Adaptive COVID-19 Screening of a Subpopulation

Methods are sought to test adaptively whether a subpopulation proportion follows the same time evolution as the population proportion. The motivating case study is the COVID-19 screening in a university community, taking into account the time evolution of the pandemic in the whole country. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.

Adaptive COVID-19 Screening of a Subpopulation

Methods are sought to test adaptively whether a subpopulation proportion follows the same time evolution as the population proportion. The motivating case study is the COVID-19 screening in a university community, taking into account the time evolution of the pandemic in the whole country. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.

Intelligence Methods to Fight Covid-19 Spread

According to the COVID-19 worldwide sickness, which has wreaked devastation, over than millions of individuals from all over the globe have been afflicted. The COVID-19 virus infected a significant number of people worldwide as a result of both the latency in detecting its existence in the female organism. A.i. (AI) and Computer Vision (ML) may assist in identifying, treatment, and assessing the severity of COVID-besides all the conventional approaches now present. In order to fully understand the role of AI and ML as a crucial tool for COVID-19 and related outbreak detection, forecasting, forecasts, contacts tracking, and therapy formulation, this study aims to offer a comprehensive review of the topic. AI revolutionises diagnostic accuracy in terms of efficiency and precision. This technology holds promise for a self-driving and visible surveillance system that can enable real - time and treat people avoiding spreading the virus to others. Digital Healthcare different applications have also been discovered. This essay investigates how AI may help fight the COVID-19 pandemic. We make an effort to provide an AI-based hospital design. Ai systems (AI) is used in the infrastructure to effectively and quickly carry out health care, assessment, and treatment. © 2022 IEEE.

Detecting early signals of COVID-19 outbreaks in 2020 in small areas by monitoring healthcare utilisation databases: first lessons learned from the Italian Alert_CoV project.

BackgroundDuring the COVID-19 pandemic, large-scale diagnostic testing and contact tracing have proven insufficient to promptly monitor the spread of infections.AimTo develop and retrospectively evaluate a system identifying aberrations in the use of selected healthcare services to timely detect COVID-19 outbreaks in small areas.MethodsData were retrieved from the healthcare utilisation (HCU) databases of the Lombardy Region, Italy. We identified eight services suggesting a respiratory infection (syndromic proxies). Count time series reporting the weekly occurrence of each proxy from 2015 to 2020 were generated considering small administrative areas (i.e. census units of Cremona and Mantua provinces). The ability to uncover aberrations during 2020 was tested for two algorithms: the improved Farrington algorithm and the generalised likelihood ratio-based procedure for negative binomial counts. To evaluate these algorithms' performance in detecting outbreaks earlier than the standard surveillance, confirmed outbreaks, defined according to the weekly number of confirmed COVID-19 cases, were used as reference. Performances were assessed separately for the first and second semester of the year. Proxies positively impacting performance were identified.ResultsWe estimated that 70% of outbreaks could be detected early using the proposed approach, with a corresponding false positive rate of ca 20%. Performance did not substantially differ either between algorithms or semesters. The best proxies included emergency calls for respiratory or infectious disease causes and emergency room visits.ConclusionImplementing HCU-based monitoring systems in small areas deserves further investigations as it could facilitate the containment of COVID-19 and other unknown infectious diseases in the future.

Insights from Review and Content Analysis of Current COVID-19 Mobile Apps and Recommendations for Future Pandemics.

