 The state of health SOH of lithium ion batteries, LIBs, is critical to their reliability and efficiency. Two new methods have recently been developed to estimate SOH using electrical impedance spectroscopy, EAS, one based on equivalent circuit models, ECMs, and another based on convolution neural networks, CNNs. Both methods were tested against a traditional Gaussian process regression, GPR, model, and found to be superior in terms of accuracy and precision. Additionally, a particle swarm optimization, PSO, based CNN by LSTM model was developed to further improve the performance of the CNN by LSTM model. This model was compared to the GPR model and found to be 27% and 35% more accurate than the GPR model. This article was authored by Dijili, Dongfang Yang, Li Wei Li, and others.