Prediction of Cancer Blood Disorder Using Adaptive Otsu Threshold and Deep Convolutional Neural Networks
S. Jeyalaksshmi, S. K. Piramu Preethika, J. Grace Hannah, S. Sathya, Sangeetha Radhakrishnan, S. Silvia Priscila · Advances in medical technologies and clinical practice book series · 2024
Cancer blood disorder affects blood cell formation and function. Blood disorders may affect platelets, plasma, and white and red blood cells. The goal of this study is to identify blood problems with cancer. In this study, cancer and blood problem images are enhanced and filtered. Remove noise from photos using image filtering. The authors recommended an adaptive anisotropic diffusion filter (2D AADF) for noise reduction. Image enhancement improves clarity. Enhancement uses de-noised photos. The authors propose picture improvement using adaptive mean adjustment (AMA). Real-time data was used for picture preprocessing. The proposed filtering approach is the most effective compared to 2D AADF, 2D adaptive log color filter, and 2D frequency domain filter. The suggested image improvement algorithm performed best compared to contrast limited adaptive histogram equalization, adaptive otsu threshold, image coherence improvement, and 2D adaptive mean adjustment.