Learning models for forecasting hospital resource utilization for COVID-19 patients in Canada.

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.

Learning accurate personalized survival models for predicting hospital discharge and mortality of COVID-19 patients.

Since it emerged in December of 2019, COVID-19 has placed a huge burden on medical care in countries throughout the world, as it led to a huge number of hospitalizations and mortalities. Many medical centers were overloaded, as their intensive care units and auxiliary protection resources proved insufficient, which made the effective allocation of medical resources an urgent matter. This study describes learned survival prediction models that could help medical professionals make effective decisions regarding patient triage and resource allocation. We created multiple data subsets from a publicly available COVID-19 epidemiological dataset to evaluate the effectiveness of various combinations of covariates-age, sex, geographic location, and chronic disease status-in learning survival models (here, "Individual Survival Distributions"; ISDs) for hospital discharge and also for death events. We then supplemented our datasets with demographic and economic information to obtain potentially more accurate survival models. Our extensive experiments compared several ISD models, using various measures. These results show that the "gradient boosting Cox machine" algorithm outperformed the competing techniques, in terms of these performance evaluation metrics, for predicting both an individual's likelihood of hospital discharge and COVID-19 mortality. Our curated datasets and code base are available at our Github repository for reproducing the results reported in this paper and for supporting future research.

Detecting Presence of PTSD Using Sentiment Analysis From Text Data.

Rates of Post-traumatic stress disorder (PTSD) have risen significantly due to the COVID-19 pandemic. Telehealth has emerged as a means to monitor symptoms for such disorders. This is partly due to isolation or inaccessibility of therapeutic intervention caused from the pandemic. Additional screening tools may be needed to augment identification and diagnosis of PTSD through a virtual medium. Sentiment analysis refers to the use of natural language processing (NLP) to extract emotional content from text information. In our study, we train a machine learning (ML) model on text data, which is part of the Audio/Visual Emotion Challenge and Workshop (AVEC-19) corpus, to identify individuals with PTSD using sentiment analysis from semi-structured interviews. Our sample size included 188 individuals without PTSD, and 87 with PTSD. The interview was conducted by an artificial character (Ellie) over a video-conference call. Our model was able to achieve a balanced accuracy of 80.4% on a held out dataset used from the AVEC-19 challenge. Additionally, we implemented various partitioning techniques to determine if our model was generalizable enough. This shows that learned models can use sentiment analysis of speech to identify the presence of PTSD, even through a virtual medium. This can serve as an important, accessible and inexpensive tool to detect mental health abnormalities during the COVID-19 pandemic.

Detecting Presence of PTSD Using Sentiment Analysis From Text Data

Rates of Post-traumatic stress disorder (PTSD) have risen significantly due to the COVID-19 pandemic. Telehealth has emerged as a means to monitor symptoms for such disorders. This is partly due to isolation or inaccessibility of therapeutic intervention caused from the pandemic. Additional screening tools may be needed to augment identification and diagnosis of PTSD through a virtual medium. Sentiment analysis refers to the use of natural language processing (NLP) to extract emotional content from text information. In our study, we train a machine learning (ML) model on text data, which is part of the Audio/Visual Emotion Challenge and Workshop (AVEC-19) corpus, to identify individuals with PTSD using sentiment analysis from semi-structured interviews. Our sample size included 188 individuals without PTSD, and 87 with PTSD. The interview was conducted by an artificial character (Ellie) over a video-conference call. Our model was able to achieve a balanced accuracy of 80.4% on a held out dataset used from the AVEC-19 challenge. Additionally, we implemented various partitioning techniques to determine if our model was generalizable enough. This shows that learned models can use sentiment analysis of speech to identify the presence of PTSD, even through a virtual medium. This can serve as an important, accessible and inexpensive tool to detect mental health abnormalities during the COVID-19 pandemic.

