-
|
Hi, My goal is to experiment with deep learning models to perform multi-class classification (CN, AD, pMCI, sMCI) on ADNI data. Your library seems perfect for this but I can't figure out how to do this.
I tried reading the docs but it's unclear for me how I can do this. I would greatly appreciate more details on how I can do this. Thanks! |
Beta Was this translation helpful? Give feedback.
Replies: 1 comment
-
|
Hello @fadyben98 , Thanks for your interest in ClinicaDL. Indeed ClinicaDL will provide you some great tools for multi-class classification on ADNI. Concerning which ADNI dataset you should use, you should maybe look at Clinica software. Especially the documentation of ADNI-2-BIDS converter. Moreover, Clinica's pipeline will provide you a TSV file with all the labels for each session (i.e. for each image). Once you have converted the ADNI dataset you need to BIDS format, you can run Clinica T1-linear pipeline to preprocess your data. And then only you can use ClinicaDL to perform your deep-learning study. Regards, Ravi |
Beta Was this translation helpful? Give feedback.
Hello @fadyben98 ,
Thanks for your interest in ClinicaDL. Indeed ClinicaDL will provide you some great tools for multi-class classification on ADNI.
Concerning which ADNI dataset you should use, you should maybe look at Clinica software. Especially the documentation of ADNI-2-BIDS converter.
Moreover, Clinica's pipeline will provide you a TSV file with all the labels for each session (i.e. for each image).
Once you have converted the ADNI dataset you need to BIDS format, you can run Clinica T1-linear pipeline to preprocess your data. And then only you can use ClinicaDL to perform your deep-learning study.
Regards,
Ravi