Advanced mobile communications and portable computation are now combined in handheld devices called “smartphones”, which are also capable of running third-party software. The number of smartphone users is growing rapidly, including among healthcare professionals. The purpose of this study was to classify smartphone-based healthcare technologies as discussed in academic literature according to their functionalities, and summarize articles in each category. In April 2011, MEDLINE was searched to identify articles that discussed the design, development, evaluation, or use of smartphone-based software for healthcare professionals, medical or nursing students, or patients. A total of 55 articles discussing 83 applications were selected for this study from 2,894 articles initially obtained from the MEDLINE searches. A total of 83 applications were documented: 57 applications for healthcare professionals focusing on disease diagnosis (21), drug reference (6), medical calculators (8), literature search (6), clinical communication (3), Hospital Information System (HIS) client applications (4), medical training (2) and general healthcare applications (7); 11 applications for medical or nursing students focusing on medical education; and 15 applications for patients focusing on disease management with chronic illness (6), ENT-related (4), fall-related (3), and two other conditions (2). The disease diagnosis, drug reference, and medical calculator applications were reported as most useful by healthcare professionals and medical or nursing students. Many medical applications for smartphones have been developed and widely used by health professionals and patients. The use of smartphones is getting more attention in healthcare day by day. Medical applications make smartphones useful tools in the practice of evidence-based medicine at the point of care, in addition to their use in mobile clinical communication. Also, smartphones can play a very important role in patient education, disease self-management, and remote monitoring of patients.

/pdf/a-systematic-review-of-healthcare-applications-for-2xnx964c6m.pdf

A Systematic Review of Healthcare Applications for Smartphones

The pandemic of coronavirus disease 2019 (COVID-19), caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), has caused an unprecedented global social and economic impact, and high numbers of deaths. Many risk factors have been identified in the progression of COVID-19 into a severe and critical stage, including old age, male gender, underlying comorbidities such as hypertension, diabetes, obesity, chronic lung diseases, heart, liver and kidney diseases, tumors, clinically apparent immunodeficiencies, local immunodeficiencies, such as early type I interferon secretion capacity, and pregnancy. Possible complications include acute kidney injury, coagulation disorders, thoromboembolism. The development of lymphopenia and eosinopenia are laboratory indicators of COVID-19. Laboratory parameters to monitor disease progression include lactate dehydrogenase, procalcitonin, high-sensitivity C-reactive protein, proinflammatory cytokines such as interleukin (IL)-6, IL-1β, Krebs von den Lungen-6 (KL-6), and ferritin. The development of a cytokine storm and extensive chest computed tomography imaging patterns are indicators of a severe disease. In addition, socioeconomic status, diet, lifestyle, geographical differences, ethnicity, exposed viral load, day of initiation of treatment, and quality of health care have been reported to influence individual outcomes. In this review, we highlight the scientific evidence on the risk factors of severity of COVID-19.

Risk factors for severe and critically ill COVID-19 patients: A review.

We present an algorithmic framework for learning local causal structure around target variables of interest in the form of direct causes/effects and Markov blankets applicable to very large data sets with relatively small samples. The selected feature sets can be used for causal discovery and classification. The framework (Generalized Local Learning, or GLL) can be instantiated in numerous ways, giving rise to both existing state-of-the-art as well as novel algorithms. The resulting algorithms are sound under well-defined sufficient conditions. In a first set of experiments we evaluate several algorithms derived from this framework in terms of predictivity and feature set parsimony and compare to other local causal discovery methods and to state-of-the-art non-causal feature selection methods using real data. A second set of experimental evaluations compares the algorithms in terms of ability to induce local causal neighborhoods using simulated and resimulated data and examines the relation of predictivity with causal induction performance.

Our experiments demonstrate, consistently with causal feature selection theory, that local causal feature selection methods (under broad assumptions encompassing appropriate family of distributions, types of classifiers, and loss functions) exhibit strong feature set parsimony, high predictivity and local causal interpretability. Although non-causal feature selection methods are often used in practice to shed light on causal relationships, we find that they cannot be interpreted causally even when they achieve excellent predictivity. Therefore we conclude that only local causal techniques should be used when insight into causal structure is sought.

