Charissa Poon, Petteri Teikari et al. (2023): "A dataset of rodent cerebrovasculature from in vivo multiphoton fluorescence microscopy imaging", Scientific Data 10, 141 doi: 10.1038/s41597-023-02048-8
Add some additional dataset to test the code for multiple datasets at the same time
You could make a pseudo dataset by inverting all the volumes for a classification ML test (task so easy to distinguish whether vessels are white and background black, or vice versa)
You could add some private dataset that would allow setting all the authentication properly to simulate better commercial R&D/production environment
Add some additional dataset to test the code for multiple datasets at the same time