Combining heterogeneous data sources for neuroimaging based diagnosis: re-weighting and selecting what is important
- Submitting institution
-
University College London
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 14621
- Type
- D - Journal article
- DOI
-
10.1016/j.neuroimage.2019.01.053
- Title of journal
- NeuroImage
- Article number
- -
- First page
- 215
- Volume
- 195
- Issue
- -
- ISSN
- 1053-8119
- Open access status
- Compliant
- Month of publication
- March
- Year of publication
- 2019
- URL
-
-
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- Yes
- Number of additional authors
-
6
- Research group(s)
-
-
- Citation count
- 5
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- In this paper we proposed a novel methodology to combine and select features from different sources of information (e.g. neuroimaging, clinical and demographic information) in order to classify patients with mental health disorders versus healthy controls. The proposed approach can be applied to other clinically relevant classification tasks such as predicting future outcomes and/or treatment response by combining larger sources of patient information, therefore it has potential to contribute to precision medicine in general. We demonstrated the advantages of the proposed approach with respect to nine different alternative approaches using two different clinical datasets.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -