• Select the area you would like to search.
  • ACTIVE INVESTIGATIONS Search for current projects using the investigator's name, institution, or keywords.
  • EXPERTS KNOWLEDGE BASE Enter keywords to search a list of questions and answers received and processed by the ADNI team.
  • ADNI PDFS Search any ADNI publication pdf by author, keyword, or PMID. Use an asterisk only to view all pdfs.
Principal Investigator  
Principal Investigator's Name: Bhavana Prakash
Institution: New York University
Department: Electrical and Computer Engineering
Proposed Analysis: I, with two other team members are trying to replicate the work of E. Jun, S. Jeong, D.-W. Heo, and H.-I. Suk titled, “Medical Transformer: Universal Brain Encoder for 3D MRI Analysis.” (arXiv: 2104.13633 [cs]). Our objective is to build a transfer learning network, namely Medical transformer, which uses a multi-view approach on sequence of 2D image slices to model a 3D image. The aim is to treat each slice along x, y, and z axes as a token and do masked token prediction as the pre-training regime to train 3 separate branches of the network, each for one of x, y, and z direction. The skeleton of the network contains a convolutional encoder, a transformer and a prediction network. The convolutional encoder and transformers are pre-trained in a self supervised learning manner on MRI data to finetune the network, and integrated with a prediction network. To perform Alzheimer’s disease (AD), mild cognitive impairment (MCI), and cognitively normal (CN) detection, we need T1-weighted structural MRI scans from ADNI dataset
Additional Investigators