Hybrid Flower Pollination Algorithm and Binary Particle Swarm Optimization Algorithm for Osteosarcoma Detection

: Generally, Osteosarcoma is represented as malignant bone sarcoma which is typified by an extensive genomic disruption and a proclivity for metastatic extend. Due to the early recognition, Osteosarcoma raises the human beings' survival rate. During the early phase, several Osteosarcoma recognition approaches are developed in order to recognize the Osteosarcoma, however, evaluating the slides in the microscope to identify the tumor necrosis degree and tumor outcome is an important challenge in the medical segment. Therefore, an effectual recognition approach is modeled by exploiting the adopted Hybrid Flower Pollination algorithm with the Binary Particle Swarm Optimization algorithm based Generative Adversarial Network (FPO-BPSO based GAN) to detect the osteosarcoma during the initial phase. Moreover, the adopted FPO-BPSO is modeled using the combination of FPA and BPSO, correspondingly. As a result, the classification of the important tumor, non-tumor, as well as necrotic tumor is performed using GAN by exploiting the histology image slides. GAN is exploited to carry out the osteosarcoma recognition based on features extracted from the image via the cell segmentation process. The GAN training process is performed by exploiting the adopted Hybrid FPO-BPSO approach. Nevertheless, the adopted FPO-BPSO attained superior performance by exploiting the measures, like accuracy, sensitivity, and specificity.