Fast algorithms for model based therapeutics
Degree GrantorUniversity of Canterbury
Degree NameBachelor of Science with Honours
In critical care tight control of glucose has been shown to lead to better clinical outcomes and reductions in length of stay. The need arises to develop new protocols for maintaining tight control of patients' glucose. The current hindrance to developing new protocols is the large amount of computational time and resources that are needed. This paper is based on finding fast analytical based methods for the insulin-glucose model. Exploiting the structure and partial solutions in a subset of the model was the key in finding accurate solutions to the full model. This successfully reduced the computing time compared to the numerical methods used currently by a factor of 6000 whilst introducing minimal error. This will allow new protocols to be rapidly developed leading to better clinical outcomes for patients in critical care.
SubjectsField of Research::08 - Information and Computing Sciences::0802 - Computation Theory and Mathematics
- Engineering: Reports