In a companion paper we examine in depth the behavior of GLL algorithms, provide extensions, and show how local techniques can be used for scalable and accurate global causal graph learning.

/pdf/local-causal-and-markov-blanket-induction-for-causal-1moumhfr0l.pdf

Local Causal and Markov Blanket Induction for Causal Discovery and Feature Selection for Classification Part I: Algorithms and Empirical Evaluation

Importance Thrombotic events are commonly reported in critically ill patients with COVID-19. Limited data exist to guide the intensity of antithrombotic prophylaxis. Objective To evaluate the effects of intermediate-dose vs standard-dose prophylactic anticoagulation among patients with COVID-19 admitted to the intensive care unit (ICU). Design, Setting, and Participants Multicenter randomized trial with a 2 × 2 factorial design performed in 10 academic centers in Iran comparing intermediate-dose vs standard-dose prophylactic anticoagulation (first hypothesis) and statin therapy vs matching placebo (second hypothesis; not reported in this article) among adult patients admitted to the ICU with COVID-19. Patients were recruited between July 29, 2020, and November 19, 2020. The final follow-up date for the 30-day primary outcome was December 19, 2020. Interventions Intermediate-dose (enoxaparin, 1 mg/kg daily) (n = 276) vs standard prophylactic anticoagulation (enoxaparin, 40 mg daily) (n = 286), with modification according to body weight and creatinine clearance. The assigned treatments were planned to be continued until completion of 30-day follow-up. Main Outcomes and Measures The primary efficacy outcome was a composite of venous or arterial thrombosis, treatment with extracorporeal membrane oxygenation, or mortality within 30 days, assessed in randomized patients who met the eligibility criteria and received at least 1 dose of the assigned treatment. Prespecified safety outcomes included major bleeding according to the Bleeding Academic Research Consortium (type 3 or 5 definition), powered for noninferiority (a noninferiority margin of 1.8 based on odds ratio), and severe thrombocytopenia (platelet count Results Among 600 randomized patients, 562 (93.7%) were included in the primary analysis (median [interquartile range] age, 62 [50-71] years; 237 [42.2%] women). The primary efficacy outcome occurred in 126 patients (45.7%) in the intermediate-dose group and 126 patients (44.1%) in the standard-dose prophylaxis group (absolute risk difference, 1.5% [95% CI, −6.6% to 9.8%]; odds ratio, 1.06 [95% CI, 0.76-1.48];P = .70). Major bleeding occurred in 7 patients (2.5%) in the intermediate-dose group and 4 patients (1.4%) in the standard-dose prophylaxis group (risk difference, 1.1% [1-sided 97.5% CI, −∞ to 3.4%]; odds ratio, 1.83 [1-sided 97.5% CI, 0.00-5.93]), not meeting the noninferiority criteria (Pfor noninferiority >.99). Severe thrombocytopenia occurred only in patients assigned to the intermediate-dose group (6 vs 0 patients; risk difference, 2.2% [95% CI, 0.4%-3.8%];P = .01). Conclusions and Relevance Among patients admitted to the ICU with COVID-19, intermediate-dose prophylactic anticoagulation, compared with standard-dose prophylactic anticoagulation, did not result in a significant difference in the primary outcome of a composite of adjudicated venous or arterial thrombosis, treatment with extracorporeal membrane oxygenation, or mortality within 30 days. These results do not support the routine empirical use of intermediate-dose prophylactic anticoagulation in unselected patients admitted to the ICU with COVID-19. Trial Registration ClinicalTrials.gov Identifier:NCT04486508

/pdf/effect-of-intermediate-dose-vs-standard-dose-prophylactic-19f7s3nxz2.pdf

Effect of Intermediate-Dose vs Standard-Dose Prophylactic Anticoagulation on Thrombotic Events, Extracorporeal Membrane Oxygenation Treatment, or Mortality Among Patients With COVID-19 Admitted to the Intensive Care Unit: The INSPIRATION Randomized Clinical Trial.