A number of mobile health apps related to coronavirus infectious disease 2019 (COVID-19) have been developed, but research into app content analytics for effective surveillance and management is still in its preliminary stages. The present study aimed to identify the purpose and functions of the currently available COVID-19 apps using content analysis. The secondary aim was to propose directions for the future development of apps that aid infectious disease surveillance and control with a focus on enhancing the app content and quality. Prior to conducting an app search in the App Store and the Google Play Store, we reviewed previous studies on COVID-19 apps found in Google Scholar and PubMed to examine the main purposes of the apps. Using the five selected keywords based on the review, we searched the two app stores to retrieve eligible COVID-19 apps including those already addressed in the reviewed literature. We conducted descriptive and content analyses of the selected apps. We classified the purpose types of the COVID-19 apps into the following five categories: Information provision, tracking, monitoring, mental health management, and engagement. We identified 890 apps from the review articles and the app stores: 47 apps met the selection criteria and were included in the content analysis. Among the selected apps, iOS apps outnumbered Android apps, 27 apps were government-developed, and most of the apps were created in the United States. The most common function for the iOS apps (63.6%) and Android apps (62.5%) was to provide COVID-19-related knowledge. The most common function among the tracking apps was to notify users of contact with infected people by the iOS apps (40.9%) and Android apps (37.5%). About 29.5% of the iOS apps and 25.0% of the Android apps were used to record symptoms and self-diagnose. Significantly fewer apps targeted mental health management and engagement. Six iOS apps (6/44, 13.6%) and four Android apps (4/24, 16.7%) provided behavioral guidelines about the pandemic. Two iOS apps (2/44, 4.5%) and two Android apps (2/24, 8.3%) featured communication functions. The present content analysis revealed that most of the apps provided unilateral information and contact tracing or location tracking. Several apps malfunctioned. Future research and development of COVID-19 apps or apps for other emerging infectious diseases should address the quality and functional improvements, which should begin with continuous monitoring and actions to mitigate any technical errors.

Efficient algorithms for real-time syndromic surveillance.

OBJECTIVE: Outbreaks of influenza-like diseases often cause spikes in the demand for hospital beds. Early detection of these outbreaks can enable improved management of hospital resources. The objective of this study was to test whether surveillance algorithms designed to be responsive to a wide range of anomalous decreases in the time between emergency department (ED) presentations with influenza-like illnesses provide efficient early detection of these outbreaks. METHODS: Our study used data on ED presentations to major public hospitals in Queensland, Australia across 2017-2020. We developed surveillance algorithms for each hospital that flag potential outbreaks when the average time between successive ED presentations with influenza-like illnesses becomes anomalously small. We designed one set of algorithms to be responsive to a wide range of anomalous decreases in the time between presentations. These algorithms concurrently monitor three exponentially weighted moving averages (EWMAs) of the time between presentations and flag an outbreak when at least one EWMA falls below its control limit. We designed another set of algorithms to be highly responsive to narrower ranges of anomalous decreases in the time between presentations. These algorithms monitor one EWMA of the time between presentations and flag an outbreak when the EWMA falls below its control limit. Our algorithms use dynamic control limits to reflect that the average time between presentations depends on the time of year, time of day, and day of the week. RESULTS: We compared the performance of the algorithms in detecting the start of two epidemic events at the hospital-level: the 2019 seasonal influenza outbreak and the early-2020 COVID-19 outbreak. The algorithm that concurrently monitors three EWMAs provided significantly earlier detection of these outbreaks than the algorithms that monitor one EWMA. CONCLUSION: Surveillance algorithms designed to be responsive to a wide range of anomalous decreases in the time between ED presentations are highly efficient at detecting outbreaks of influenza-like diseases at the hospital level.

LIBKDV: A Versatile Kernel Density Visualization Library for Geospatial Analytics

Kernel density visualization (KDV) has been widely used in many geospatial analysis tasks, including traffic accident hotspot detection, crime hotspot detection, and disease outbreak detection. Although KDV can be supported by many scientific, geographical, and visualization software tools, none of these tools can support high-resolution KDV with large-scale datasets. Therefore, we develop the first versatile programming library, called LIBKDV, based on the set of our complexity-optimized algorithms. Given the high efficiency of these algorithms, LIBKDV not only accelerates the KDV computation but also enriches KDV-based geospatial analytics, including bandwidth-tuning analysis and spatiotemporal analysis, which cannot be natively and feasibly supported by existing software tools. In this demonstration, participants will be invited to use our programming library to explore interesting hotspot patterns on large-scale traffic accident, crime, and COVID-19 datasets. © 2022, VLDB Endowment. All rights reserved.

Detection of Infectious Disease Outbreaks in Search Engine Time Series Using Non-Specific Syndromic Surveillance with Effect-Size Filtering

Novel infectious disease outbreaks, including most recently that of the COVID-19 pandemic, could be detected by non-specific syndromic surveillance systems. Such systems, utilizing a variety of data sources ranging from Electronic Health Records to internet data such as aggregated search engine queries, create alerts when unusually high rates of symptom reports occur. This is especially important for the detection of novel diseases, where their manifested symptoms are unknown. Here we improve upon a set of previously-proposed non-specific syndromic surveillance methods by taking into account both how unusual a preponderance of symptoms is and their effect size. We demonstrate that our method is as accurate as previously-proposed methods for low dimensional data and show its effectiveness for high-dimensional aggregated data by applying it to aggregated time-series health-related search engine queries. We find that in 2019 the method would have raised alerts related to several disease outbreaks earlier than health authorities did. During the COVID-19 pandemic the system identified the beginning of pandemic waves quickly, through combinations of symptoms which varied from wave to wave. Thus, the proposed method could be used as a practical tool for decision makers to detect new disease outbreaks using time series derived from search engine data even in the absence of specific information on the diseases of interest and their symptoms. © 2022 ACM.

Early detection of COVID-19 outbreaks using textual analysis of electronic medical records.

PURPOSE: Our objective was to develop a tool promoting early detection of COVID-19 cases by focusing epidemiological investigations and PCR examinations during a period of limited testing capabilities. METHODS: We developed an algorithm for analyzing medical records recorded by healthcare providers in the Israeli Defense Forces. The algorithm utilized textual analysis to detect patients presenting with suspicious symptoms and was tested among 92 randomly selected units. Detection of a potential cluster of patients in a unit prompted a focused epidemiological investigation aided by data provided by the algorithm. RESULTS: During a month of follow up, the algorithm has flagged 17 of the units for investigation. The subsequent epidemiological investigations led to the testing of 78 persons and the detection of eight cases in four clusters that were previously gone unnoticed. The resulting positive test rate of 10.25% was five time higher than the IDF average at the time of the study. No cases of COVID-19 in the examined units were missed by the algorithm. CONCLUSIONS: This study depicts the successful development and large scale deployment of a textual analysis based algorithm for early detection of COVID-19 cases, demonstrating the potential of natural language processing of medical text as a tool for promoting public health.

Catwalk: identifying closely related sequences in large microbial sequence databases.

There is a need to identify microbial sequences that may form part of transmission chains, or that may represent importations across national boundaries, amidst large numbers of SARS-CoV-2 and other bacterial or viral sequences. Reference-based compression is a sequence analysis technique that allows both a compact storage of sequence data and comparisons between sequences. Published implementations of the approach are being challenged by the large sample collections now being generated. Our aim was to develop a fast software detecting highly similar sequences in large collections of microbial genomes, including millions of SARS-CoV-2 genomes. To do so, we developed Catwalk, a tool that bypasses bottlenecks in the generation, comparison and in-memory storage of microbial genomes generated by reference mapping. It is a compiled solution, coded in Nim to increase performance. It can be accessed via command line, rest api or web server interfaces. We tested Catwalk using both SARS-CoV-2 and Mycobacterium tuberculosis genomes generated by prospective public-health sequencing programmes. Pairwise sequence comparisons, using clinically relevant similarity cut-offs, took about 0.39 and 0.66 µs, respectively; in 1 s, between 1 and 2 million sequences can be searched. Catwalk operates about 1700 times faster than, and uses about 8 % of the RAM of, a Python reference-based compression and comparison tool in current use for outbreak detection. Catwalk can rapidly identify close relatives of a SARS-CoV-2 or M. tuberculosis genome amidst millions of samples.

