Skip to content

Latest commit

 

History

History
117 lines (93 loc) · 4.24 KB

README.md

File metadata and controls

117 lines (93 loc) · 4.24 KB
Raw

Uncharted

Web based medical graph application

Uncharted plans to use Semantic Versioning. Everything is scoped under the uncharted un namespace.

Live here

Purpose

To replace inpatient hospital paper-based charts

Such charts include:

  • Vital Signs
  • Fluid Balance
  • drain output
  • urine output
  • fluids
  • Blood Sugar

Uncharted brings these benefits:

  • cleaner graphs which are easier to interpret
  • automatic calculations, totals, and early warning scores
  • access on any device anywhere

MVP guidelines:

  • login screen
  • user patient list:
  • add new patient by NHI
  • remove patient
  • patient screen:
  • load patient data
  • patient header bar
  • change to another patient
  • chart list
  • select another chart
  • add a new chart

MVP Charts:

  • vital signs

  • HR, BP, Temp, O2 sats and delivery, RR

  • input and output

  • IV and oral intake, NG feeds

  • urine output

  • drains

  • type and volume of output

  • fluid balance

  • totals in and out, and the difference across days

  • weight

  • blood sugar

  • fingerprick sugar level

  • insulin type & dose

Step-by-step plan:

  • set up git repository
  • write step-by-step plan
  • install a bootstrap template page, such as the 'dashboard' example
  • split the page into a headerBar, chartList, and chartArea
  • make the chart list and chart Area scroll independently of each other
  • populate the chart list with a few buttons which respond to clicks by changing the chartArea
  • put this example chart into one of the chart views
  • it should disappear if another chart is selected, and reload when re-selected.
  • make it scroll and scale appropriately if the page is panned or resized
  • duplicate the example chart

Chart Stream:

  • Make one graph of this chart have bars for blood pressure

  • change the other into a line graph for heart rate

  • replace the d3fc example data with some more realistic dummy clinical measurements

  • display d3fc line annotations to give better context to the data

  • also display the value labels for the maximum, minimum, and most recent data points

  • make 'unpinned' graphs disappear when there is no data in-view.

  • separate the x axis into its own separate 'graph'

  • take the team out for dinner

  • write more step-by-step instructions, the next focus is data entry

  • Enable data entry for vital signs, then send the entries to the server

  • ...
  • Develop a fluid chart with urine output, IV and oral intake etc..
  • ...
  • Display early warning scores (EWS) and calculate them when new measurements are added
  • Add respiratory vitals and temperature
  • Look back through Jack's old paper notes and see if there is anything important missing from the vitals signs chart
  • Finish the vitals signs MVP chart!
  • now make it look pretty, and work on small touch screens

Client Stream:

  • install a login page (with trivial authentication)
  • fabricate some example patient demographic data for an example user patient list
  • make a user patient list page, possibly using a suitable bootstrap template
  • clicking on a patient sends you to the chart viewing page (data irrespective of the patient)
  • the selected patient's name, age, sex, and NHI appear in the headerBar
  • provide a way to get back to the patient list
  • in the patient list, implement a way to add patients (if you know their NHI)
  • and remove patients from the list
  • make the list of charts (in patient view) shrink to a single icon column on small screens, possibly like this bootstrap example

Server Stream:

  • implement a real login/authentication scheme

  • write an API for user and patient data exchange

  • implement the API on the server and client, sending dummy data

  • implement some kind of database or the like for the data

  • retrieve different data for a couple of example patients and users

  • accept new measurements entered via the client, save to the database

  • push new measurements to other clients viewing that patient

  • should you be using Grunt, or CodeKit?