SIMLR: Machine Learning inside the SIR Model for COVID-19 Forecasting

Accurate forecasts of the number of newly infected people during an epidemic are critical for making effective timely decisions. This paper addresses this challenge using the SIMLR model, which incorporates machine learning (ML) into the epidemiological SIR model. For each region, SIMLR tracks the changes in the policies implemented at the government level, which it uses to estimate the time-varying parameters of an SIR model for forecasting the number of new infections one to four weeks in advance. It also forecasts the probability of changes in those government policies at each of these future times, which is essential for the longer-range forecasts. We applied SIMLR to data from in Canada and the United States, and show that its mean average percentage error is as good as state-of-the-art forecasting models, with the added advantage of being an interpretable model. We expect that this approach will be useful not only for forecasting COVID-19 infections, but also in predicting the evolution of other infectious diseases.

A longitudinal dataset of incidence and intervention policy impacts regarding the COVID-19 pandemic in Canadian provinces.

One year after identifying the first case of the 2019 coronavirus disease (COVID-19) in Canada, federal and provincial governments are still struggling to manage the pandemic. Provincial governments across Canada have experimented with widely varying policies in order to limit the burden of COVID-19. However, to date, the effectiveness of these policies has been difficult to ascertain. This is partly due to the lack of a publicly available, high-quality dataset on COVID-19 interventions and outcomes for Canada. The present paper provides a dataset containing important, Canadian-specific data that is known to affect COVID-19 outcomes, including sociodemographic, climatic, mobility and health system related information for all 10 Canadian provinces and their health regions. This dataset also includes longitudinal data on the daily number of COVID-19 cases, deaths, and the constantly changing intervention policies that have been implemented by each province in an attempt to control the pandemic.

Dataset of COVID-19 outbreak and potential predictive features in the USA.

This dataset provides information related to the outbreak of COVID-19 disease in the United States, including data from each of 3142 US counties from the beginning of the outbreak (January 2020) until June 2021. This data is collected from many public online databases and includes the daily number of COVID-19 confirmed cases and deaths, as well as 46 features that may be relevant to the pandemic dynamics: demographic, geographic, climatic, traffic, public-health, social-distancing-policy adherence, and political characteristics of each county. We anticipate many researchers will use this dataset to train models that can predict the spread of COVID-19 and to identify the key driving factors.

Gender Differences in Satisfaction With a Text Messaging Program (Text4Hope) and Anticipated Receptivity to Technology-Based Health Support During the COVID-19 Pandemic: Cross-sectional Survey Study.

BACKGROUND: In March 2020, Text4Hope-a community health service-was provided to Alberta residents. This free service aims to promote psychological resilience and alleviate pandemic-associated stress, anxiety, and depression symptoms during the COVID-19 pandemic. OBJECTIVE: This study aimed to evaluate the feedback, satisfaction, experience, and perceptions of Text4Hope subscribers and to examine any differences based on gender after subscribers received 6 weeks of daily supportive text messages. Additionally, this study examined subscribers' anticipated receptivity to technology-based medical services that could be offered during major crises, emergencies, or pandemics. METHODS: Individuals self-subscribed to Text4Hope to receive daily supportive text messages for 3 months. Subscribers were invited to complete a web-based survey at 6 weeks postintervention to provide service satisfaction-related information. Overall satisfaction was assessed on a scale of 0-10, and satisfaction scores were analyzed using a related-measures t test. Likert scale satisfaction responses were used to assess various aspects of the Text4Hope program. Gender differences were analyzed using one-way analysis of variance (ANOVA) and Chi-square analyses. RESULTS: A total of 2032 subscribers completed the baseline and 6-week surveys; 1788 (88%) were female, 219 (10.8%) were male, and 25 (1.2%) were other gender. The mean age of study participants was 44.58 years (SD 13.45 years). The mean overall satisfaction score was 8.55 (SD 1.78), suggesting high overall satisfaction with Text4Hope. The ANOVA analysis, which was conducted using the Welch test (n=1716), demonstrated that females had significantly higher mean satisfaction scores than males (8.65 vs 8.11, respectively; mean difference=0.546; 95% CI 0.19 to 0.91; P<.001) and nonsignificantly lower satisfaction scores than other gender respondents (mean difference=-0.938; 95% CI -0.37 to 2.25; P=.15). More than 70% of subscribers agreed that Text4Hope helped them cope with stress (1334/1731, 77.1%) and anxiety (1309/1728, 75.8%), feel connected to a support system (1400/1729, 81%), manage COVID-19-related issues (1279/1728, 74%), and improve mental well-being (1308/1731, 75.6%). Similarly, subscribers agreed that messages were positive, affirmative, and succinct. Messages were always or often read by 97.9% (1681/1716) of respondents, and more than 20% (401/1716, 23.4%) always or often returned to messages. The majority of subscribers (1471/1666, 88.3%) read the messages and either reflected upon them or took a positive action. Subscribers welcomed almost all technology-based services as part of their health care during crisis or emergency situations. Text4Hope was perceived to be effective by many female subscribers, who reported higher satisfaction and improved coping after receiving text messages for 6 weeks. CONCLUSIONS: Respondents affirmed the high quality of the text messages with their positive feedback. Technology-based services can provide remotely accessible and population-level interventions that align with the recommended physical distancing practices for pandemics. Text4Hope subscriber feedback revealed high satisfaction and acceptance at 6 weeks postintervention. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): RR2-10.2196/19292.