Finding disease outbreak locations from human mobility data.

Finding the origin location of an infectious disease outbreak quickly is crucial in mitigating its further dissemination. Current methods to identify outbreak locations early on rely on interviewing affected individuals and correlating their movements, which is a manual, time-consuming, and error-prone process. Other methods such as contact tracing, genomic sequencing or theoretical models of epidemic spread offer help, but they are not applicable at the onset of an outbreak as they require highly processed information or established transmission chains. Digital data sources such as mobile phones offer new ways to find outbreak sources in an automated way. Here, we propose a novel method to determine outbreak origins from geolocated movement data of individuals affected by the outbreak. Our algorithm scans movement trajectories for shared locations and identifies the outbreak origin as the most dominant among them. We test the method using various empirical and synthetic datasets, and demonstrate that it is able to single out the true outbreak location with high accuracy, requiring only data of N = 4 individuals. The method can be applied to scenarios with multiple outbreak locations, and is even able to estimate the number of outbreak sources if unknown, while being robust to noise. Our method is the first to offer a reliable, accurate out-of-the-box approach to identify outbreak locations in the initial phase of an outbreak. It can be easily and quickly applied in a crisis situation, improving on previous manual approaches. The method is not only applicable in the context of disease outbreaks, but can be used to find shared locations in movement data in other contexts as well. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1140/epjds/s13688-021-00306-6.

Geographically weighted generalized Farrington algorithm for rapid outbreak detection over short data accumulation periods.

The demand for rapid surveillance and early detection of local outbreaks has been growing recently. The rapid surveillance can select timely and appropriate interventions toward controlling the spread of emerging infectious diseases, such as the coronavirus disease 2019 (COVID-19). The Farrington algorithm was originally proposed by Farrington et al (1996), extended by Noufaily et al (2012), and is commonly used to estimate excess death. However, one of the major challenges in implementing this algorithm is the lack of historical information required to train it, especially for emerging diseases. Without sufficient training data the estimation/prediction accuracy of this algorithm can suffer leading to poor outbreak detection. We propose a new statistical algorithm-the geographically weighted generalized Farrington (GWGF) algorithm-by incorporating both geographically varying and geographically invariant covariates, as well as geographical information to analyze time series count data sampled from a spatially correlated process for estimating excess death. The algorithm is a type of local quasi-likelihood-based regression with geographical weights and is designed to achieve a stable detection of outbreaks even when the number of time points is small. We validate the outbreak detection performance by using extensive numerical experiments and real-data analysis in Japan during COVID-19 pandemic. We show that the GWGF algorithm succeeds in improving recall without reducing the level of precision compared with the conventional Farrington algorithm.

Risk Diagrams Based on Primary Care Electronic Medical Records and Linked Real-Time PCR Data to Monitor Local COVID-19 Outbreaks During the Summer 2020: A Prospective Study Including 7,671,862 People in Catalonia.

Monitoring transmission is a prerequisite for containing COVID-19. We report on effective potential growth (EPG) as a novel measure for the early identification of local outbreaks based on primary care electronic medical records (EMR) and PCR-confirmed cases. Secondly, we studied whether increasing EPG precedes local hospital and intensive care (ICU) admissions and mortality. Population-based cohort including all Catalan citizens' PCR tests, hospitalization, intensive care (ICU) and mortality between 1/07/2020 and 13/09/2020; linked EMR covering 88.6% of the Catalan population was obtained. Nursing home residents were excluded. COVID-19 counts were ascertained based on EMR and PCRs separately. Weekly empirical propagation (ρ7) and 14-day cumulative incidence (A14) and 95% confidence intervals were estimated at care management area (CMA) level, and combined as EPG = ρ7 × A14. Overall, 7,607,201 and 6,798,994 people in 43 CMAs were included for PCR and EMR measures, respectively. A14, ρ7, and EPG increased in numerous CMAs during summer 2020. EMR identified 2.70-fold more cases than PCRs, with similar trends, a median (interquartile range) 2 (1) days earlier, and better precision. Upticks in EPG preceded increases in local hospital admissions, ICU occupancy, and mortality. Increasing EPG identified localized outbreaks in Catalonia, and preceded local hospital and ICU admissions and subsequent mortality. EMRs provided similar estimates to PCR, but some days earlier and with better precision. EPG is a useful tool for the monitoring of community transmission and for the early identification of COVID-19 local outbreaks.

