Prognostic significance of pulmonary arterial wedge pressure estimated by deep learning in acute heart failure.

AIMS: Acute decompensated heart failure (ADHF) presents with pulmonary congestion, which is caused by an increased pulmonary arterial wedge pressure (PAWP). PAWP is strongly associated with prognosis, but its quantitative evaluation is often difficult. Our prior work demonstrated that a deep learning approach based on chest radiographs can calculate estimated PAWP (ePAWP) in patients with cardiovascular disease. Therefore, the present study aimed to assess the prognostic value of ePAWP and compare it with other indices of haemodynamic congestion. METHODS AND RESULTS: We conducted a post hoc analysis of a single-centre, prospective, observational heart failure registry and analysed data from 534 patients admitted for ADHF between January 2018 and December 2019. The deep learning approach was used to calculate ePAWP from chest radiographs at admission and discharge. Patients were divided into three groups based on the ePAWP tertiles at discharge, as follows: first tertile group (ePAWP ≤ 11.2 mm Hg, n = 178), second tertile group (11.2 < ePAWP < 13.5 mm Hg, n = 170), and third tertile group (ePAWP ≥ 13.5 mm Hg, n = 186). The third tertile group had a higher prevalence of atrial fibrillation and lower systolic blood pressure at admission; a lower platelet count and higher total bilirubin at both admission and discharge; and a higher left atrial diameter, peak early diastolic transmitral flow velocity, right ventricular end-diastolic diameter, and maximal inferior vena cava diameter at discharge. During the median follow-up period of 289 days, 223 (41.7%) patients reached the primary endpoint (a composite of all-cause mortality or rehospitalization for heart failure). Kaplan-Meier analysis revealed a significantly higher composite event rate in the third tertile group (log-rank test, P = 0.006). Even when adjusted for clinically relevant factors, a higher ePAWP at discharge and a smaller decrease in ePAWP from admission to discharge were significantly associated with higher event rates [ePAWP at discharge: hazard ratio, 1.10; 95% confidence interval (CI), 1.02-1.19; P = 0.010; and size of ePAWP decrease: hazard ratio, 0.94; 95% CI, 0.89-0.99; P = 0.038]. CONCLUSIONS: Our study suggests that ePAWP calculated by a deep learning approach may be useful for identifying and monitoring pulmonary congestion during hospitalization for ADHF.

Quantitative estimation of pulmonary artery wedge pressure from chest radiographs by a regression convolutional neural network.

Recent studies reported that a convolutional neural network (CNN; a deep learning model) can detect elevated pulmonary artery wedge pressure (PAWP) from chest radiographs, the diagnostic images most commonly used for assessing pulmonary congestion in heart failure. However, no method has been published for quantitatively estimating PAWP from such radiographs. We hypothesized that a regression CNN, an alternative type of deep learning, could be a useful tool for quantitatively estimating PAWP in cardiovascular diseases. We retrospectively enrolled 936 patients with cardiovascular diseases who had undergone right heart catheterization (RHC) and chest radiography and estimated PAWP by constructing a regression CNN based on the VGG16 model. We randomly categorized 80% of the data as training data (training group, n = 748) and 20% as test data (test group, n = 188). Moreover, we tuned the learning rate-one of the model parameters-by 5-hold cross-validation of the training group. Correlations between PAWP measured by RHC [ground truth (GT) PAWP] and PAWP derived from the regression CNN (estimated PAWP) were tested. To visualize how the regression CNN assessed the images, we created a regression activation map (RAM), a visualization technique for regression CNN. Estimated PAWP correlated significantly with GT PAWP in both the training (r = 0.76, P < 0.001) and test group (r = 0.62, P < 0.001). Bland-Altman plots found a mean (SEM) difference between GT and estimated PAWP of - 0.23 (0.16) mm Hg in the training and - 0.05 (0.41) mm Hg in the test group. The RAM showed that our regression CNN model estimated high PAWP by focusing on the cardiomegaly and pulmonary congestion. In the test group, the area under the curve (AUC) for detecting elevated PAWP (≥ 18 mm Hg) produced by the regression CNN model was similar to the AUC of an experienced cardiologist (0.86 vs 0.83, respectively; P = 0.24). This proof-of-concept study shows that regression CNN can quantitatively estimate PAWP from standard chest radiographs in cardiovascular diseases.

SIRVVD model-based verification of the effect of first and second doses of COVID-19/SARS-CoV-2 vaccination in Japan.

As of August 2021, COVID-19 is still spreading in Japan. Vaccination, one of the key measures to bring COVID-19 under control, began in February 2021. Previous studies have reported that COVID-19 vaccination reduces the number of infections and mortality rates. However, simulations of spreading infection have suggested that vaccination in Japan is insufficient. Therefore, we developed a susceptible-infected-recovered-vaccination1-vaccination2-death model to verify the effect of the first and second vaccination doses on reducing the number of infected individuals in Japan; this includes an infection simulation. The results confirm that appropriate vaccination measures will sufficiently reduce the number of infected individuals and reduce the mortality rate.

Fast screening framework for infection control scenario identification.

Due to the emergence of the novel coronavirus disease, many recent studies have investigated prediction methods for infectious disease transmission. This paper proposes a framework to quickly screen infection control scenarios and identify the most effective scheme for reducing the number of infected individuals. Analytical methods, as typified by the SIR model, can conduct trial-and-error verification with low computational costs; however, they must be reformulated to introduce additional constraints, and thus are inappropriate for case studies considering detailed constraint parameters. In contrast, multi-agent system (MAS) simulators introduce detailed parameters but incur high computation costs per simulation, making them unsuitable for extracting effective measures. Therefore, we propose a framework that implements an MAS for constructing a training dataset, and then trains a support vector regression (SVR) model to obtain effective measure results. The proposed framework overcomes the weaknesses of conventional methods to produce effective control measure recommendations. The constructed SVR model was experimentally verified by comparing its performance on datasets with expected and unexpected outputs. Although datasets producing an unexpected output decreased the prediction accuracy, by removing randomness from the training dataset, the accuracy of the proposed method was still high in these cases. High-precision predictions of the MAS-based simulation output were obtained for both test datasets in under one second of the computational time. Furthermore, the experimental results establish that the proposed framework can obtain intuitively correct outputs for unknown inputs, and produces sufficiently high-precision prediction with lower computation costs than an existing method.

SIR model-based verification of effect of COVID-19 Contact-Confirming Application (COCOA) on reducing infectors in Japan.

As of April 2021, the coronavirus disease (COVID-19) continues to spread in Japan. To overcome COVID-19, the Ministry of Health, Labor, and Welfare of the Japanese government developed and released the COVID-19 Contact-Confirming Application (COCOA) on June 19, 2020. COCOA users can know whether they have come into contact with infectors. If persons who receive a contact notification through COCOA undertake self-quarantine, the number of infectors in Japan will decrease. However, the effectiveness of COCOA in reducing the number of infectors depends on the usage rate of COCOA, the rate of fulfillment of contact condition, the rate of undergoing the reverse transcription polymerase chain reaction (RT-PCR) test, the false negative rate of the RT-PCR test, the rate of infection registration, and the self-quarantine rate. Therefore, we developed a Susceptible-Infected-Removed (SIR) model to estimate the effectiveness of COCOA. In this paper, we introduce the SIR model and report the simulation results for different scenarios that were assumed for Japan.