4.5 Simulation Environment and Relevant Software Packages

Throughout this work, we used MATLAB (version R2019b) to simulate the in silico physiological model of rat cardiac system. The LSTM model was developed in Python (version 3.7) using TensorFlow2. The controller was synthesized in Python (version 3.7) and the formulated optimization problems were solved using the sequential least squares programming (SLSQP) algorithm in Scipy (version 1.6). The closed-loop simulations were performed through the interaction of Python with MATLAB. For the healthy cases, the optimization solver’s tolerance was set to 10e − 5 and the maximum number of iterations was set to 50. While, for the intra-patient and diseased cases, the solver’s tolerance was set to 10e − 4 and the maximum number of iterations was set to 500 (although the algorithm converged in less than 50 iterations in our simulations).