30:T44a,html`

Data

We'll use 2015 hospital inpatient discharges data with Lung Cancer diagnosis from the four counties of New Your City, plus the two additional counties to the east, in Long Island. To illustrate the extend to which public and private data can be analysed together when OAUTH2 governed API ecosystems are used, and the way this can be achieved without compromise of the privacy of the latter, we'll treat two of the counties as public data, Queens and Manhattan, and the remainder as private data. Of course in reality all data is already available publicly and deidentified, and can be obtaied directly from New York State Department of Health at health.data.ny.gov/resource/82xm-y6g8. In fact, as detailed in (Almeida 2019), and demonstrated below, populating the variable lungNY2015, all 9,261 reccords for 2015's inpatiant reccords for Lung Cancer Diagnosis (ccs_diagnosis_code=19) can be readily obtained by composing the API call with the corresponsing parameter.

`31:[] 2e:{"title":"paper","cells":[{"id":0,"value":"md`# paper`","pinned":false,"hidden":false,"mode":"ojs"},{"id":5,"value":"html`\n

Suplementary paper to EpiSphere, enabling user-governed Epidemiology data ecosystems for a new generation of large scale cohort studies.

EpiSphere: API ecosystems for real-time Epidemiology

Jonas S Almeida, Praful, Montse

\n`\n ","pinned":false,"hidden":false,"mode":"ojs"},{"id":45,"value":"html`\n

Abstract

This live paper is supplementary to the report under peer review on user-governed analysis of epidemiology data. The argument that well governed interoperability is the most effective mechanism to promote semantic convergence and analytical reproducibility requires daemonstration. Accordingly, this suplementary live paper was conceived to provide that verification. It proposes an exercise where the reader separates datasets on lung cancer inpatients in the state of New York into two groups. One group will be treated as public data and the remainder will be kept in a private folder in Box.com, a popular user-facing platform to manage and share sensitive health data. The test will be to show that the same report can be assembled in real time for all data sets that a reader has access to, whichever they may be - i.e. data access and analytics without any gate beyong authentication.

\n`","pinned":false,"hidden":false,"mode":"ojs"},{"id":10,"value":"html`\n

Methods

\n\n\n- \n`","pinned":false,"hidden":false,"mode":"ojs"},{"id":154,"value":"style_paper=html(`\n`)","pinned":false,"hidden":false,"mode":"ojs"},{"id":58,"value":"$30","pinned":false,"hidden":false,"mode":"ojs"},{"id":81,"value":"lungCancerNY2019 = d3.json('https://health.data.ny.gov/resource/82xm-y6g8.json?ccs_diagnosis_code=19&$limit=10000')","pinned":true,"hidden":false,"mode":"ojs"},{"id":110,"value":"html`\n

Figure 1 - all deidentified inpatient data for Lung Cancer diagnosis in the state of New York is available in the public dommain. It is hosted by the state's Dept of health at health.data.ny.gov/resource/82xm-y6g8.json?ccs_diagnosis_code=19. Click on arrow heads to open and traverse data structure. This data is loaded in real time, if reading device connected wi the web, when this document is opened.

\n`","pinned":false,"hidden":false,"mode":"ojs"},{"id":150,"value":"html`\n

Graphic representation

\n`","pinned":false,"hidden":false,"mode":"ojs"},{"id":147,"value":"plotly= require(\"https://cdn.plot.ly/plotly-latest.min.js\")","pinned":false,"hidden":false,"mode":"ojs"},{"id":131,"value":"plotIt=async function(trace,layout){\n if(typeof(plotly)==\"undefined\"){\n let plotly= await require(\"https://cdn.plot.ly/plotly-latest.min.js\")\n }\n let div = DOM.element('div')\n trace=trace||false\n return plotly.plot(div,trace,layout)\n}","pinned":false,"hidden":false,"mode":"ojs"},{"id":92,"value":"html`\n

Computation

All computational methods used here observe the FAIR criteria (Wilkinson 2018) in its narrowest interpretation of being about reproducible dereferenceability and comunication of both data and analysis. Specifically, this implies analytical solutions reproducible in a regular web browser running in a mobile device (the dominant digital platform), without any requirements of downloading data or installing applications.

\n`","pinned":false,"hidden":false,"mode":"ojs"},{"id":83,"value":"d3=require('d3')","pinned":true,"hidden":false,"mode":"ojs"},{"id":62,"value":"html`\n

Results

\n`","pinned":false,"hidden":false,"mode":"ojs"}],"files":"$W31","stdlib":"1","titling":"auto"}