Dataset Information

Patient-specific COVID-19 resource utilization prediction using fusion AI model.

ABSTRACT: The strain on healthcare resources brought forth by the recent COVID-19 pandemic has highlighted the need for efficient resource planning and allocation through the prediction of future consumption. Machine learning can predict resource utilization such as the need for hospitalization based on past medical data stored in electronic medical records (EMR). We conducted this study on 3194 patients (46% male with mean age 56.7 (±16.8), 56% African American, 7% Hispanic) flagged as COVID-19 positive cases in 12 centers under Emory Healthcare network from February 2020 to September 2020, to assess whether a COVID-19 positive patient's need for hospitalization can be predicted at the time of RT-PCR test using the EMR data prior to the test. Five main modalities of EMR, i.e., demographics, medication, past medical procedures, comorbidities, and laboratory results, were used as features for predictive modeling, both individually and fused together using late, middle, and early fusion. Models were evaluated in terms of precision, recall, F1-score (within 95% confidence interval). The early fusion model is the most effective predictor with 84% overall F1-score [CI 82.1-86.1]. The predictive performance of the model drops by 6 % when using recent clinical data while omitting the long-term medical history. Feature importance analysis indicates that history of cardiovascular disease, emergency room visits in the past year prior to testing, and demographic factors are predictive of the disease trajectory. We conclude that fusion modeling using medical history and current treatment data can forecast the need for hospitalization for patients infected with COVID-19 at the time of the RT-PCR test.

SUBMITTER: Tariq A

PROVIDER: S-EPMC8175333 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:BackgroundThe COVID-19 pandemic has highlighted the inability of health systems to leverage existing system infrastructure in order to rapidly develop and apply broad analytical tools that could inform state- and national-level policymaking, as well as patient care delivery in hospital settings. The COVID-19 pandemic has also led to highlighted systemic disparities in health outcomes and access to care based on race or ethnicity, gender, income-level, and urban-rural divide. Although the United States seems to be recovering from the COVID-19 pandemic owing to widespread vaccination efforts and increased public awareness, there is an urgent need to address the aforementioned challenges.ObjectiveThis study aims to inform the feasibility of leveraging broad, statewide datasets for population health-driven decision-making by developing robust analytical models that predict COVID-19-related health care resource utilization across patients served by Indiana's statewide Health Information Exchange.MethodsWe leveraged comprehensive datasets obtained from the Indiana Network for Patient Care to train decision forest-based models that can predict patient-level need of health care resource utilization. To assess these models for potential biases, we tested model performance against subpopulations stratified by age, race or ethnicity, gender, and residence (urban vs rural).ResultsFor model development, we identified a cohort of 96,026 patients from across 957 zip codes in Indiana, United States. We trained the decision models that predicted health care resource utilization by using approximately 100 of the most impactful features from a total of 1172 features created. Each model and stratified subpopulation under test reported precision scores >70%, accuracy and area under the receiver operating curve scores >80%, and sensitivity scores approximately >90%. We noted statistically significant variations in model performance across stratified subpopulations identified by age, race or ethnicity, gender, and residence (urban vs rural).ConclusionsThis study presents the possibility of developing decision models capable of predicting patient-level health care resource utilization across a broad, statewide region with considerable predictive performance. However, our models present statistically significant variations in performance across stratified subpopulations of interest. Further efforts are necessary to identify root causes of these biases and to rectify them.

Project description:IntroductionKey measures in preventing spread of the virus that causes coronavirus disease 2019 (COVID-19) are social distancing and stay-at-home mandates. These measures along with other stressors have the potential to increase incidences of intimate partner violence (IPV), sexual assault, and child maltreatment.MethodsWe performed a retrospective review of county police dispatches, emergency department (ED) visits, Sexual Assault Nurse Examiner (SANE) consults, Domestic Violence Healthcare Project (DVHP) team consults, and Child Protection Team consults at a large, tertiary, Level I trauma center. We queried International Classification of Diseases Revision 10 codes most specific to IPV, sexual assault, and child maltreatment from March-October 2020 compared to 2019. Similarly, the number of consults performed by SANE, DVHP, and our Child Protection Team were collected. We compared all ED visits and consultations to total ED visits for the reviewed time period. Finally, the total number of calls and referrals to a child advocacy center and resource call line for victims were recorded during this timeframe.ResultsPolice dispatches for IPV-related assaults increased by 266 reports from 2019 to 2020 (P = 0.015). Emergency department visits related to IPV increased from 0.11% of visits in 2019 to 0.15% in 2020 (P = 0.032), and DVHP consults increased from 0.31% in 2019 to 0.48% in 2020 of ED visits in the first three months (P < 0.001). Child maltreatment visits increased from 0.47% of visits in 2019 to 0.81% of visits in 2020 (P = 0.028), and a higher percentage of patients required Child Protection team consults from 1% in 2019 to 1.6% in 2020 (P = 0.004). Sexual assault-related visits and SANE consults both showed a small increase that was not statistically significant. Fewer calls and referrals were made to our child advocacy center and resource call line, decreasing by 99 referrals and 252 calls, respectively.ConclusionDespite decreased ED volumes throughout the pandemic, we observed an increase in police dispatches, ED visits, and utilization of hospital consult services related to IPV and child maltreatment following the initiation of stay-at-home orders. However, use of community resources, such as the local child advocacy center, declined.

