@judywawira @ivange94 I am beginning this thread for all discussions concerning my Gsoc 2018 project; Implementing Reporting Workflow for Radiology as an Open Web App and Integrating Voice Dictation for Radiology.
Project: Implementing Reporting Workflow for Radiology as an Open Web App and Integrating Voice Dictation for Radiology
I have set up a blog using wordpress to document my activities during this period. It is not complete with the look and feel just yet, I will do that progressively. Here’s the link Ryan’s blog
I have been thinking about the components that will make up this OWA, I have a couple in mind already
A component for handling report templates, importing, display and its completion
A component for displaying all reports currently in the system, could display the reports in a tabular manner, with the ID and some metadata of a particular report making a row in the table
A component for viewing images corresponding to a particular order, and allowing the possibility to refer to these images in when creating a report for this order
Please do a mock up design and ask for feedback.
@ivange94 That is a quick mockup of what I was talking about.
- The Radiology template component allows importation of an MRRT template and occupation of that table with already imported MRRT templates.
- The Reports component displays all reports currently in the system. Would be displayed in a tabular manner similar to that of the report template. Each row would contain meta data about a particular report icluding its ID, the name of the radiologist who carried out the report and its status( Finalized, Draft, Preliminary, etc)
- The Study component would be used to display images from a particular order, and its status( whether it has been claimed by a radiologist who is currently producing the report)
- The Order component to get radiology orders
That was what I was trying to portray in the earlier post. Any thing I’m missing?
Can you look at what @Zatoner_Dev worked on , existing functionality for reporting and the user stories for Radiology ?
Then the sketch is not representative of all possible components
Yes I know it is little scanty and summarizes a lot of things. My aim for showing the sketch was just to portray a general idea i had in mind, to add a visual representation of the few basic components I wrote about, so that we could brain storm in depth on what to add and what to leave out, what sub components to include in the basic components i had pointed out , before I come up with a more detailed sketch/mock up
Speaking with my mentor @judywawira, we have concluded that Radiology should have just one functional OWA at the end of this GSOC session. That functional OWA would be a result of the project @lehone is working on. So we would have to figure out a common approach to this.
Also, given the fact that we are completely moving away from OpenMRS , we will have to figure out a way by which the templates will be stored against the spring data framework being developed by @yashdsaraf
We had a had with @sunbiz this evening and went back to two open web apps - your portion for reporting should be separate
We want you to use tinymce - https://www.tinymce.com in this one , you will have a voice button added that integrates with deep speech
so by midterms i should be able to use your own with voice and allowing for template selection
templates should be saved in a folder within the application - we didn’t feel like we need to save the html files into spring data
and then after midterm you will integrate the dictate study button to load your owa, allow dictation and save the report as a fhir diagnostic report into spring data
with what you know now , sketch out a better timetable to get this done
One or many OWAs, I dont think matters. They being in the same style, may be matters a little more. They being reusable and working easily with other webcomponents or other OWAs is absolutely essential.
@judywawira wrote above most of the things I wanted to say
Please edit anything i am missing @sunbiz – Ryan i should be able to use your own to dictate other notes in the EHR – lets say a clinic note … so think about this when developing and making your implementations
This is the final week of community bonding and I would like to make known a few decisions I made on how I intend to tackle this project.
First, two options for voice recognition software were given to my by @judywawira, from which I was to decide on which I would use for the project. The choices were Deep Voice and CMU Sphinx. I chose CMU Sphinx because it is actually a toolkit which contains a host of voice recognition packages, each of them suitable for use in different applications. The package I intend to use is Sphinx 4, because it is a java library and provides a java API that would be easy to implement in a java based project such as Librehealth radiology.
This will make implementation fairly easy because as any library in Java, all you need to do to use sphinx4 is to add the jars to the dependencies of your project and then you can write code using the API.
Inorder to limit the length of this post, I’d like to refer you to my project timeline from my GSOC proposal, detailing how I intend to achieve this. Additions and comments are highly welcomed. Lets make this last week of Community Bonding count! Heres the link to my GSOC Proposal
How will this fit in with the new architecture we are working on with spring data
I must confess I wasn’t totally aware of the scope of the Spring Data project. I had thought the project was mainly to change the database implementation in toolkit from hibernate to Spring Data, so I assumed it wouldn’t affect my work. I only learned recently that they are developing a completely new app based on Spring Data, and given the fact that they will be implementing just the standard fhir endpoints, which do not include some of the radiology REST endpoints I will need, I would have to implement them myself. But I intend to commence with work on the UI for the OWA first, and once I am done with that, which would not take long, then I would implement the REST endpoints I need, that is if indeed they are not being implemented in the Spring Data project, after which I would continue with my implementation as on my proposal
There is no clarity of the project and work to be done
Can you provide a step by step approach for your project
Saying you were unaware of spring data project despite active discussion on the GSOC threads either shows you are distracted or not interested
I expect better before the bonding time closes
@judywawira @ivange94 I just updated my blog with updates on my work so far. Here’s the link https://tenas97blog.wordpress.com/2018/05/28/gsoc-first-coding-phase-week-2/