Optimization and Deep Learning–Based Content Retrieval, Indexing, and Metric Learning Approach for Medical Images
Kumar K. Suresh, S. Sundaresan, R. Nishanth, Kumar T. Ananth · 2021
Clinical imaging has extended multifaceted in recent days and has got progressively increasing instructive about the patient's life frameworks. Medical imaging data is worked on, and the way it is used can enhance the understanding of various diseases. Content-based image retrieval (CBIR) serves as one of the most promising systems for image recovery. This model is based on an advanced deep learning model incorporating the learned semantic features from a collection of labelled patient images. The algorithm is in the unsupervised mode, which matches words in a visual sequence. The spatial differences are computed by using an improved Accept Difference Index (ADI). For enhancing the texture function, an advanced self-organizing map and metric-based learning are used on the surface texture to recognize the similarity between various brain MRI features. The swarm optimization algorithm is applied to the medical data, which can be enhanced with pattern mining techniques. To pick out the frame's desired documents, the key has been detuned with a cosine model. The proposed method shows substantial spatial similarity for medical images and provides large datasets with accuracy and efficiency for retrieval and indexing.