Making commercial bracelet smarter with a biochemical button module

Despite the rapid development of mobile health based on wearable devices in recent years, lack of access to biochemical detection remains a vital challenge for most commercial wearable devices, which hinders the provision of effective electronic health records (EHRs) for disease control strategies, and further constraining the development of personalized precision medicine. Herein, we propose a strategy to graft biochemical detection function onto commercial bracelet. Different from the conventional development process of designing a completely new wearable biochemical device, we prefer to upgrade existing commercial wearable device to achieve simpler, faster, and more effective research and commercialization processes. An affordable and user-friendly biochemical button module has been designed that enables to integrate sensitive, specific, and rapid biochemical detection function into the idle space on the strap of the bracelet without increasing the size of the main body. This "Smart Bracelet Plus" shows the ability to simultaneously monitor physical and biochemical signals, and will serve as a reliable and systematic personal diagnostics and monitoring platform for providing real-time EHRs for disease control strategies and improving the efficiency of the healthcare system.

Assessing the relationship between cyanobacterial blooms and respiratory-related hospital visits: Green bay, Wisconsin 2017-2019

Potential acute and chronic human health effects associated with exposure to cyanobacteria and cyanotoxins, including respiratory symptoms, are an understudied public health concern. We examined the relationship between estimated cyanobacteria biomass and the frequency of respiratory-related hospital visits for residents living near Green Bay, Lake Michigan, Wisconsin during 2017-2019. Remote sensing data from the Cyanobacteria Assessment Network was used to approximate cyanobacteria exposure through creation of a metric for cyanobacteria chlorophyll-a (ChlBS). We obtained counts of hospital visits for asthma, wheezing, and allergic rhinitis from the Wisconsin Hospital Association for ZIP codes within a 3-mile radius of Green Bay. We analyzed weekly counts of hospital visits versus cyanobacteria, which was modelled as a continuous measure (ChlBS) or categorized according to World Health Organization's (WHO) alert levels using Poisson generalized linear models. Our data included 2743 individual hospital visits and 114 weeks of satellite derived cyanobacteria biomass indicator data. Peak values of ChlBS were observed between the months of June and October. Using the WHO alert levels, 60% of weeks were categorized as no risk, 19% as Vigilance Level, 15% as Alert Level 1, and 6% as Alert Level 2. In Poisson regression models adjusted for temperature, dewpoint, season, and year, there was no association between ChlBS and hospital visits (rate ratio [RR] [95% Confidence Interval (CI)] = 0.98 [0.77, 1.24]). There was also no consistent association between WHO alert level and hospital visits when adjusting for covariates (Vigilance Level: RR [95% CI] 0.88 [0.74, 1.05], Alert Level 1: 0.82 [0.67, 0.99], Alert Level 2: 0.98 [0.77, 1.24], compared to the reference no risk category). Our methodology and model provide a template for future studies that assess the association between cyanobacterial blooms and respiratory health.

Potential applications and performance of machine learning techniques and algorithms in clinical practice: A systematic review

PURPOSE

The advent of clinically adapted machine learning algorithms can solve numerous problems ranging from disease diagnosis and prognosis to therapy recommendations. This systematic review examines the performance of machine learning (ML) algorithms and evaluates the progress made to date towards their implementation in clinical practice.

METHODS

Systematic searching of databases (PubMed, MEDLINE, Scopus, Google Scholar, Cochrane Library and WHO Covid-19 database) to identify original articles published between January 2011 and October 2021. Studies reporting ML techniques in clinical practice involving humans and ML algorithms with a performance metric were considered.

RESULTS

Of 873 unique articles identified, 36 studies were eligible for inclusion. The XGBoost (extreme gradient boosting) algorithm showed the highest potential for clinical applications (n = 7 studies); this was followed jointly by random forest algorithm, logistic regression, and the support vector machine, respectively (n = 5 studies). Prediction of outcomes (n = 33), in particular Inflammatory diseases (n = 7) received the most attention followed by cancer and neuropsychiatric disorders (n = 5 for each) and Covid-19 (n = 4). Thirty-three out of the thirty-six included studies passed more than 50% of the selected quality assessment criteria in the TRIPOD checklist. In contrast, none of the studies could achieve an ideal overall bias rating of 'low' based on the PROBAST checklist. In contrast, only three studies showed evidence of the deployment of ML algorithm(s) in clinical practice.

