Dataset Information

Real-World Implications of a Rapidly Responsive COVID-19 Spread Model with Time-Dependent Parameters via Deep Learning: Model Development and Validation.

ABSTRACT: BACKGROUND:The COVID-19 pandemic has caused major disruptions worldwide since March 2020. The experience of the 1918 influenza pandemic demonstrated that decreases in the infection rates of COVID-19 do not guarantee continuity of the trend. OBJECTIVE:The aim of this study was to develop a precise spread model of COVID-19 with time-dependent parameters via deep learning to respond promptly to the dynamic situation of the outbreak and proactively minimize damage. METHODS:In this study, we investigated a mathematical model with time-dependent parameters via deep learning based on forward-inverse problems. We used data from the Korea Centers for Disease Control and Prevention (KCDC) and the Center for Systems Science and Engineering (CSSE) at Johns Hopkins University for Korea and the other countries, respectively. Because the data consist of confirmed, recovered, and deceased cases, we selected the susceptible-infected-recovered (SIR) model and found approximated solutions as well as model parameters. Specifically, we applied fully connected neural networks to the solutions and parameters and designed suitable loss functions. RESULTS:We developed an entirely new SIR model with time-dependent parameters via deep learning methods. Furthermore, we validated the model with the conventional Runge-Kutta fourth order model to confirm its convergent nature. In addition, we evaluated our model based on the real-world situation reported from the KCDC, the Korean government, and news media. We also crossvalidated our model using data from the CSSE for Italy, Sweden, and the United States. CONCLUSIONS:The methodology and new model of this study could be employed for short-term prediction of COVID-19, which could help the government prepare for a new outbreak. In addition, from the perspective of measuring medical resources, our model has powerful strength because it assumes all the parameters as time-dependent, which reflects the exact status of viral spread.

SUBMITTER: Jung SY

PROVIDER: S-EPMC7486001 | biostudies-literature | 2020 Sep

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Real-World Implications of a Rapidly Responsive COVID-19 Spread Model with Time-Dependent Parameters via Deep Learning: Model Development and Validation.

Jung Se Young SY Jo Hyeontae H Son Hwijae H Hwang Hyung Ju HJ

Journal of medical Internet research 20200909 9

<h4>Background</h4>The COVID-19 pandemic has caused major disruptions worldwide since March 2020. The experience of the 1918 influenza pandemic demonstrated that decreases in the infection rates of COVID-19 do not guarantee continuity of the trend.<h4>Objective</h4>The aim of this study was to develop a precise spread model of COVID-19 with time-dependent parameters via deep learning to respond promptly to the dynamic situation of the outbreak and proactively minimize damage.<h4>Methods</h4>In t ...[more]

PMID: 32877350

Similar Datasets

Project description:BackgroundModeling plays a critical role in mitigating impacts of seasonal influenza epidemics. Complex simulation models are currently at the forefront of evaluating optimal mitigation strategies at multiple scales and levels of organization. Given their evaluative role, these models remain limited in their ability to predict and forecast future epidemics leading some researchers and public-health practitioners to question their usefulness. The objective of this study is to evaluate the predictive ability of an existing complex simulation model of influenza spread.Methods and findingsWe used extensive data on past epidemics to demonstrate the process of predictive validation. This involved generalizing an individual-based model for influenza spread and fitting it to laboratory-confirmed influenza infection data from a single observed epidemic (1998-1999). Next, we used the fitted model and modified two of its parameters based on data on real-world perturbations (vaccination coverage by age group and strain type). Simulating epidemics under these changes allowed us to estimate the deviation/error between the expected epidemic curve under perturbation and observed epidemics taking place from 1999 to 2006. Our model was able to forecast absolute intensity and epidemic peak week several weeks earlier with reasonable reliability and depended on the method of forecasting-static or dynamic.ConclusionsGood predictive ability of influenza epidemics is critical for implementing mitigation strategies in an effective and timely manner. Through the process of predictive validation applied to a current complex simulation model of influenza spread, we provided users of the model (e.g. public-health officials and policy-makers) with quantitative metrics and practical recommendations on mitigating impacts of seasonal influenza epidemics. This methodology may be applied to other models of communicable infectious diseases to test and potentially improve their predictive ability.

