×
  • 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: Sara Colantonio
Institution: National Research Council of Italy
Department: Institute of Information Science and Technologies
Country:
Proposed Analysis: The purpose of our research is to investigate the feasibility of automatic AI-driven classification of AD and MCI with respect of normal control. One of the major issues of AI models applied to medical images is their black-box nature and therefore the possible lack of trust by physicians. Recent advances in the literature suggest that explainable models are available and could be adopted to overcome this limitation. Moreover, another frequently arising problem in this field is the performance drop due to dataset-shift. This can arise when trained models are deployed in different institutions possibly with different scanners. We plan to carry out our analysis by means of Convolutional Neural Networks (CNNs), but by leveraging the explaining capabilities featured by the ProtoPNet architecture. In addition, a modified version of ProtoPNet will be designed and implemented to address the dataset-shift issue and to provide the user with causality-based explanations.
Additional Investigators  
Investigator's Name: Gianluca Carloni
Proposed Analysis: Ph.D. student at Pisa University and research fellow at ISTI-CNR. The analysis will be the same as for the principal investigator.
Investigator's Name: Andrea Berti
Proposed Analysis: Ph.D. student at Pisa University and research fellow at ISTI-CNR. The analysis will be the same as for the principal investigator.