The African Network for Improved Diagnostics, Epidemiology and Management of common infectious Agents.

BACKGROUND: In sub-Saharan Africa, acute respiratory infections (ARI), acute gastrointestinal infections (GI) and acute febrile disease of unknown cause (AFDUC) have a large disease burden, especially among children, while respective aetiologies often remain unresolved. The need for robust infectious disease surveillance to detect emerging pathogens along with common human pathogens has been highlighted by the ongoing novel coronavirus disease 2019 (COVID-19) pandemic. The African Network for Improved Diagnostics, Epidemiology and Management of Common Infectious Agents (ANDEMIA) is a sentinel surveillance study on the aetiology and clinical characteristics of ARI, GI and AFDUC in sub-Saharan Africa. METHODS: ANDEMIA includes 12 urban and rural health care facilities in four African countries (Côte d'Ivoire, Burkina Faso, Democratic Republic of the Congo and Republic of South Africa). It was piloted in 2018 in Côte d'Ivoire and the initial phase will run from 2019 to 2021. Case definitions for ARI, GI and AFDUC were established, as well as syndrome-specific sampling algorithms including the collection of blood, naso- and oropharyngeal swabs and stool. Samples are tested using comprehensive diagnostic protocols, ranging from classic bacteriology and antimicrobial resistance screening to multiplex real-time polymerase chain reaction (PCR) systems and High Throughput Sequencing. In March 2020, PCR testing for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and analysis of full genomic information was included in the study. Standardised questionnaires collect relevant clinical, demographic, socio-economic and behavioural data for epidemiologic analyses. Controls are enrolled over a 12-month period for a nested case-control study. Data will be assessed descriptively and aetiologies will be evaluated using a latent class analysis among cases. Among cases and controls, an integrated analytic approach using logistic regression and Bayesian estimation will be employed to improve the assessment of aetiology and associated risk factors. DISCUSSION: ANDEMIA aims to expand our understanding of ARI, GI and AFDUC aetiologies in sub-Saharan Africa using a comprehensive laboratory diagnostics strategy. It will foster early detection of emerging threats and continued monitoring of important common pathogens. The network collaboration will be strengthened and site diagnostic capacities will be reinforced to improve quality management and patient care.

Monitoring scheme for early detection of coronavirus and other respiratory virus outbreaks.

In December 2019, an outbreak of pneumonia caused by a novel coronavirus (severe acute respiratory syndrome coronavirus 2 [SARS-CoV-2]) began in Wuhan, China. SARS-CoV-2 exhibited efficient person-to-person transmission of what became labeled as COVID-19. It has spread worldwide with over 83,000,000 infected cases and more than 1,800,000 deaths to date (December 31, 2020). This research proposes a statistical monitoring scheme in which an optimized np control chart is utilized by sentinel metropolitan airports worldwide for early detection of coronavirus and other respiratory virus outbreaks. The sample size of this chart is optimized to ensure the best overall performance for detecting a wide range of shifts in the infection rate, based on the available resources, such as the inspection rate and the allowable false alarm rate. The effectiveness of the proposed optimized np chart is compared with that of the traditional np chart with a predetermined sample size under both sampling inspection and 100% inspection. For a variety of scenarios including a real case, the optimized np control chart is found to substantially outperform its traditional counterpart in terms of the average number of infections. Therefore, this control chart has potential to be an effective tool for early detection of respiratory virus outbreaks, promoting early outbreak investigation and mitigation.

