×
  • 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: Christoph M. Friedrich
Institution: University of Applied Science and Arts Dortmund
Department: Computer Science
Country:
Proposed Analysis: i want to do further analysis of the ADNI data and unfortunately my old access from my former position at Fraunhofer Institute is no longer working. Therefore i reapply for access to the ADNI dataset to do further research on selection of representative subsets of SNPs with metaheuristic methods. The main idea is using metaheuristics from publications and using phenotypic data. I plan to additionally let one or two of my students, that plan to do their master thesis in "medical informatics" to work with the data and i will add them to the collaboraters list if they agree. Former successful work with the ADNI data that has been published and approved by the ADNI committee: G. Üstünkar, S. Özögür-Akyüz, G. W. Weber, C. M. Friedrich und Y. A. Son, „Selection of Representative SNP Sets for Genome-Wide Association Studies: A Metaheuristic Approach“, DOI:10.1007/s11590-011-0419-7, Optimization Letters, Volume 6(6), Seite 1207-1218, 2012. and can be accessed at: http://dx.doi.org/10.1007/s11590-011-0419-7 with best regards Christoph M. Friedrich
Additional Investigators  
Investigator's Name: Louise Bloch
Proposed Analysis: Louise Bloch will proceed with the work done by Sarah Pagliardini and will make it publishable. Her special focus will be more Early prediction of AD based on images and less on the demographics and test batteries, as we have a reasonable model from Sarah Pagliardini for this. The proposed methods are Deep Learning, RandomForests, XGBoost and GLMNet. My group is equipped with a NVIDIA DGX-1 machine, therefore we are sure to be able to process the amount of data with deep learning methods. AIBL data will be used for replication of results.
Investigator's Name: Obioma Pelka
Proposed Analysis: Obioma Pelka has developed a method to fuse image data with structured data for deep learning methods. This method has been developed with data from the Heinz-Nixdorf Recall Study in germany for the classification task of amnestic mid cognitive impairment. A paper has been submitted to the recent PloS One special collection on Alzheimers Disease, which is edited by Mr. Weiner. The Reviewers recommended that the results should be replicated on the ADNI dataset. Obioma Pelka needs access to the data and will together with Louise Bloch replicate her results on the ADNI dataset. The publication will be send for approval to the committee.