CONCLUSIONS

ML is potentially a reliable tool for clinical decision support. Although advocated widely in clinical practice, work is still in progress to validate clinically adapted ML algorithms. Improving quality standards, transparency, and interpretability of ML models will further lower the barriers to acceptability.

Using Primary Care Data to Report Real-World Pancreatic Cancer Survival and Symptomatology

Pancreatic cancer is the 10th most common cancer diagnosed; despite recent advances in many areas of oncology, survival remains poor, in part owing to late diagnosis. Whilst primary care data are used widely for epidemiology and pharmacovigilance, they are less used for observing survival. In this study we extracted a pancreatic cancer cohort from a nationally representative English primary care database of electronic health records (EHRs) and reported on their symptom and mortality data. A total of 11, 649 cases were identified within the Oxford Royal College of General Practitioners (RCGP) Clinical Informatics Digital Hub network. All-cause mortality data was recorded for 4623 (39.69%). Mean age at recording of cancer diagnosis was 71.4 years (SD 12.0 years). 1-year and 5-year survival was 22.06% and 3.27% respectively. Within a multivariate model, age had a significant impact on survival; those diagnosed under the age of 60 had the longest survival, as compared to those age 60 - 79 (HR: 1.36, 95% CI: 1.20 - 1.54, p < 0.001) and 80+ (HR: 2.13, 95% CI: 1.86 - 2.44, p < 0.01). Symptomatology was examined; at any time point abdominal pain was the most commonly reported symptom present in 5271 cases (45.2%), but within the 12 months preceding diagnosis jaundice was the most common feature, present in 2587 patients (22.2%). Future studies clarifying other contributing factors on survival outcomes and patterns of symptomatology are needed; primary care EHRs provide an opportunity to evaluate real-world cancer patient cohort data.

Personalized predictive models for symptomatic COVID-19 patients using basic preconditions: Hospitalizations, mortality, and the need for an ICU or ventilator

BACKGROUND

The rapid global spread of the SARS-CoV-2 virus has provoked a spike in demand for hospital care. Hospital systems across the world have been over-extended, including in Northern Italy, Ecuador, and New York City, and many other systems face similar challenges. As a result, decisions on how to best allocate very limited medical resources and design targeted policies for vulnerable subgroups have come to the forefront. Specifically, under consideration are decisions on who to test, who to admit into hospitals, who to treat in an Intensive Care Unit (ICU), and who to support with a ventilator. Given today's ability to gather, share, analyze and process data, personalized predictive models based on demographics and information regarding prior conditions can be used to (1) help decision-makers allocate limited resources, when needed, (2) advise individuals how to better protect themselves given their risk profile, (3) differentiate social distancing guidelines based on risk, and (4) prioritize vaccinations once a vaccine becomes available.

OBJECTIVE

To develop personalized models that predict the following events: (1) hospitalization, (2) mortality, (3) need for ICU, and (4) need for a ventilator. To predict hospitalization, it is assumed that one has access to a patient's basic preconditions, which can be easily gathered without the need to be at a hospital and hence serve citizens and policy makers to assess individual risk during a pandemic. For the remaining models, different versions developed include different sets of a patient's features, with some including information on how the disease is progressing (e.g., diagnosis of pneumonia).

MATERIALS AND METHODS

National data from a publicly available repository, updated daily, containing information from approximately 91,000 patients in Mexico were used. The data for each patient include demographics, prior medical conditions, SARS-CoV-2 test results, hospitalization, mortality and whether a patient has developed pneumonia or not. Several classification methods were applied and compared, including robust versions of logistic regression, and support vector machines, as well as random forests and gradient boosted decision trees.

RESULTS

Interpretable methods (logistic regression and support vector machines) perform just as well as more complex models in terms of accuracy and detection rates, with the additional benefit of elucidating variables on which the predictions are based. Classification accuracies reached 72 %, 79 %, 89 %, and 90 % for predicting hospitalization, mortality, need for ICU and need for a ventilator, respectively. The analysis reveals the most important preconditions for making the predictions. For the four models derived, these are: (1) for hospitalization:age, pregnancy, diabetes, gender, chronic renal insufficiency, and immunosuppression; (2) for mortality: age, immunosuppression, chronic renal insufficiency, obesity and diabetes; (3) for ICU need: development of pneumonia (if available), age, obesity, diabetes and hypertension; and (4) for ventilator need: ICU and pneumonia (if available), age, obesity, and hypertension.