Early detection of influenza outbreak using time derivative of incidence.

For mitigation strategies of an influenza outbreak, it can be helpful to understand the characteristics of regional and age-group-specific spread. In South Korea, however, there has been no official statistic related to it. In this study, we extract the time series of influenza incidence from National Health Insurance Service claims database, which consists of all medical and prescription drug-claim records for all South Korean population. The extracted time series contains the number of new patients by region (250 city-county-districts) and age-group (0-4, 5-19, 20-64, 65+) within a week. The number of cases of influenza (2009-2017) is 12,282,356. For computing an onset of influenza outbreak by region and age-group, we propose a novel method for early outbreak detection, in which the onset of outbreak is detected as a sudden change in the time derivative of incidence. The advantage of it over the cumulative sum and the exponentially weighted moving average control charts, which have been widely used for the early outbreak detection of infectious diseases, is that information on the previous non-epidemic periods are not necessary. Then, we show that the metro area and 5-19 age-group are earlier than the rural area and other age-groups for the start of the influenza outbreak. Also, the metro area and 5-19 age-group peak earlier than the rural area and other age-groups. These results would be helpful to design a surveillance system for timely early warning of an influenza outbreak in South Korea.

Citizen-Centered Mobile Health Apps Collecting Individual-Level Spatial Data for Infectious Disease Management: Scoping Review.

BACKGROUND: The novel coronavirus SARS-CoV-2 rapidly spread around the world, causing the disease COVID-19. To contain the virus, much hope is placed on participatory surveillance using mobile apps, such as automated digital contact tracing, but broad adoption is an important prerequisite for associated interventions to be effective. Data protection aspects are a critical factor for adoption, and privacy risks of solutions developed often need to be balanced against their functionalities. This is reflected by an intensive discussion in the public and the scientific community about privacy-preserving approaches. OBJECTIVE: Our aim is to inform the current discussions and to support the development of solutions providing an optimal balance between privacy protection and pandemic control. To this end, we present a systematic analysis of existing literature on citizen-centered surveillance solutions collecting individual-level spatial data. Our main hypothesis is that there are dependencies between the following dimensions: the use cases supported, the technology used to collect spatial data, the specific diseases focused on, and data protection measures implemented. METHODS: We searched PubMed and IEEE Xplore with a search string combining terms from the area of infectious disease management with terms describing spatial surveillance technologies to identify studies published between 2010 and 2020. After a two-step eligibility assessment process, 27 articles were selected for the final analysis. We collected data on the four dimensions described as well as metadata, which we then analyzed by calculating univariate and bivariate frequency distributions. RESULTS: We identified four different use cases, which focused on individual surveillance and public health (most common: digital contact tracing). We found that the solutions described were highly specialized, with 89% (24/27) of the articles covering one use case only. Moreover, we identified eight different technologies used for collecting spatial data (most common: GPS receivers) and five different diseases covered (most common: COVID-19). Finally, we also identified six different data protection measures (most common: pseudonymization). As hypothesized, we identified relationships between the dimensions. We found that for highly infectious diseases such as COVID-19 the most common use case was contact tracing, typically based on Bluetooth technology. For managing vector-borne diseases, use cases require absolute positions, which are typically measured using GPS. Absolute spatial locations are also important for further use cases relevant to the management of other infectious diseases. CONCLUSIONS: We see a large potential for future solutions supporting multiple use cases by combining different technologies (eg, Bluetooth and GPS). For this to be successful, however, adequate privacy-protection measures must be implemented. Technologies currently used in this context can probably not offer enough protection. We, therefore, recommend that future solutions should consider the use of modern privacy-enhancing techniques (eg, from the area of secure multiparty computing and differential privacy).