Automated detection of pneumonia in lung ultrasound using deep video classification for COVID-19.

There is a crucial need for quick testing and diagnosis of patients during the COVID-19 pandemic. Lung ultrasound is an imaging modality that is cost-effective, widely accessible, and can be used to diagnose acute respiratory distress syndrome in patients with COVID-19. It can be used to find important characteristics in the images, including A-lines, B-lines, consolidation, and pleural effusion, which all inform the clinician in monitoring and diagnosing the disease. With the use of portable ultrasound transducers, lung ultrasound images can be easily acquired, however, the images are often of poor quality. They often require an expert clinician interpretation, which may be time-consuming and is highly subjective. We propose a method for fast and reliable interpretation of lung ultrasound images by use of deep learning, based on the Kinetics-I3D network. Our learned model can classify an entire lung ultrasound scan obtained at point-of-care, without requiring the use of preprocessing or a frame-by-frame analysis. We compare our video classifier against ground truth classification annotations provided by a set of expert radiologists and clinicians, which include A-lines, B-lines, consolidation, and pleural effusion. Our classification method achieves an accuracy of 90% and an average precision score of 95% with the use of 5-fold cross-validation. The results indicate the potential use of automated analysis of portable lung ultrasound images to assist clinicians in screening and diagnosing patients.

Accurate long-range forecasting of COVID-19 mortality in the USA.

The need for improved models that can accurately predict COVID-19 dynamics is vital to managing the pandemic and its consequences. We use machine learning techniques to design an adaptive learner that, based on epidemiological data available at any given time, produces a model that accurately forecasts the number of reported COVID-19 deaths and cases in the United States, up to 10 weeks into the future with a mean absolute percentage error of 9%. In addition to being the most accurate long-range COVID predictor so far developed, it captures the observed periodicity in daily reported numbers. Its effectiveness is based on three design features: (1) producing different model parameters to predict the number of COVID deaths (and cases) from each time and for a given number of weeks into the future, (2) systematically searching over the available covariates and their historical values to find an effective combination, and (3) training the model using "last-fold partitioning", where each proposed model is validated on only the last instance of the training dataset, rather than being cross-validated. Assessments against many other published COVID predictors show that this predictor is 19-48% more accurate.

Text4Hope: Receiving Daily Supportive Text Messages for 3 Months During the COVID-19 Pandemic Reduces Stress, Anxiety, and Depression.

