Quantum Inspired Marine Predator Algorithm with Long Short-Term Memory for Detection and Classification of Chronic Kidney Disease

Chronic Kidney Disease (CKD) is a continual decline of the kidney working time being and it is a progressive condition. The main problem with CKD is, it is asymptomatic in its early stages, this makes the early detection challenging. In order to detect the symptoms of the disease and to figure out the exact stage of the disease advance technologies have been introduced. The traditional methods failed to diagnosing the disease accurately in early stage. The paper proposed a Quantum Inspired Marine Predator Algorithm (QIMPA) in order to predict and classify the chronic kidney disease. This algorithm is employed with the marine predators using Brownian and Levy flights that uses hunting strategies. This strategy works in three phases, firstly, prey's movements and strategies are noted how it escapes from the predator, then a balanced approach is created between predator and prey. Finally, in this phase, optimization process produces a Levy flight pattern to capture the predator and prey moves at low velocity. Long Short-Term Memory (LSTM) is utilized for the classification, it works by capturing the long-range values of features that enhances the diagnosing kidney disease. The proposed QIMPA model obtained 99.04% for CKD dataset compared to the existing methods.