Challenge Overview
INTRODUCTION
This is the first in a series of challenges to develop a desktop web application to enable artificial intelligence aided dental diagnosis and treatment planning. We will be using IBM’s Watson to provide the computer learning aspect of the application.
USE CASE
Dental offices and individual practitioners who wish to increase diagnostic accuracy and create evidence based treatment plans. Staff should be able to upload radiographs, pocket depth probing scores, and high definition photographs for Watson to review and aid the dental professional in diagnosis. Once the diagnoses are established and confirmed, a treatment plan would be co-created with Watson.
REQUIREMENTS
For this challenge, you will be building a prototype web application with the MEAN (Mongo, Express, Angular, Node) stack. IBM Watson must be used for analyzing images and providing recommended diagnosis and treatments.
Your prototype must include basic functionality for:
-- Uploading an image to Watson
-- Receiving a diagnosis back from Watson
-- User confirming or rejecting the diagnosis and posting back to Watson for training
-- Creating a treatment plan based on the confirmed/corrected diagnosis
-- Saving data to MongoDB for redisplay on different screens
You need only provide basic styling (Bootstrap suggested) and client branding for the prototype. Colors should coincide with the client’s logo located in the challenge forums.
Main Screen
-- First screen displayed to users
-- Display a list of previously saved diagnosis and treatment plans in a table
-- Displaying by date created is OK for this challenge
-- Selecting an item displays the corresponding treatment plan(s)
-- Provide navigation for creating a new diagnosis (made prominent)
Diagnosis Creation Screen
-- Diagnosis begins on this screen
-- Provide the ability to upload multiple images to Watson
-- Display activity indicator while data is being processed
Diagnosis Confirmation Screen
-- Display results in a table (see page 13 in forums PDF)
-- Each tooth should have its own diagnosis with the following data:
-- Tooth #
-- Diagnosis
-- Confirmed (yes/no dropdown from user)
-- Corrected (free text if Confirmed is “no”)
-- Tx Plan (treatment)
-- There could be multiple treatment plans per tooth
-- User will confirm or correct each diagnosis
-- Results posted back to Watson for training
-- Results stored in MongoDB for redisplay
-- Display the treatment plan screen once all diagnosis have been confirmed or corrected
Treatment Plan Screen
-- Display results per tooth in a table (see page 14 in forums PDF)
-- Display treatment plan(s) for each tooth with the following data:
-- Tooth # and Treatment Option
-- Risks and Benefits
-- Cost
-- Years of Health
-- Cost per year (Cost / Years of Health)
-- Generally only one plan will be available per tooth, however there could be multiple
-- Results stored in MongoDB for redisplay
-- Navigate back to the main screen upon completion
RESOURCES
The following resources are located in the challenge forums:
-- PDF describing goals of the system
-- Client branding image
Final Submission Guidelines
-- Deployment guide describing how to run/install your prototype
-- Prototype must be deployable to an IBM Bluemix virtual server
-- Any sample data required to run your submission
-- Upload your source project and supporting files as a zip archive