mxochicale
commented
1 year ago Thanks @michellepi for pointed out the fetal plane classifier from Sendra-Balcells et al. 2023. Sounds like a great idea to have a classifier that can be used for different datasets (e.g. Spain common maternal-fetal, African countries datasets and our synthesised datasets). Feel free to make contributions.
Based on the model benchmark presented by Burgos et al.18 for fetal plane classification, the best performing method, a Densenet-16925, was implemented in this study.
Sendra-Balcells, Carla, Víctor M. Campello, Jordina Torrents-Barrena, Yahya Ali Ahmed, Mustafa Elattar, Benard Ohene-Botwe, Pempho Nyangulu et al. "Generalisability of fetal ultrasound deep learning models to low-resource imaging settings in five African countries." Scientific reports 13, no. 1 (2023): 2728.
If we go for config files, these are few pointers (but might be better options):
- https://medium.com/analytics-vidhya/how-to-write-configuration-files-in-your-machine-learning-project-47bc840acc19
- https://livebook.manning.com/concept/deep-learning/config-file
- https://www.borealisai.com/research-blogs/using-config-files-run-machine-learning-experiments/
- https://zindi.africa/learn/how-to-write-configuration-files-in-your-machine-learning-project
Raising this one to implement ResNet18 classifier based on the work of Montero et al. 2021. See "Supplementary Materials The following are available at https://www.mdpi.com/article/10.3390/s21237975/s1, Section S1: ResNet18 Fastai Architecture" : https://www.mdpi.com/1424-8220/21/23/7975
Description of Maternal fetal ultrasound planes from low-resource imaging settings in five African countries https://github.com/budai4medtech/xfetus/tree/main/data
Other relevant papers: