Computer vision digitization of smartphone images of anesthesia paper health records from low-middle income countries

Table 1 Blood pressure YOLOv8 dataset

Set	Total images	Images with no annotations	Training instances
Training	18990	12799	31476
Validation	3870	2797	5244
Testing	2970	1842	5973

A breakdown of the dataset used to train the handwritten digit detection model. The dataset consisted of tiles from the larger images, so one marker on the larger image would appear in multiple tiles in varying locations. Test set error is not reported on this dataset, but instead on a 30 chart holdout set for more interpretable results

ISSN: 1471-2105