This project is aimed for an automatic detection and evaluation of urine dipsticks from images. Steps performed during the analysis:
- Detection of the stick: Feature matching & Mask R-CNN
- Detection of the reference card: Feature matching
- Localisation of the single test and reference fields
- Color analysis and comparison
The main scripts to perform the detection and colour evaluation can be found in the urinalysis folder. All scripts that support the main scripts can be found in the helper subfolder, scripts analyzing the calculated results in the evaluation subfolder. The used implementation of Mask R-CNN is in the mask-rcnn subfolder.
Required python libraries can be found in requirements.txt
M. Flaucher et al., "Smartphone-Based Colorimetric Analysis of Urine Test Strips for At-Home Prenatal Care," in IEEE Journal of Translational Engineering in Health and Medicine, vol. 10, pp. 1-9, 2022, Art no. 2800109, doi: 10.1109/JTEHM.2022.3179147.
- Madeleine Flaucher - Initial work
- Matterport Inc. - Mask R-CNN implementation