BACKGROUND: This study reports on the changes in stress, anxiety, and depressive symptoms of subscribers after 3 months using Text4Hope, a supportive text messaging program designed to provide support during the pandemic. METHODS: Standardized self-report measures were used to evaluate perceived stress (measured with the Perceived Stress Scale-10 [PSS-10]), anxiety (measured with the General Anxiety Disorder Scale 7 [GAD-7]), and depressive symptoms (measured with the Patient Health Questionnaire [PHQ-9]), at baseline and 3rd month (n = 373). RESULTS: After 3 months of using Text4Hope, subscribers' self-reports revealed significant (p< 0.001) mean score reductions compared with baseline on: the GAD-7 by 22.7%, PHQ-9 by 10.3%, and PSS-10 scores by 5.7%. Reductions in inferred prevalence rates for moderate to high symptoms were also observed, with anxiety demonstrating the largest reduction (15.7%). CONCLUSIONS: Observed Text4Hope-related reductions in psychological distress during COVID-19 indicate that Text4Hope is an effective, convenient, and accessible means of implementing a population-level psychological intervention.

Mental Health Outreach via Supportive Text Messages during the COVID-19 Pandemic: Improved Mental Health and Reduced Suicidal Ideation after Six Weeks in Subscribers of Text4Hope Compared to a Control Population.

Background: In March 2020, Alberta Health Services launched Text4Hope, a free mental health text-message service. The service aimed to alleviate pandemic-associated stress, generalized anxiety disorder (GAD), major depressive disorder (MDD), and suicidal propensity. The effectiveness of Text4Hope was evaluated by comparing psychiatric parameters between two subscriber groups. Methods: A comparative cross-sectional study with two arms: Text4Hope subscribers who received daily texts for six weeks, the intervention group (IG); and new Text4Hope subscribers who were yet to receive messages, the control group (CG). Logistic regression models were used in the analysis. Results: Participants in the IG had lower prevalence rates for moderate/high stress (78.8% vs. 88.0%), likely GAD (31.4% vs. 46.5%), and likely MDD (36.8% vs. 52.1%), respectively, compared to respondents in the CG. After controlling for demographic variables, the IG remained less likely to self-report symptoms of moderate/high stress (OR = 0.56; 95% CI = 0.41-0.75), likely GAD (OR = 0.55; 95% CI = 0.44-0.68), and likely MDD (OR = 0.50; 95% CI = 0.47-0.73). The mean Composite Mental Health score, the sum of mean scores on the PSS, GAD-7, and PHQ-9 was 20.9% higher in the CG. Conclusions: Text4Hope is an effective population-level intervention that helps reduce stress, anxiety, depression, and suicidal thoughts during the COVID-19 pandemic. Similar texting services should be implemented during global crises.

Health Care Providers' Performance, Mindset, and Attitudes Toward a Neonatal Resuscitation Computer-Based Simulator: Empirical Study.

