Addressing the potential of artificial intelligence and biosensors
The UK’s 20 year AMR vision for AMR, “Contained and Controlled”, states the ambition to ensure that “ all decisions to use antimicrobials are informed by a diagnostic test, clinical decision support tool or relevant data and where antimicrobial treatment is indicated”. Our Unit has a number of ongoing externally funded projects which are looking to develop artificial intelligence tools and biosensors to monitor antibiotic drug levels non-invasively, to support prescribing decisions and ultimately optimise and personalise antimicrobial dosing through the addition of pharmacodynamic and pharmacokinetic models.
Our work to date in this area has developed a number of prototype tools. Our supervised machine learning algorithm has been, in a prospective observational trial of 104 patients, to be able to diagnose infection in individuals using only routinely available blood parameters (C-reactive protein, white cell count, bilirubin, creatinine, ALT and alkaline phosphatase) with 84% accuracy. Our case-based reasoning system has been shown to be capable of returning appropriate prescribing recommendations which matched those of expert clinicians approximately 90% of the time. In addition, we are also developing a subcutaneous sensor for monitoring of drug levels, a closed-loop control system for precision antimicrobial delivery and in-silico platform for evaluating vancomycin dosing strategies.
Rawson, Timothy; Hernandez, Bernard; Moore, Luke; et al Supervised machine learning for the prediction of infection on admission to hospital: a prospective observational cohort study Journal of Antimicrobial Chemotherapy, December 2018
Rawson TM, Charani E, Moore LSP, et al. Exploring the Use of C-Reactive Protein to Estimate the Pharmacodynamics of Vancomycin. Therapeutic Drug Monitoring. 2018;40(3):315-21.
Hernandez B, Herrero P, Rawson TM, Moore LSP, et al. Supervised learning for infection risk inference using pathology data. BMC Medical Informatics and Decision Making 2017;17(1):168.
Herrero P, Rawson TM, Philip A, et al, Closed-loop Control for Precision Antimicrobial Delivery: an In Silico Proof-of-Concept. IEEE Transactions on Biomedical Engineering, vol. PP, no. 99, pp. 1-1
Rawson TM, O'Hare D, Herrero P, et al. Delivering precision antimicrobial therapy through closed-loop control systems. Journal of Antimicrobial Chemotherapy, 2017.
Rawson TM, Sharma S, Georgiou P, et al. Towards a minimally invasive device for beta-lactam monitoring in humans. Electrochemistry Communications, Volume 82, 2017, Pages 1-5, ISSN 1388-2481