https://www.jmir.org/stats/feed
Purchases: 0 ]]> https://www.jmir.org/2018/2/e47 Mon, 12 Feb 2018 09:00:03 EST https://www.jmir.org/2018/2/e47
Purchases: 0 ]]> https://www.jmir.org/2013/9/e214 Tue, 24 Sep 2013 10:46:11 EDT https://www.jmir.org/2013/9/e214
Purchases: 0 ]]> https://www.jmir.org/2016/5/e134 Fri, 27 May 2016 11:00:04 EDT https://www.jmir.org/2016/5/e134
Purchases: 0 ]]> https://www.jmir.org/2018/6/e10202 Thu, 07 Jun 2018 09:30:02 EDT https://www.jmir.org/2018/6/e10202
Purchases: 0 ]]> https://www.jmir.org/2017/8/e304 Tue, 29 Aug 2017 11:12:45 EDT https://www.jmir.org/2017/8/e304
Purchases: 0 ]]> https://www.jmir.org/2017/2/e34 Wed, 15 Feb 2017 10:00:03 EST https://www.jmir.org/2017/2/e34 Background: Performing systematic reviews is a time-consuming and resource-intensive process. Objective: We investigated whether a machine learning system could perform systematic reviews more efficiently. Methods: All systematic reviews and meta-analyses of interventional randomized controlled trials cited in recent clinical guidelines from the American Diabetes Association, American College of Cardiology, American Heart Association (2 guidelines), and American Stroke Association were assessed. After reproducing the primary screening data set according to the published search strategy of each, we extracted correct articles (those actually reviewed) and incorrect articles (those not reviewed) from the data set. These 2 sets of articles were used to train a neural network–based artificial intelligence engine (Concept Encoder, Fronteo Inc). The primary endpoint was work saved over sampling at 95% recall (WSS@95%). Results: Among 145 candidate reviews of randomized controlled trials, 8 reviews fulfilled the inclusion criteria. For these 8 reviews, the machine learning system significantly reduced the literature screening workload by at least 6-fold versus that of manual screening based on WSS@95%. When machine learning was initiated using 2 correct articles that were randomly selected by a researcher, a 10-fold reduction in workload was achieved versus that of manual screening based on the WSS@95% value, with high sensitivity for eligible studies. The area under the receiver operating characteristic curve increased dramatically every time the algorithm learned a correct article. Conclusions: Concept Encoder achieved a 10-fold reduction of the screening workload for systematic review after learning from 2 randomly selected studies on the target topic. However, few meta-analyses of randomized controlled trials were included. Concept Encoder could facilitate the acquisition of evidence for clinical guidelines.

Purchases: 0 ]]> https://www.jmir.org/2020/12/e22422 Wed, 30 Dec 2020 09:00:28 EST https://www.jmir.org/2020/12/e22422
Purchases: 0 ]]> https://www.jmir.org/2005/5/e56 Thu, 17 Nov 2005 00:00:00 EST https://www.jmir.org/2005/5/e56
Purchases: 0 ]]> https://www.jmir.org/2020/12/e21815 Wed, 30 Dec 2020 10:00:03 EST https://www.jmir.org/2020/12/e21815 Background: The COVID-19 pandemic has resulted in profound mental health impacts among the general population worldwide. As many in-person mental health support services have been suspended or transitioned online to facilitate physical distancing, there have been numerous calls for the rapid expansion of asynchronous virtual mental health (AVMH) resources. These AVMH resources have great potential to provide support for people coping with negative mental health impacts associated with the pandemic; however, literature examining use prior to COVID-19 illustrates that the uptake of these resources is consistently low. Objective: The aim of this paper is to examine the use of AVMH resources in Canada during the COVID-19 pandemic among the general population and among a participant subgroup classified as experiencing an adverse mental health impact related to the pandemic. Methods: Data from this study were drawn from the first wave of a large multiwave cross-sectional monitoring survey, distributed from May 14 to 29, 2020. Participants (N=3000) were adults living in Canada. Descriptive statistics were used to characterize the sample, and bivariate cross-tabulations were used to examine the relationships between the use of AVMH resources and self-reported indicators of mental health that included a range of emotional and coping-related responses to the pandemic. Univariate and fully adjusted multivariate logistic regression models were used to examine associations between sociodemographic and health-related characteristics and use of AVMH resources in the subgroup of participants who reported experiencing one or more adverse mental health impacts identified in the set of self-reported mental health indicators. Results: Among the total sample, 2.0% (n=59) of participants reported accessing AVMH resources in the prior 2 weeks to cope with stress related to the COVID-19 pandemic, with the highest rates of use among individuals who reported self-harm (n=5, 10.4%) and those who reported coping “not well” with COVID-19–related stress (n=22, 5.5%). Within the subgroup of 1954 participants (65.1% of the total sample) who reported an adverse mental health impact related to COVID-19, 54 (2.8%) reported use of AVMH resources. Individuals were more likely to have used AVMH resources if they had reported receiving in-person mental health supports, were connecting virtually with a mental health worker or counselor, or belonged to a visible minority group. Conclusions: Despite substantial government investment into AVMH resources, uptake is low among both the general population and individuals who may benefit from the use of these resources as a means of coping with the adverse mental health impacts of the COVID-19 pandemic. Further research is needed to improve our understanding of the barriers to use.

Purchases: 0 ]]> https://www.jmir.org/2020/12/e24868 Wed, 30 Dec 2020 09:45:03 EST https://www.jmir.org/2020/12/e24868