Project description:Hospitals in Canada are facing a crisis-level shortage of critical supplies and equipment during the COVID-19 pandemic. This motivates us to create predictive models that can use Canada COVID-19 data and pandemic-related factors to accurately forecast 5 quantities-three related to hospital resource utilization (i.e., the number of hospital beds, ICU beds, and ventilators that will be needed by COVID-19 patients) and two to the pandemic progress (i.e., the number of COVID-19 cases and COVID-19 deaths)-several weeks in advance. We developed a machine learning method that can use information (i.e., resource utilization, pandemic progress, population mobility, weather condition, and public policy) currently known about a region since March 2020, to learn multiple temporal convolutional network (TCN) models every week; each used for forecasting the weekly average of one of these 5 quantities in Canada (respectively, in six specific provinces) for each, in the next 1 (resp., 2,3,4) weeks. To validate the effectiveness of our method, we compared our method, versus other standard models, on the COVID-19 data and hospital resource data, on the tasks of predicting the 116 values (for Canada and its six most populated provinces), every week from Oct 2020 to July 2021, and the 20 values (only for Canada) for four specific times within 9 July to 31 Dec 2021. Experimental results show that our 4640 TCN models (each forecasting a regional target for a specific future time, on a specific date) can produce accurate 1,2,3,4-week forecasts of the utilization of every hospital resource and pandemic progress for each week from 2 Oct 2020 to 2 July 2021, as well as 80 TCN models for each of the four specified times within 9 July and 31 Dec 2021. Compared to other baseline and state-of-the-art predictive models, our TCN models yielded the best forecasts, with the lowest mean absolute percentage error (MAPE). Additional experiments, on the IHME COVID-19 data, demonstrate the effectiveness of our TCN models, in comparison with IHME forecasts. Each of our TCN models used a pre-defined set of features; we experimentally validate the effectiveness of these features by showing that these models perform better than other models that instead used other features. Overall, these experimental results demonstrate that our method can accurately forecast hospital resource utilization and pandemic progress for Canada and for each of the six provinces.

Project description:BackgroundGiven the severity and scope of the current COVID-19 pandemic, it is critical to determine predictive features of COVID-19 mortality and medical resource usage to effectively inform health, risk-based physical distancing, and work accommodation policies. Non-clinical sociodemographic features are important explanatory variables of COVID-19 outcomes, revealing existing disparities in large health care systems.Methods and findingsWe use nation-wide multicenter data of COVID-19 patients in Brazil to predict mortality and ventilator usage. The dataset contains hospitalized patients who tested positive for COVID-19 and had either recovered or were deceased between March 1 and June 30, 2020. A total of 113,214 patients with 50,387 deceased, were included. Both interpretable (sparse versions of Logistic Regression and Support Vector Machines) and state-of-the-art non-interpretable (Gradient Boosted Decision Trees and Random Forest) classification methods are employed. Death from COVID-19 was strongly associated with demographics, socioeconomic factors, and comorbidities. Variables highly predictive of mortality included geographic location of the hospital (OR = 2.2 for Northeast region, OR = 2.1 for North region); renal (OR = 2.0) and liver (OR = 1.7) chronic disease; immunosuppression (OR = 1.7); obesity (OR = 1.7); neurological (OR = 1.6), cardiovascular (OR = 1.5), and hematologic (OR = 1.2) disease; diabetes (OR = 1.4); chronic pneumopathy (OR = 1.4); immunosuppression (OR = 1.3); respiratory symptoms, ranging from respiratory discomfort (OR = 1.4) and dyspnea (OR = 1.3) to oxygen saturation less than 95% (OR = 1.7); hospitalization in a public hospital (OR = 1.2); and self-reported patient illiteracy (OR = 1.1). Validation accuracies (AUC) for predicting mortality and ventilation need reach 79% and 70%, respectively, when using only pre-admission variables. Models that use post-admission disease progression information reach accuracies (AUC) of 86% and 87% for predicting mortality and ventilation use, respectively.ConclusionsThe results highlight the predictive power of socioeconomic information in assessing COVID-19 mortality and medical resource allocation, and shed light on existing disparities in the Brazilian health care system during the COVID-19 pandemic.

Dataset Information

Patient-specific COVID-19 resource utilization prediction using fusion AI model.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets