Virtual patient with temporal evolution for mechanical ventilation trial studies: A stochastic model approach.

BACKGROUND AND OBJECTIVE: Healthcare datasets are plagued by issues of data scarcity and class imbalance. Clinically validated virtual patient (VP) models can provide accurate in-silico representations of real patients and thus a means for synthetic data generation in hospital critical care settings. This research presents a realistic, time-varying mechanically ventilated respiratory failure VP profile synthesised using a stochastic model. METHODS: A stochastic model was developed using respiratory elastance (Ers) data from two clinical cohorts and averaged over 30-minute time intervals. The stochastic model was used to generate future Ers data based on current Ers values with added normally distributed random noise. Self-validation of the VPs was performed via Monte Carlo simulation and retrospective Ers profile fitting. A stochastic VP cohort of temporal Ers evolution was synthesised and then compared to an independent retrospective patient cohort data in a virtual trial across several measured patient responses, where similarity of profiles validates the realism of stochastic model generated VP profiles. RESULTS: A total of 120,000 3-hour VPs for pressure control (PC) and volume control (VC) ventilation modes are generated using stochastic simulation. Optimisation of the stochastic simulation process yields an ideal noise percentage of 5-10% and simulation iteration of 200,000 iterations, allowing the simulation of a realistic and diverse set of Ers profiles. Results of self-validation show the retrospective Ers profiles were able to be recreated accurately with a mean squared error of only 0.099 [0.009-0.790]% for the PC cohort and 0.051 [0.030-0.126]% for the VC cohort. A virtual trial demonstrates the ability of the stochastic VP cohort to capture Ers trends within and beyond the retrospective patient cohort providing cohort-level validation. CONCLUSION: VPs capable of temporal evolution demonstrate feasibility for use in designing, developing, and optimising bedside MV guidance protocols through in-silico simulation and validation. Overall, the temporal VPs developed using stochastic simulation alleviate the need for lengthy, resource intensive, high cost clinical trials, while facilitating statistically robust virtual trials, ultimately leading to improved patient care and outcomes in mechanical ventilation.

Compound W-Net with Fully Accumulative Residual Connections for Liver Segmentation Using CT Images.

Computed tomography (CT) is a common modality for liver diagnosis, treatment, and follow-up process. Providing accurate liver segmentation using CT images is a crucial step towards those tasks. In this paper, we propose a stacked 2-U-Nets model with three different types of skip connections. The proposed connections work to recover the loss of high-level features on the convolutional path of the first U-Net due to the pooling and the loss of low-level features during the upsampling path of the first U-Net. The skip connections concatenate all the features that are generated at the same level from the previous paths to the inputs of the convolutional layers in both paths of the second U-Net in a densely connected manner. We implement two versions of the model with different number of filters at each level of each U-Net by maximising the Dice similarity between the predicted liver region and that of the ground truth. The proposed models were trained with 3Dircadb public dataset that were released for Sliver and 3D liver and tumour segmentation challenges during MICCAI 2007-2008 challenge. The experimental results show that the proposed model outperformed the original U-Net and 2-U-Nets variants, and is comparable to the state-of-the-art mU-Net, DC U-Net, and Cascaded UNET.

SPTV sheds light on flow dynamics of fractal-induced turbulence over a plate-fin array forced convection.

A 3D stationary particle tracking velocimetry (SPTV) with a unique recursive corrective algorithm has been successfully established to detect the instantaneous regional fluid flow characteristics. The veracity of SPTV is corroborated by conducting actual displacement measurement validation, which gives a maximum percentage deviation of about 0.8%. This supports the accuracy of the current SPTV system in 3D position detection. More importantly, the SPTV detected velocity fluctuations are highly repeatable. In this study, SPTV is proven to be able to express the nature of chaotic fractal grid-induced regional turbulence, namely: the high turbulence intensity attributed to multilength-scale wake interactions, the Kolmogorov's -5/3 law decay, vortex shedding, and the Gaussian flow undulations immediately leeward of the grid followed by non-Gaussian behaviour further downstream. Moreover, by comparing the flow fields between control no-grid and fractal grid-generated turbulence of a plate-fin array, SPTV reveals vigorous turbulence intensity, smaller regional integral-length-scale, and energetic vortex shedding at higher frequency for the latter, particularly between fins. Thereupon, it allows the unravelling of detailed thermofluid interplays of plate-fin heat sink heat transfer augmentation. The novelty of SPTV lies in its simplicity, use of low-cost off-the-shelf components, and most remarkably, low computational complexity in detecting fundamental characteristics of turbulent fluid flow.

Fractal grid-induced turbulence strength characterization via piezoelectric thin-film flapping velocimetry.

The centerline streamwise and cross-sectional (x/Dh = 0.425) turbulence characteristics of a 2D planar space-filling square-fractal-grid (SFG) composed of self-similar patterns superimposed at multiple length-scales is experimentally unveiled via piezoelectric thin-film flapping velocimetry (PTFV). The fluid-structure-interaction between a flexible piezoelectric thin-film and SFG-generated turbulent flow at ReDh = 4.1 × 104 is investigated by analysis of the thin-film's mechanical response. Measurements of the thin-film-tip deflection δ and induced voltage V demonstrate increasing flow fluctuation strength in the turbulence generation region, followed by rapid decay further downstream of the SFG. Interestingly, SFG-induced turbulence enables the generation of maximum centerline thin-film's response (Vrms, δrms) and millinewton turbulence-forcing (turbulence-induced excitation force acting on the thin-film) Frms which are respectively, 7× and 2× larger than the classical square-regular-grid of similar blockage ratio. The low frequency, large-scale energy-containing eddies at SFG's central opening plays a critical role in driving the thin-film vibration. Most importantly, the SFG-generated turbulence at (y/T = 0.106, z/T = 0.125) away from the centerline allows equivalent mechanical characteristics of turbulence generation and decay, with peak of 1.9× nearer from grid. In short, PTFV provides a unique expression of the SFG-generated turbulence, of which, the equivalent turbulence length-scale and induced-forcing deduced could aid in deciphering the flow dynamics for effective turbulence management.

Pneumocephalus and air travel: an experimental investigation on the effects of aircraft cabin pressure on intracranial pressure.

This study investigates the effects of aircraft cabin pressure on intracranial pressure (ICP) elevation of a pneumocephalus patient. We propose an experimental setup that simulates the intracranial hydrodynamics of a pneumocephalus patient during flight. It consists of an acrylic box (skull), air-filled balloon [intracranial air (ICA)], water-filled balloon (cerebrospinal fluid and blood) and agarose gel (brain). The cabin was replicated using a custom-made pressure chamber. The setup can measure the rise in ICP during depressurization to levels similar to that inside the cabin at cruising altitude. ΔICP, i.e. the difference between mean cruising ICP and initial ICP, was found to increase with ICA volume and ROC. However, ΔICP was independent of the initial ICP. The largest ΔICP was 5 mmHg; obtained when ICA volume and ROC were 20 ml and 1,600 ft/min, respectively. The postulated ICA expansion and the subsequent increase in ICP in pneumocephalus patients during flight were successfully quantified in a laboratory setting. Based on the quantitative and qualitative analyses of the results, an ICA volume of 20 ml and initial ICP of 15 mmHg were recommended as conservative thresholds that are required for safe air travel among pneumocephalus patients. This study provides laboratory data that may be used by doctors to advise post-neurosurgical patients if they can safely fly.

Impact of myocardial infarction on intraventricular vortex and flow energetics assessed using computational simulations.

Flow energetics have been proposed as early indicators of progressive left ventricular (LV) functional impairment in patients with myocardial infarction (MI), but its correlation with individual MI parameters has not been fully explored. Using electro-fluid-structure interaction LV models, this study investigated the correlation between four MI parameters: infarct size, infarct multiplicity, regional enhancement of contractility at the viable myocardium area (RECVM), and LV mechanical dyssynchrony (LVMD) with intraventricular vortex and flow energetics. In LV with small infarcts, our results showed that infarct appearance amplified the energy dissipation index (DI), where substantial viscous energy loss was observed in areas with high flow velocity and near the infarct-vortex interface. The LV with small multiple infarcts and RECVM showed remarkable DI increment during systole and diastole. In correlation analysis, the systolic kinetic energy fluctuation index (E') was positively related to ejection fraction (EF) (R2  = 0.982) but negatively correlated with diastolic E' (R2  = 0.970). Diastolic E' was inversely correlated with vortex kinetic energy (R2  = 0.960) and vortex depth (R2  = 0.876). We showed an excessive systolic DI could differentiate infarcted LV with normal EF from healthy LV. Strong flow acceleration, LVMD, and vortex-infarct interactions were predominant factors that induced excessive DI in infarcted LVs. Instead of causing undesired flow turbulence, high systolic E' suggested the existence of energetic flow acceleration, while high diastolic E' implied an inefficient diastolic filling. Thus, systolic E' is not a suitable early indicator for progressive LV dysfunction in MI patients, while diastolic E' may be a useful index to indicate diastolic impairment in these patients.

A mathematical framework for minimally invasive tumor ablation therapies.

Minimally invasive tumor ablations (MITAs) are an increasingly important tool in the treatment of solid tumors across multiple organs. The problems experienced in modeling different types of MITAs are very similar, but the development of mathematical models is mostly performed in isolation according to modality. Fundamental research into the modeling of specific types of MITAs is indeed required, but to choose the optimal treatment for an individual the primary clinical requirement is to have reliable predictions for a range of MITAs. In this review of the mathematical modeling of MITAs 4 modalities are considered: radiofrequency ablation, microwave ablation, cryoablation, and irreversible electroporation. The similarities in the mathematical modeling of these treatments are highlighted, and the analysis of the models within a general framework is discussed. This will aid in developing a deeper understanding of the sensitivity of MITA models to physiological parameters and the impact of uncertainty on predictions of the ablation zone. Through robust validation and analysis of the models it will be possible to choose the best model for a given application. This is important because many different models exist with no objective comparison of their performance. The collection of relevant in vivo experimental data is also critical to parameterize such models accurately. This approach will be necessary to translate the field into clinical practice.

A boundary element model of the human eye undergoing laser thermokeratoplasty.

In the present paper, a three-dimensional radially symmetric boundary element model of the human eye is proposed for simulating changes in corneal temperature during treatment of laser thermokeratoplasty. Energy absorption inside the cornea is modeled using the Beer-Lambert law. Heat transfer inside the eye is assumed to be governed by the classical heat diffusion equation. The resulting initial-boundary value problem is solved numerically using a time-stepping boundary element method. The temperature field is calculated for heating by both the pulsed laser and the continuous wave laser. The results obtained are compared with those from other models found in the literature.

Simulation of aqueous humor hydrodynamics in human eye heat transfer.

The need to develop accurate representation of the human eye for the purpose of physiological studies is important to ensure that the predicted results are reliable. The presence of natural circulation of aqueous humor (AH) is evident from clinical, experimental and simulated observations. Most of the thermal models of the human eye that are found in the literature, however, had assumed a stagnant AH inside the anterior chamber. In this paper, a two-dimensional model of the human eye is developed where the circulation of AH inside the anterior chamber is included. The effects of the AH flow on the temperature distribution inside the eye are investigated. The natural circulation of AH is found to increase the temperature and distorts the temperature profile in the cornea and anterior chamber. Further investigations, where an artificial heat source is introduced inside the human eye suggest that AH flow plays a role in the heat transfer at the anterior region of the eye although this has yet to be quantified experimentally.