Project description:ObjectiveThe RNS System is a direct brain-responsive neurostimulation system that is US Food and Drug Administration-approved for adults with medically intractable focal onset seizures based on safety and effectiveness data from controlled clinical trials. The purpose of this study was to retrospectively evaluate the real-world safety and effectiveness of the RNS System.MethodsEight comprehensive epilepsy centers conducted a chart review of patients treated with the RNS System for at least 1 year, in accordance with the indication for use. Data included device-related serious adverse events and the median percent change in disabling seizure frequency from baseline at years 1, 2, and 3 of treatment and at the most recent follow-up.ResultsOne hundred fifty patients met the criteria for analysis. The median reduction in seizures was 67% (interquartile range [IQR] = 33%-93%, n = 149) at 1 year, 75% (IQR = 50%-94%, n = 93) at 2 years, 82% (IQR = 50%-96%, n = 38) at ≥3 years, and 74% (IQR = 50%-96%, n = 150) at last follow-up (mean = 2.3 years). Thirty-five percent of patients had a ≥90% seizure frequency reduction, and 18% of patients reported being clinically seizure-free at last follow-up. Seizure frequency reductions were similar regardless of patient age, age at epilepsy onset, duration of epilepsy, seizure onset in mesial temporal or neocortical foci, magnetic resonance imaging findings, prior intracranial monitoring, prior epilepsy surgery, or prior vagus nerve stimulation treatment. The infection rate per procedure was 2.9% (6/150 patients); five of the six patients had an implant site infection, and one had osteomyelitis. Lead revisions were required in 2.7% (4/150), and 2.0% (3/150) of patients had a subdural hemorrhage, none of which had long-lasting neurological consequences.SignificanceIn this real-world experience, safety was similar and clinical seizure outcomes exceeded those of the prospective clinical trials, corroborating effectiveness of this therapy and suggesting that clinical experience has informed more effective programming.

Project description:BackgroundFor medical artificial intelligence (AI) training and validation, human expert labels are considered the gold standard that represents the correct answers or desired outputs for a given data set. These labels serve as a reference or benchmark against which the model's predictions are compared.ObjectiveThis study aimed to assess the accuracy of a custom deep learning (DL) algorithm on classifying diabetic retinopathy (DR) and further demonstrate how label errors may contribute to this assessment in a nationwide DR-screening program.MethodsFundus photographs from the Lifeline Express, a nationwide DR-screening program, were analyzed to identify the presence of referable DR using both (1) manual grading by National Health Service England-certificated graders and (2) a DL-based DR-screening algorithm with validated good lab performance. To assess the accuracy of labels, a random sample of images with disagreement between the DL algorithm and the labels was adjudicated by ophthalmologists who were masked to the previous grading results. The error rates of labels in this sample were then used to correct the number of negative and positive cases in the entire data set, serving as postcorrection labels. The DL algorithm's performance was evaluated against both pre- and postcorrection labels.ResultsThe analysis included 736,083 images from 237,824 participants. The DL algorithm exhibited a gap between the real-world performance and the lab-reported performance in this nationwide data set, with a sensitivity increase of 12.5% (from 79.6% to 92.5%, P<.001) and a specificity increase of 6.9% (from 91.6% to 98.5%, P<.001). In the random sample, 63.6% (560/880) of negative images and 5.2% (140/2710) of positive images were misclassified in the precorrection human labels. High myopia was the primary reason for misclassifying non-DR images as referable DR images, while laser spots were predominantly responsible for misclassified referable cases. The estimated label error rate for the entire data set was 1.2%. The label correction was estimated to bring about a 12.5% enhancement in the estimated sensitivity of the DL algorithm (P<.001).ConclusionsLabel errors based on human image grading, although in a small percentage, can significantly affect the performance evaluation of DL algorithms in real-world DR screening.

Dataset Information

Real-World Implications of a Rapidly Responsive COVID-19 Spread Model with Time-Dependent Parameters via Deep Learning: Model Development and Validation.

Publications

Real-World Implications of a Rapidly Responsive COVID-19 Spread Model with Time-Dependent Parameters via Deep Learning: Model Development and Validation.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets