Application of Modeling and Optimization Methods in Biomedicine and Biorefineries

Application of modeling and optimization methods in biomedicine and biorefineries

Kirti M. Yenkie, University of Wisconsin-Madison Current Position: Second year Postdoctoral Researcher Research Interests: My research focuses on development of methodologies for predicting policies or forecasting decisions which can enhance the performance of existing processes as well as act as guidelines for developing novel processes in chemical and biological systems. I plan to incorporate methods such as mathematical modeling, simulation and numerical techniques for the process analysis and then apply principles from optimization and control for prediction of better operating policies and decisions. Policy prediction is particularly advantageous for processes in pharmaceutical industries where the product specifications are stringent and it is essential to maintain the product quality and purity. Also, medical treatments can be benefitted by prediction of drug dosing policies for chronic problems such as diabetes or treatments with adverse side effects such as cancer chemotherapy. These methods are also applicable to selection of best alternatives in a process; for example, selection of the best option among the available alternatives for wastewater treatment, based on initial conditions and final water quality requirements. To ensure the robustness of the results predicted, I intend to apply methods for uncertainty characterization and incorporation into process models through stochastic modeling and control methods.   Postdoctoral Project: Modeling, simulation, analysis and optimization of biorefineries Advisor: Dr. Christos T. Maravelias, Chemical and Biological Engineering, University of Wisconsin-Madison. - Assessment of separation technologies for efficient recovery of bio-based chemicals from microbial cultures - Collaboration with experimentalists from Dr. Brian Pfleger’s group for information about microbial strains, substrates, culture conditions, titers, productivity and yield. -  Developing guidelines for selection of best possible separation scheme based on physical properties of selected chemicals by application of mathematical modeling, mixed-integer non-linear programming optimization and high performance computing methods. PhD Dissertation: Stochastic processes from batch crystallization to in-vitro fertilization (IVF) Advisor: Dr. Urmila M. Diwekar, Bioengineering, University of Illinois at Chicago (UIC). -  Studied effects of uncertainties in batch crystallization, developed stochastic models using Ito processes and applied optimal control methods for predicting robust operating policies for enhanced product quality. -  Analyzed clinical data for superovulation (first stage in IVF) cycles. Developed mathematical model for multiple ovulation using differential equations and applied optimization and control for predicting optimal hormone dosing policies. -  Developed a new approach for parameter estimation in stochastic differential equations and applied it to develop stochastic models for multiple ovulation and follicle growth. Other Research Projects:

(i)     Developed models for cellular systems in terms of stochastic differential equations as an alternative to discrete event simulation methods such as the Gillespie algorithm.

(ii)   Application of optimal control methods to research on sustainability by using representative models for earth’s ecological-economic interactions in terms of differential-algebraic equations and regulatory policies. Future Research Plans: Optimizing cancer chemotherapy - I plan to work on modeling the cancer propagation by using deterministic and stochastic models. Using these models, treatment therapies in terms of drug dosing regimens as well the time-gaps between dosing periods will be designed. I would also like to explore the possibility of treatment by combination of multiple drugs or multiple therapies such as chemotherapy, radiotherapy and gene therapy. Optimizing wastewater treatment methods - Processes such as activated sludge, advanced oxidation, ion-exchange and anaerobic digestion are used to treat wastewater for compliance with the environmental norms. I plan to use optimal control and MINLP optimization methods for prediction of treatment conditions as well as selection of best alternatives. Teaching Interests: My B. Tech. and M. Tech. degrees in Chemical Engineering and Ph.D. in Bioengineering have provided me with the opportunity to learn core chemical engineering subjects, like thermodynamics, mass transfer, reaction engineering, transport phenomena, etc. as well as some interdisciplinary topics like metabolic engineering, biochemical engineering and computational biology. Also, my interdisciplinary research facilitated me to develop skills in applied optimization, biomedicine, crystallization, separation processes, sustainability and advanced mathematics. I have been a teaching assistant for several courses such as chemical engineering thermodynamics, plant utility, reaction engineering laboratory, senior process design, and cell and tissue engineering during my graduate studies. I have graded assignments, conducted problem solving sessions and supervised experiments. I have also conducted tutorials for programming tools like MATLAB and GAMS. I have been a teaching associate for the introduction to optimization course at UIC and have taken lectures on genetic algorithms and simulated annealing. I have mentored undergraduate and graduate students for their research projects. I believe that I am capable of teaching most of the chemical engineering subjects, particularly the ones requiring computer programming and numerical methods such as transport phenomena, process design and optimization, bioprocess and biomedical engineering. I am also interested in developing and teaching innovative courses like modeling and control in biomedicine and stochastic processes in chemical engineering. Selected Publications

1)     Yenkie, K. M.; Diwekar, U. 2013. Stochastic optimal control of seeded batch crystallizer applying Ito process. Ind. Eng. Chem. Res., 52:108-122.

2)     Yenkie, K. M.; Diwekar, U.; Bhalerao, V. 2013. Modeling the superovulation stage in in-vitro fertilization. IEEE Trans. Biomed. Eng., 60(11): 3003-3008.

3)     Yenkie, K. M.; Diwekar, U. 2014. Optimal control for predicting customized drug dosage for superovulation stage of in-vitro fertilization. Journal of Theoretical Biology, 355: 219-228.

4)     Yenkie, K. M.; Diwekar, U.; Bhalerao, V. 2014. Modeling and prediction of outcome for the superovulation stage in in-vitro fertilization. JFIV Reprod.Med.Genet. 2(2):1000122(1-8).

5)     Yenkie, K. M.; Diwekar, U. 2014. Comparison of optimal control methods for customized drug dosage prediction in superovulation stage of in-vitro fertilization. Computers and Chemical Engineering, 71: 708-714.

6)     Yenkie, K. M.; Diwekar, U. 2014. Uncertainty in clinical data and stochastic model for in-vitro fertilization. Journal of Theoretical Biology, 367: 76-85.

7)     Yenkie, K. M.; Diwekar, U.; Linninger, A. 2016. Simulation-free estimation of reaction propensities in cellular reactions and gene signaling networks. Computers and Chemical Engineering, 87: 154-163.