Bio-medical image segmentation is one of the promising sectors where nuclei segmentation from high-resolution histopathological images enables extraction of very high-quality features for nuclear morphometrics and other analysis metrics in the field of digital pathology. The traditional methods including Otsu thresholding and watershed methods do not work properly in different challenging cases. However, Deep Learning (DL) based approaches are showing tremendous success in different modalities of bio-medical imaging including computation pathology. Recently, the Recurrent Residual U-Net (R2U-Net) has been proposed, which has shown state-of-the-art (SOTA) performance in different modalities (retinal blood vessel, skin cancer, and lung segmentation) in medical image segmentation. However, in this implementation, the R2U-Net is applied to nuclei segmentation for the first time on a publicly available dataset that was collected from the Data Science Bowl Grand Challenge in 2018. The R2U-Net shows around 92.15% segmentation accuracy in terms of the Dice Coefficient (DC) during the testing phase. In addition, the qualitative results show accurate segmentation, which clearly demonstrates the robustness of the R2U-Net model for the nuclei segmentation task.

Nuclei Segmentation with Recurrent Residual Convolutional Neural Networks based U-Net (R2U-Net)

This study examines the incidence of and risk factors for venous and arterial thrombosis in patients hospitalized with COVID-19 in 4 New York City hospitals.

/pdf/thrombosis-in-hospitalized-patients-with-covid-19-in-a-new-11zbj5m0rv.pdf

Thrombosis in Hospitalized Patients With COVID-19 in a New York City Health System.

Studies in Health Technology and Informatics

https://onlinelibrary.wiley.com/doi/pdf/10.1111/jth.15534

Platelets contribute to disease severity in COVID-19.

Rapid increases in e-cigarette use and potential exposure to harmful byproducts have shifted public health focus to e-cigarettes as a possible drug of abuse. Effective surveillance of use and prevalence would allow appropriate regulatory responses. An ideal surveillance system would collect usage data in real time, focus on populations of interest, include populations unable to take the survey, allow a breadth of questions to answer, and enable geo-location analysis. Social media streams may provide this ideal system. To realize this use case, a foundational question is whether we can detect e-cigarette use at all. This work reports two pilot tasks using text classification to identify automatically Tweets that indicate e-cigarette use and/or e-cigarette use for smoking cessation. We build and define both datasets and compare performance of 4 state of the art classifiers and a keyword search for each task. Our results demonstrate excellent classifier performance of up to 0.90 and 0.94 area under the curve in each category. These promising initial results form the foundation for further studies to realize the ideal surveillance solution.

/pdf/text-classification-for-automatic-detection-of-e-cigarette-4wnzpz33iu.pdf

Text classification for automatic detection of e-cigarette use and use for smoking cessation from twitter: a feasibility pilot.

We present a feasibility study using text classification to classify tweets about alcohol use Alcohol use is the most widely used substance in the US and is the leading risk factor for premature morbidity and mortality globally Understanding use patterns and locations is an important step toward prevention, moderation, and control of alcohol outlets Social media may provide an alternate way to measure alcohol use in real time This feasibility study explores text classification methodologies for identifying alcohol use tweets We labeled 34,563 geo-located New York City tweets collected in a 24 hour period over New Year's Day 2012 We preprocessed the tweets into stem/ not stemmed and unigram/ bigram representations We then applied multinomial naive Bayes, a linear SVM, Bayesian logistic regression, and random forests to the classification task Using 10 fold cross-validation, the algorithms performed with area under the receiver operating curve of 066, 091, 093, and 094 respectively We also compare to a human constructed Boolean search for the same tweets and the text classification method is competitive with this hand crafted search In conclusion, we show that the task of automatically identifying alcohol related tweets is highly feasible and paves the way for future research to improve these classifiers

Yin Aphinyanaphongs

Papers

Thrombosis in Hospitalized Patients With COVID-19 in a New York City Health System.

Studies in Health Technology and Informatics

Platelets contribute to disease severity in COVID-19.

Text classification for automatic detection of e-cigarette use and use for smoking cessation from twitter: a feasibility pilot.

Text classification for automatic detection of alcohol use-related tweets: A feasibility study