Anemia treatment

This research introduces novel methods for optimizing erythropoietin (EPO) dosing in chronic kidney disease patients with anemia. The methods incorporate hemoglobin response model uncertainty and applies physics-informed neural networks (PINN) to build a precise hemoglobin response model under EPO treatment, outperforming traditional modeling approaches. The work enhances EPO dosing precision for anemia management in chronic kidney disease patients.

Relevant publications:

Z. Zhang, Z. Li Hemoglobin Response Modelling under Erythropoietin Treatment: Physiological Model-Informed Machine Learning Method. The Canadian Journal of Chemical Engineering. 2023, in press.

J. McAllister, Z. Li, J. Liu, U. Simonsmeier. Erythropoietin Dose Optimization for Anemia in Chronic Kidney Disease using Recursive Zone Model Predictive Control. IEEE Transactions on Control Systems Technology. 2019, 27, 1181-1193.

J. McAllister, Z. Li, J. Liu, U. Simonsmeier. EPO Dosage Optimization for Anemia Management: Stochastic Control under Uncertainty Using Conditional Value at Risk. Processes. 2018, 6, 60.

J. McAllister, J. Liu, Z. Li, U. Simonsmeier. Erythropoiesis-stimulating-agent Dose Optimization for Anemia Management in Chronic Kidney Disease using Recursive Constrained Modeling and Zone Model Predictive Control. 2018 American Control Conference, Milwaukee, 2018.

J. Ren, J. McAllister, Z. Li, J. Liu, U. Simonsmeier. Modeling of Hemoglobin Response to Erythropoietin Therapy through Constrained Optimization. 6th International Symposium on Advanced Control of Industrial Processes, Taipei, 2017.