Extra Testing Data for paper "OC_Finder: A deep learning-based software for osteoclast segmentation, classification, and counting"

Here we have 9 datasets we used to validate OC_Finder's performance on various imaging settings.
The 9 datasets are inside the folder named "9 datasets for validation experiment".
Each dataset is composed of image files and csv files for the coordination of osteoclasts and non-osteoclasts that were manually labelled by human examiner.
csv files ending "_posi" has coordination of osteoclasts and "_nega" has coordination of non-osteoclasts.
Images in dataset #4, #5, #6, #7, #8, and #9 were resized so the scale of the images matched to the OC_Finder's training dataset.
Images in original size before resizing are also provided in "Original images before resizing".
Detailed capture setting and resizing information of images in each dataset can be found in "capture setting.xlsx".
The number of images in each dataset are as following:
#1: 18
#2: 18
#3: 18
#4: 36
#5: 36
#6: 36
#7: 16
#8: 16
#9: 16