Case Study: King's College London achieves 6.4x labeling efficiency and 97% automated labels with Encord

6x Faster Video Annotation for King's College London

King's College London faced prohibitively high costs using clinicians to annotate pre-cancerous polyp GI videos for large datasets. To address this, King's College London deployed Encord’s micro-model module to increase clinician efficiency and automate labeling.

Encord’s micro-model module automated 97% of produced labels and raised average labeling efficiency 6.4x, with the highest-cost clinician seeing a 16x improvement. Encord’s solution also cut model development time from one year to two months and accelerated time to AI in production (roughly 6x faster).

King's College London