Search

Diagnostics 13(12), 2098 (2023). doi:10.3390/diagnostics13122098 special issue: "AI-Driven Intelligent Health Care Diagnostic Solutions: A Machine Learning Approach", Acute Respiratory Distress Syndrome (ARDS) is a condition that endangers the lives of many Intensive Care Unit patients through gradual reduction of lung function. Due to its heterogeneity, this condition has been difficult to diagnose and treat, although it has been the subject of continuous research, leading to the development of several tools for modeling disease progression on the one hand, and guidelines for diagnosis on the other, mainly the “Berlin Definition”. This paper describes the development of a deep learning-based surrogate model of one such tool for modeling ARDS onset in a virtual patient: the Nottingham Physiology Simulator. The model-development process takes advantage of current machine learning and data-analysis techniques, as well as efficient hyperparameter-tuning methods, within a high-performance computing-enabled data science platform. The lightweight models developed through this process present comparable accuracy to the original simulator (per-parameter R2 > 0.90). The experimental process described herein serves as a proof of concept for the rapid development and dissemination of specialised diagnosis support systems based on pre-existing generalised mechanistic models, making use of supercomputing infrastructure for the development and testing processes and supported by open-source software for streamlined implementation in clinical routines., Published by MDPI, Basel

Additional file 1. Detailed model description, additional figures and tables.

Additional file 1. Full description of computational simulator, additional results and figures.

Configuring the simulator to patient data using global optimization and the tabulated configuration results. The file presents the optimization strategy used in matching the model to the ARDS patient data and the resultant parameter values for the different models tabulated for each patient.

Additional figures and experiments. The file reports the effects on PaO2/FIO2 of changes in hemoglobin levels, cardiac output, and FIO2 for the MRS-10 RM applied to patient A and presents some further model validation results.

Bacteria can be harnessed to synthesise high-value chemicals. A promising strategy for increasing productivity uses inducible control systems to switch metabolism from growth to chemical synthesis once a large population of cell factories are generated. However, use of expensive chemical inducers limits scalability of this approach for biotechnological applications. Switching using cheap nutrients is an appealing alternative, but their tightly regulated uptake and consumption again limits scalability. Here, using mathematical models of fatty acid uptake in E. coli as an exemplary case study, we unravel how the cell’s native regulation and program of induction can be engineered to minimise inducer usage. We show that integrating positive feedback loops into the circuitry creates an irreversible metabolic switch, which, requiring only temporary induction, drastically reduces inducer usage. Our proposed switch should be widely applicable, irrespective of the product of interest, and brings closer the realization of scalable and sustainable microbial chemical production., A promising strategy to increase product synthesis from bacteria uses inducible systems to switch metabolism to production. Here, the authors use models to show how engineering positive feedback loops into the genetic circuitry creates a switch that requires only temporary induction with a cheap nutrient to switch metabolism irreversibly, and so drastically reduce inducer use and cost.