×
  • 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: Ali Abedi
Institution: Toronto Rehabilitation Institute
Department: KITE
Country:
Proposed Analysis: There is a large volume of healthcare-related data generated by hospitals and institutions all over the world. This valuable data can be used to train machine-learning and deep-learning models. In conventional machine-learning methods, all institutions send their data to a centralized server, and model training is performed centrally on the server. However, medical data is private in nature, and sharing this data with the server causes different legal, privacy, and data-ownership problems, especially between international institutions. In our project, we try to develop a new federated-learning framework for training our machine-learning models in a decentralized manner. In our scenario, institutions do not send their raw private data to the server. Each institution trains a model locally using its raw data. Then, all institutions send their locally-trained models to the server. Finally, the curator server aggregates all received models and create a global model that can be utilized by all institutions. Since the ADNI dataset is multi-institutional, it is ideal to be used in our experiments about developing a federated-learning algorithm.
Additional Investigators