Proposing New Blockchain Challenges in eHealth

The blockchain technology has reached a great boom in the health sector, due to its importance to overcome interoperability and security challenges of the EHR and EMR systems in eHealth. The main objective of this work is to show a review of the existing research works in the literature, referring to the new blockchain technology applied in ehealth and exposing the possible research lines and trends in which this technology can be focused. The search for blockchain studies in eHealth field was carried out in the following databases: IEEE Xplore, Google Scholar, Science Direct, PubMed, Web of Science and ResearchGate from 2010 to the present. Different search criteria were established such as: "Blockchain" AND ("eHealth" OR "EHR" OR "electronic health records" OR "medicine") selecting the papers considered of most interest. A total of 84 publications on blockchain in eHealth were found, of which 18 have been identified as relevant works, 5.56% correspond to the year 2016, 22.22% to 2017 and 72.22% to 2018. Many of the publications found show how this technology is being developed and applied in the health sector and the benefits it provides. The new blockchain technology applied in eHealth identifies new ways to share the distributed view of health data and promotes the advancement of precision medicine, improving health and preventing diseases.

A Proposed Solution and Future Direction for Blockchain-Based Heterogeneous Medicare Data in Cloud Environment

The healthcare data is an important asset and rich source of healthcare intellect. Medical databases, if created properly, will be large, complex, heterogeneous and time varying. The main challenge nowadays is to store and process this data efficiently so that it can benefit humans. Heterogeneity in the healthcare sector in the form of medical data is also considered to be one of the biggest challenges for researchers. Sometimes, this data is referred to as large-scale data or big data. Blockchain technology and the Cloud environment have proved their usability separately. Though these two technologies can be combined to enhance the exciting applications in healthcare industry. Blockchain is a highly secure and decentralized networking platform of multiple computers called nodes. It is changing the way medical information is being stored and shared. It makes the work easier, keeps an eye on the security and accuracy of the data and also reduces the cost of maintenance. A Blockchain-based platform is proposed that can be used for storing and managing electronic medical records in a Cloud environment.

Extracting and utilizing electronic health data from Epic for research

Many institutions would like to harness their electronic health record (EHR) data for research. However, with many EHR systems, this process is remarkably difficult. We have been using our vast EHR system for research very effectively, with substantial research support and many publications. Herein we share our process and provide recommendations for others wanting to utilize their EHR data for research.

Electronic health records implementation: an evaluation of information system impact and contingency factors

OBJECTIVE

This paper provides a review of EHR (electronic health record) implementations around the world and reports on findings including benefits and issues associated with EHR implementation.

MATERIALS AND METHODS

A systematic literature review was conducted from peer-reviewed scholarly journal publications from the last 10 years (2001-2011). The search was conducted using various publication collections including: Scopus, Embase, Informit, Medline, Proquest Health and Medical Complete. This paper reports on our analysis of previous empirical studies of EHR implementations. We analysed data based on an extension of DeLone and McLean's information system (IS) evaluation framework. The extended framework integrates DeLone and McLean's dimensions, including information quality, system quality, service quality, intention of use and usage, user satisfaction and net benefits, together with contingent dimensions, including systems development, implementation attributes and organisational aspects, as identified by Van der Meijden and colleagues.

RESULTS

A mix of evidence-based positive and negative impacts of EHR was found across different evaluation dimensions. In addition, a number of contingent factors were found to contribute to successful implementation of EHR.

LIMITATIONS

This review does not include white papers or industry surveys, non-English papers, or those published outside the review time period.

CONCLUSION

This review confirms the potential of this technology to aid patient care and clinical documentation; for example, in improved documentation quality, increased administration efficiency, as well as better quality, safety and coordination of care. Common negative impacts include changes to workflow and work disruption. Mixed observations were found on EHR quality, adoption and satisfaction. The review warns future implementers of EHR to carefully undertake the technology implementation exercise. The review also informs healthcare providers of contingent factors that potentially affect EHR development and implementation in an organisational setting. Our findings suggest a lack of socio-technical connectives between the clinician, the patient and the technology in developing and implementing EHR and future developments in patient-accessible EHR. In addition, a synthesis of DeLone and McLean's framework and Van der Meijden and colleagues' contingent factors has been found useful in comprehensively understanding and evaluating EHR implementations.