Why Workflow Matters
These days, deep Learning research in medical imaging is booming with improved models being developed day by day to enable Al-assisted diagnosis. However, most of this Al research is being done in isolation and with limited datasets which may lead to "overly simplified models". Apart from that, even when a fully validated model is available, it is still a challenge to deploy the algorithm in a local environment with existing settings because creating a new specialized workflow for Al is probably expensive and compartmentalizes the workflow and breaks its smoothness.