There are many active research projects accessing and applying shared ADNI data. Use the search above to find specific research focuses on the active ADNI investigations. This information is requested annually as a requirement for data access.
Principal Investigator | |
Principal Investigator's Name: | Katarina Trojacanec |
Institution: | Faculty of Computer Science and Engineering |
Department: | Department of Software Engineering |
Country: | |
Proposed Analysis: | The aim of the research using ADNI image data set is investigation and performance evaluation of multiple query image retrieval in the field of medical imaging. The objectives of the research are: •To investigate the strategies used for early fusion of information extracted from the queries - different strategies for combining the feature vectors that represents the visual image content and/or including different weighing schemes during the combination; •To examine methods for late fusion of the results obtained from the separate queries - different methods for combining the results obtained from the retrieval based on the separate queries; •To investigate the combination of the strategies for early and late fusion; •To explore the structural unit of the query (the whole images or the regions of interest (ROIs) marked on the image) •To explore the number of examples in the query and its influence on the retrieval performance. The contributions of the proposed research will be in the direction of getting clinically more relevant answer in the retrieval process on the bases of the semantically more precise query. Additionally, the results from the research will show which of the analyzed strategies/algorithms (or mix of them) improves the retrieval performance in the field of medical images, taking into consideration their specific nature. Every investigation in the direction of the performance improvement of content based medical image retrieval and getting more precise, faster and more relevant answer means bringing more accurate information for relatively short period of time for the clinicians, from one point of view, and for the researchers and students, from the other point of view. |
Additional Investigators | |
Investigator's Name: | Suzana Loshkovska |
Proposed Analysis: | The aim of the research using ADNI image data set is investigation and performance evaluation of multiple query image retrieval in the field of medical imaging. The objectives of the research are: •To investigate the strategies used for early fusion of information extracted from the queries - different strategies for combining the feature vectors that represents the visual image content and/or including different weighing schemes during the combination; •To examine methods for late fusion of the results obtained from the separate queries - different methods for combining the results obtained from the retrieval based on the separate queries; •To investigate the combination of the strategies for early and late fusion; •To explore the structural unit of the query (the whole images or the regions of interest (ROIs) marked on the image) •To explore the number of examples in the query and its influence on the retrieval performance. The contributions of the proposed research will be in the direction of getting clinically more relevant answer in the retrieval process on the bases of the semantically more precise query. Additionally, the results from the research will show which of the analyzed strategies/algorithms (or mix of them) improves the retrieval performance in the field of medical images, taking into consideration their specific nature. Every investigation in the direction of the performance improvement of content based medical image retrieval and getting more precise, faster and more relevant answer means bringing more accurate information for relatively short period of time for the clinicians, from one point of view, and for the researchers and students, from the other point of view. |
Investigator's Name: | Ivica Dimitrovski |
Proposed Analysis: | The aim of the research using ADNI image data set is investigation and performance evaluation of multiple query image retrieval in the field of medical imaging. The objectives of the research are: •To investigate the strategies used for early fusion of information extracted from the queries - different strategies for combining the feature vectors that represents the visual image content and/or including different weighing schemes during the combination; •To examine methods for late fusion of the results obtained from the separate queries - different methods for combining the results obtained from the retrieval based on the separate queries; •To investigate the combination of the strategies for early and late fusion; •To explore the structural unit of the query (the whole images or the regions of interest (ROIs) marked on the image) •To explore the number of examples in the query and its influence on the retrieval performance. The contributions of the proposed research will be in the direction of getting clinically more relevant answer in the retrieval process on the bases of the semantically more precise query. Additionally, the results from the research will show which of the analyzed strategies/algorithms (or mix of them) improves the retrieval performance in the field of medical images, taking into consideration their specific nature. Every investigation in the direction of the performance improvement of content based medical image retrieval and getting more precise, faster and more relevant answer means bringing more accurate information for relatively short period of time for the clinicians, from one point of view, and for the researchers and students, from the other point of view. |
