Back

The SENSOR System: Using Standardized Data Entry and Dashboards for Review of Scientific Studies Using the Clinical Applications of Psychedelics as an Illustration

Kamal, L.; Godsell, M. P.; Mulligan, B.; Eberspaecher, S.; Myint, D.; Besemann, M.; Minerbi, A.; Gupta, G.

2022-01-21 health informatics
10.1101/2022.01.14.22269304 medRxiv
Show abstract

IntroductionLiterature reviews are useful tools for communicating the breadth of scientific discovery for a given topic. Irrespective of the nature of the review, data should be communicated in effective, easy to understand ways. In trying to address these limitations of traditional scientific reviews, we propose using dynamic data driven displays that have been used in multiple other industries to improve communication and decision making. Given the recent interest in the clinical applications of psychedelics for various mental health issues, we chose to test the SENSOR System (Standardized Data Entry and Dashboards for Review of Scientific Studies) as an alternative for an existing review article. MethodsTo validate the SENSOR System, an existing review with a topical, heterogenous, and growing set of studies was selected. In this case we chose the Wheeler et al. review on psychedelics in clinical practice where articles had already been preselected and reviewed. Detailed discussion of this review and the cited papers preceded designing the content and shared links for a Google Form for data intake, Google Drive for article access, and Google Sheets linked to the form intake data. ResultsA total of 46 study entries were made by 2 team members, including 3 articles published since the review to demonstrate the ease of updating the system Various representations of the Google Forms intake data in the SENSOR System dashboard are presented. DiscussionVisual representation of review studies using a dashboard proved feasible and advantageous for numerous reasons. As the technology and guidelines for these systems evolve there is an opportunity to standardize reporting, centralize legacy datasets, streamline the submission process, improve collaboration between researchers, measure relative contribution of participating authors, and improve patient involvement. For the use case of clinical applications of psychedelics, limitations of conveying data accurately includes heterogeneity of study design, dosing, indications, and outcome measures. ConclusionCreation of a system for standardized data entry and dashboards for reviews of scientific studies is a feasible alternative and/or adjunct to the dissemination of summaries through traditional scientific review. There are numerous proposed advantages of the flexible, dynamic, and graphical display that requires further validation.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.