Difference between revisions of "AI-Assisted Quality Control of CTD Data"
Jump to navigation
Jump to search
| Line 4: | Line 4: | ||
== Use Case Objectives == | == Use Case Objectives == | ||
| − | + | * Machine Learning Task: Flag in advance the scans to be deleted during CTD quality control | |
| − | + | * Business Value: Flagged scans allow the analyst to quickly focus attention on crucial areas, reducing the time and effort required to delete scans | |
| − | + | * Measures of Success: | |
| − | + | ** Accuracy of model predictions | |
| − | + | **Client feedback on quality control speed-ups | |
| − | * Aspirational Goals: | + | *Aspirational Goals: |
| − | + | ** Mitigation of uncertainty in human decisions | |
| − | + | ** Semi or full automation of scan deletions | |
| Line 17: | Line 17: | ||
[[File:Business process.png|Flow diagram for the model integration into the business process.]] | [[File:Business process.png|Flow diagram for the model integration into the business process.]] | ||
| − | == Machine Learning Pipeline == | + | ==Machine Learning Pipeline== |
[[File:Ml pipeline.png|Three-step process used in the machine learning pipeline.]] | [[File:Ml pipeline.png|Three-step process used in the machine learning pipeline.]] | ||
| − | == Experimental Model Performance == | + | ==Experimental Model Performance== |
[[File:Performance.png|Model performance and dataset histogram over the depth range from which CTD scans are collected.]] | [[File:Performance.png|Model performance and dataset histogram over the depth range from which CTD scans are collected.]] | ||
| − | == Model Deployment and Integration == | + | ==Model Deployment and Integration== |
[[File:Model communications.png|Information flow in the integration of the model deployment into the business process.]] | [[File:Model communications.png|Information flow in the integration of the model deployment into the business process.]] | ||
| − | == Next Steps == | + | ==Next Steps == |
Revision as of 11:09, 22 December 2022
Use Case Objectives
- Machine Learning Task: Flag in advance the scans to be deleted during CTD quality control
- Business Value: Flagged scans allow the analyst to quickly focus attention on crucial areas, reducing the time and effort required to delete scans
- Measures of Success:
- Accuracy of model predictions
- Client feedback on quality control speed-ups
- Aspirational Goals:
- Mitigation of uncertainty in human decisions
- Semi or full automation of scan deletions
Machine Learning Pipeline
Experimental Model Performance
Model Deployment and Integration