Investigator's Name: | Ivan Kitanovski |
Proposed Analysis: | The aim of the research using ADNI image data set is investigation and performance evaluation of multiple query image retrieval in the field of medical imaging. The objectives of the research are: •To investigate the strategies used for early fusion of information extracted from the queries - different strategies for combining the feature vectors that represents the visual image content and/or including different weighing schemes during the combination; •To examine methods for late fusion of the results obtained from the separate queries - different methods for combining the results obtained from the retrieval based on the separate queries; •To investigate the combination of the strategies for early and late fusion; •To explore the structural unit of the query (the whole images or the regions of interest (ROIs) marked on the image) •To explore the number of examples in the query and its influence on the retrieval performance. The contributions of the proposed research will be in the direction of getting clinically more relevant answer in the retrieval process on the bases of the semantically more precise query. Additionally, the results from the research will show which of the analyzed strategies/algorithms (or mix of them) improves the retrieval performance in the field of medical images, taking into consideration their specific nature. Every investigation in the direction of the performance improvement of content based medical image retrieval and getting more precise, faster and more relevant answer means bringing more accurate information for relatively short period of time for the clinicians, from one point of view, and for the researchers and students, from the other point of view. |
Investigator's Name: | Biljana Tojtovska |
Proposed Analysis: | The aim of the research using ADNI image data set is to enable efficient medical image organization and retrieval based on longitudinal data for Alzheimer’s disease. The objectives of the research are: •To perform statistical analysis; •To investigate and propose strategies for efficient representation of the information extracted from the images. Two directions will be considered: single image representation in a given moment of time, and representation of longitudinal information extracted from multiple images in a given time period; •To find the most appropriate similarity measurement for comparison of the query image/case and all images/cases in the database; •To explore and find the most appropriate classification technique. The contributions of the proposed research will be in the direction of overcoming the disadvantages and limitations of the existing algorithms and approaches used for image retrieval in the context of Alzheimer’s disease. Moreover, obtaining the information contained in the result of the retrieval is very important in terms of providing a second opinion in the diagnosis process, analysis of treatment response by monitoring the progress of the disease, both in clinical and research environment. |
Investigator's Name: | Slobodan Kalajdziski |
Proposed Analysis: | The aim of the research using ADNI image data set is to enable efficient medical image organization and retrieval based on longitudinal data for Alzheimer’s disease. The objectives of the research are: •Knowledge discovery; •To investigate and propose strategies for efficient representation of the information extracted from the images. Two directions will be considered: single image representation in a given moment of time, and representation of longitudinal information extracted from multiple images in a given time period; •To find the most appropriate similarity measurement for comparison of the query image/case and all images/cases in the database; •To explore and find the most appropriate classification technique. The contributions of the proposed research will be in the direction of overcoming the disadvantages and limitations of the existing algorithms and approaches used for image retrieval in the context of Alzheimer’s disease. Moreover, obtaining the information contained in the result of the retrieval is very important in terms of providing a second opinion in the diagnosis process, analysis of treatment response by monitoring the progress of the disease, both in clinical and research environment. |
Investigator's Name: | Jana Miceva |
Proposed Analysis: | The aim of this research is to enhance medical case retrieval for Alzheimer’s disease. The objectives of the research are: •To investigate longitudinal image representation strategies and their influence on longitudinal image retrieval. •Dealing with missing data and their influence on longitudinal image retrieval. •To investigate deep learning strategies in the context of longitudinal images for Alzheimer's disease •To investigate human brain connectomics in the context of Alzheimer’s Disease. The main contribution of the research will be towards improvement of the longitudinal image retrieval for Alzheimer’s disease. |
Investigator's Name: | Mateja Petkovska |
Proposed Analysis: | The aim of this research is to enhance medical case retrieval for Alzheimer’s disease. The objectives of the research are: •To investigate longitudinal image representation strategies and their influence on longitudinal image retrieval. •Dealing with missing data and their influence on longitudinal image retrieval. •To investigate deep learning strategies in the context of longitudinal images for Alzheimer's disease •To investigate human brain connectomics in the context of Alzheimer’s Disease. The main contribution of the research will be towards improvement of the longitudinal image retrieval for Alzheimer’s disease. |