization employed. Then pre-processed subjected feature extraction intensity histogram, average pixel RGB channels, first-order statistics, Grey Level Co-Occurrence Matrix, Discrete Wavelet Transform, Local Binary Pattern measures extracted. After extracting these sets features, May Fly optimization technique adopted select relevant features. selected features fed into algorithm terms classifying similar input images classes. evaluated accuracy, precision, recall, f-measure. investigation evident performance incorporating better than conventional techniques.