Unsupervised Anomaly Detection in Multivariate Spatio-Temporal Data Using Deep Learning: Early Detection of COVID-19 Outbreak in Italy.

Unsupervised anomaly detection for spatio-temporal data has extensive use in a wide variety of applications such as earth science, traffic monitoring, fraud and disease outbreak detection. Most real-world time series data have a spatial dimension as an additional context which is often expressed in terms of coordinates of the region of interest (such as latitude - longitude information). However, existing techniques are limited to handle spatial and temporal contextual attributes in an integrated and meaningful way considering both spatial and temporal dependency between observations. In this paper, a hybrid deep learning framework is proposed to solve the unsupervised anomaly detection problem in multivariate spatio-temporal data. The proposed framework works with unlabeled data and no prior knowledge about anomalies are assumed. As a case study, we use the public COVID-19 data provided by the Italian Department of Civil Protection. Northern Italy regions' COVID-19 data are used to train the framework; and then any abnormal trends or upswings in COVID-19 data of central and southern Italian regions are detected. The proposed framework detects early signals of the COVID-19 outbreak in test regions based on the reconstruction error. For performance comparison, we perform a detailed evaluation of 15 algorithms on the COVID-19 Italy dataset including the state-of-the-art deep learning architectures. Experimental results show that our framework shows significant improvement on unsupervised anomaly detection performance even in data scarce and high contamination ratio scenarios (where the ratio of anomalies in the data set is more than 5%). It achieves the earliest detection of COVID-19 outbreak and shows better performance on tracking the peaks of the COVID-19 pandemic in test regions. As the timeliness of detection is quite important in the fight against any outbreak, our framework provides useful insight to suppress the resurgence of local novel coronavirus outbreaks as early as possible.

Social media based surveillance systems for healthcare using machine learning: A systematic review.

BACKGROUND: Real-time surveillance in the field of health informatics has emerged as a growing domain of interest among worldwide researchers. Evolution in this field has helped in the introduction of various initiatives related to public health informatics. Surveillance systems in the area of health informatics utilizing social media information have been developed for early prediction of disease outbreaks and to monitor diseases. In the past few years, the availability of social media data, particularly Twitter data, enabled real-time syndromic surveillance that provides immediate analysis and instant feedback to those who are charged with follow-ups and investigation of potential outbreaks. In this paper, we review the recent work, trends, and machine learning(ML) text classification approaches used by surveillance systems seeking social media data in the healthcare domain. We also highlight the limitations and challenges followed by possible future directions that can be taken further in this domain. METHODS: To study the landscape of research in health informatics performing surveillance of the various health-related data posted on social media or web-based platforms, we present a bibliometric analysis of the 1240 publications indexed in multiple scientific databases (IEEE, ACM Digital Library, ScienceDirect, PubMed) from the year 2010-2018. The papers were further reviewed based on the various machine learning algorithms used for analyzing health-related text posted on social media platforms. FINDINGS: Based on the corpus of 148 selected articles, the study finds the types of social media or web-based platforms used for surveillance in the healthcare domain, along with the health topic(s) studied by them. In the corpus of selected articles, we found 26 articles were using machine learning technique. These articles were studied to find commonly used ML techniques. The majority of studies (24%) focused on the surveillance of flu or influenza-like illness (ILI). Twitter (64%) is the most popular data source to perform surveillance research using social media text data, and Support Vector Machine (SVM) (33%) being the most used ML algorithm for text classification. CONCLUSIONS: The inclusion of online data in surveillance systems has improved the disease prediction ability over traditional syndromic surveillance systems. However, social media based surveillance systems have many limitations and challenges, including noise, demographic bias, privacy issues, etc. Our paper mentions future directions, which can be useful for researchers working in the area. Researchers can use this paper as a library for social media based surveillance systems in the healthcare domain and can expand such systems by incorporating the future works discussed in our paper.