BACKGROUND: Neonatal resuscitation involves a complex sequence of actions to establish an infant's cardiorespiratory function at birth. Many of these responses, which identify the best action sequence in each situation, are taught as part of the recurrent Neonatal Resuscitation Program training, but they have a low incidence in practice, which leaves health care providers (HCPs) less prepared to respond appropriately and efficiently when they do occur. Computer-based simulators are increasingly used to complement traditional training in medical education, especially in the COVID-19 pandemic era of mass transition to digital education. However, it is not known how learners' attitudes toward computer-based learning and assessment environments influence their performance. OBJECTIVE: This study explores the relation between HCPs' attitudes toward a computer-based simulator and their performance in the computer-based simulator, RETAIN (REsuscitation TrAINing), to uncover the predictors of performance in computer-based simulation environments for neonatal resuscitation. METHODS: Participants were 50 neonatal HCPs (45 females, 4 males, 1 not reported; 16 respiratory therapists, 33 registered nurses and nurse practitioners, and 1 physician) affiliated with a large university hospital. Participants completed a demographic presurvey before playing the game and an attitudinal postsurvey after completing the RETAIN game. Participants' survey responses were collected to measure attitudes toward the computer-based simulator, among other factors. Knowledge on neonatal resuscitation was assessed in each round of the game through increasingly difficult neonatal resuscitation scenarios. This study investigated the moderating role of mindset on the association between the perceived benefits of understanding the terminology used in the computer-based simulator, RETAIN, and their performance on the neonatal resuscitation tasks covered by RETAIN. RESULTS: The results revealed that mindset moderated the relation between participants' perceived terminology used in RETAIN and their actual performance in the game (F3,44=4.56, R2=0.24, adjusted R2=0.19; P=.007; estimate=-1.19, SE=0.38, t44=-3.12, 95% CI -1.96 to -0.42; P=.003). Specifically, participants who perceived the terminology useful also performed better but only when endorsing more of a growth mindset; they also performed worse when endorsing more of a fixed mindset. Most participants reported that they enjoyed playing the game. The more the HCPs agreed that the terminology in the tutorial and in the game was accessible, the better they performed in the game, but only when they reported endorsing a growth mindset exceeding the average mindset of all the participants (F3,44=6.31, R2=0.30, adjusted R2=0.25; P=.001; estimate=-1.21, SE=0.38, t44=-3.16, 95% CI -1.99 to -0.44; P=.003). CONCLUSIONS: Mindset moderates the strength of the relationship between HCPs' perception of the role that the terminology employed in a game simulator has on their performance and their actual performance in a computer-based simulator designed for neonatal resuscitation training. Implications of this research include the design and development of interactive learning environments that can support HCPs in performing better on neonatal resuscitation tasks.

Changes in Stress, Anxiety, and Depression Levels of Subscribers to a Daily Supportive Text Message Program (Text4Hope) During the COVID-19 Pandemic: Cross-Sectional Survey Study.

BACKGROUND: In addition to the obvious physical medical impact of COVID-19, the disease poses evident threats to people's mental health, psychological safety, and well-being. Provision of support for these challenges is complicated by the high number of people requiring support and the need to maintain physical distancing. Text4Hope, a daily supportive SMS text messaging program, was launched in Canada to mitigate the negative mental health impacts of the pandemic among Canadians. OBJECTIVE: This paper describes the changes in the stress, anxiety, and depression levels of subscribers to the Text4Hope program after 6 weeks of exposure to daily supportive SMS text messages. METHODS: We used self-administered, empirically supported web-based questionnaires to assess the demographic and clinical characteristics of Text4Hope subscribers. Perceived stress, anxiety, and depression were measured with the 10-Item Perceived Stress Scale (PSS-10), the Generalized Anxiety Disorder-7 (GAD-7) scale, and the Patient Health Questionnaire-9 (PHQ-9) scale at baseline and sixth week time points. Moderate or high perceived stress, likely generalized anxiety disorder, and likely major depressive disorder were assessed using cutoff scores of ≥14 for the PSS-10, ≥10 for the GAD-7, and ≥10 for the PHQ-9, respectively. At 6 weeks into the program, 766 participants had completed the questionnaires at both time points. RESULTS: At the 6-week time point, there were statistically significant reductions in mean scores on the PSS-10 and GAD-7 scales but not on the PHQ-9 scale. Effect sizes were small overall. There were statistically significant reductions in the prevalence rates of moderate or high stress and likely generalized anxiety disorder but not likely major depressive disorder for the group that completed both the baseline and 6-week assessments. The largest reductions in mean scores and prevalence rates were for anxiety (18.7% and 13.5%, respectively). CONCLUSIONS: Text4Hope is a convenient, cost-effective, and accessible means of implementing a population-level psychological intervention. This service demonstrated significant reductions in anxiety and stress levels during the COVID-19 pandemic and could be used as a population-level mental health intervention during natural disasters and other emergencies. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): RR2-10.2196/19292.