Clinical Prediction from Structural Brain MRI Scans: A Large-Scale Empirical Study
Multivariate pattern analysis (MVPA) methods have become an important tool in neuroimaging, revealing complex associations and yielding powerful prediction models. Despite methodological developments and novel application domains, there has been little effort to compile benchmark results that resear...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | English |
| Published: |
Springer US
2016
|
| Online Access: | http://hdl.handle.net/1721.1/105865 https://orcid.org/0000-0002-5002-1227 |
| _version_ | 1826197298786336768 |
|---|---|
| author | Konukoglu, Ender Sabuncu, Mert R |
| author2 | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory |
| author_facet | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory Konukoglu, Ender Sabuncu, Mert R |
| author_sort | Konukoglu, Ender |
| collection | MIT |
| description | Multivariate pattern analysis (MVPA) methods have become an important tool in neuroimaging, revealing complex associations and yielding powerful prediction models. Despite methodological developments and novel application domains, there has been little effort to compile benchmark results that researchers can reference and compare against. This study takes a significant step in this direction. We employed three classes of state-of-the-art MVPA algorithms and common types of structural measurements from brain Magnetic Resonance Imaging (MRI) scans to predict an array of clinically relevant variables (diagnosis of Alzheimer’s, schizophrenia, autism, and attention deficit and hyperactivity disorder; age, cerebrospinal fluid derived amyloid-β levels and mini-mental state exam score). We analyzed data from over 2,800 subjects, compiled from six publicly available datasets. The employed data and computational tools are freely distributed (https://www.nmr.mgh.harvard.edu/lab/mripredict), making this the largest, most comprehensive, reproducible benchmark image-based prediction experiment to date in structural neuroimaging. Finally, we make several observations regarding the factors that influence prediction performance and point to future research directions. Unsurprisingly, our results suggest that the biological footprint (effect size) has a dramatic influence on prediction performance. Though the choice of image measurement and MVPA algorithm can impact the result, there was no universally optimal selection. Intriguingly, the choice of algorithm seemed to be less critical than the choice of measurement type. Finally, our results showed that cross-validation estimates of performance, while generally optimistic, correlate well with generalization accuracy on a new dataset. |
| first_indexed | 2024-09-23T10:45:34Z |
| format | Article |
| id | mit-1721.1/105865 |
| institution | Massachusetts Institute of Technology |
| language | English |
| last_indexed | 2024-09-23T10:45:34Z |
| publishDate | 2016 |
| publisher | Springer US |
| record_format | dspace |
| spelling | mit-1721.1/1058652022-09-30T22:50:41Z Clinical Prediction from Structural Brain MRI Scans: A Large-Scale Empirical Study Konukoglu, Ender Sabuncu, Mert R Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory Sabuncu, Mert R Multivariate pattern analysis (MVPA) methods have become an important tool in neuroimaging, revealing complex associations and yielding powerful prediction models. Despite methodological developments and novel application domains, there has been little effort to compile benchmark results that researchers can reference and compare against. This study takes a significant step in this direction. We employed three classes of state-of-the-art MVPA algorithms and common types of structural measurements from brain Magnetic Resonance Imaging (MRI) scans to predict an array of clinically relevant variables (diagnosis of Alzheimer’s, schizophrenia, autism, and attention deficit and hyperactivity disorder; age, cerebrospinal fluid derived amyloid-β levels and mini-mental state exam score). We analyzed data from over 2,800 subjects, compiled from six publicly available datasets. The employed data and computational tools are freely distributed (https://www.nmr.mgh.harvard.edu/lab/mripredict), making this the largest, most comprehensive, reproducible benchmark image-based prediction experiment to date in structural neuroimaging. Finally, we make several observations regarding the factors that influence prediction performance and point to future research directions. Unsurprisingly, our results suggest that the biological footprint (effect size) has a dramatic influence on prediction performance. Though the choice of image measurement and MVPA algorithm can impact the result, there was no universally optimal selection. Intriguingly, the choice of algorithm seemed to be less critical than the choice of measurement type. Finally, our results showed that cross-validation estimates of performance, while generally optimistic, correlate well with generalization accuracy on a new dataset. BrightFocus Foundation (Alzheimer’s Disease pilot grant (AHAF A2012333)) National Institutes of Health (U.S.) (K25 grant (NIBIB 1K25EB013649-01)) National Center for Research Resources (U.S.) (U24 RR021382) National Institutes of Health. National Institute for Biomedical Imaging and Bioengineering (R01EB006758) National Institute of Neurological Disorders and Stroke (U.S.) (R01 NS052585-01, 1R21NS072652-01, 1R01NS070963, R01NS083534) National Institutes of Health (U.S.) (Blueprint for Neuroscience Research (5U01-MH093765)) 2016-12-16T23:44:38Z 2016-12-16T23:44:38Z 2014-07 2016-08-18T15:45:47Z Article http://purl.org/eprint/type/JournalArticle 1539-2791 1559-0089 http://hdl.handle.net/1721.1/105865 Sabuncu, Mert R., and Ender Konukoglu. “Clinical Prediction from Structural Brain MRI Scans: A Large-Scale Empirical Study.” Neuroinformatics 13, no. 1 (July 22, 2014): 31–46. doi:10.1007/s12021-014-9238-1. https://orcid.org/0000-0002-5002-1227 en http://dx.doi.org/10.1007/s12021-014-9238-1 Neuroinformatics Article is made available in accordance with the publisher's policy and may be subject to US copyright law. Please refer to the publisher's site for terms of use. Springer Science+Business Media New York application/pdf Springer US Springer US |
| spellingShingle | Konukoglu, Ender Sabuncu, Mert R Clinical Prediction from Structural Brain MRI Scans: A Large-Scale Empirical Study |
| title | Clinical Prediction from Structural Brain MRI Scans: A Large-Scale Empirical Study |
| title_full | Clinical Prediction from Structural Brain MRI Scans: A Large-Scale Empirical Study |
| title_fullStr | Clinical Prediction from Structural Brain MRI Scans: A Large-Scale Empirical Study |
| title_full_unstemmed | Clinical Prediction from Structural Brain MRI Scans: A Large-Scale Empirical Study |
| title_short | Clinical Prediction from Structural Brain MRI Scans: A Large-Scale Empirical Study |
| title_sort | clinical prediction from structural brain mri scans a large scale empirical study |
| url | http://hdl.handle.net/1721.1/105865 https://orcid.org/0000-0002-5002-1227 |
| work_keys_str_mv | AT konukogluender clinicalpredictionfromstructuralbrainmriscansalargescaleempiricalstudy AT sabuncumertr clinicalpredictionfromstructuralbrainmriscansalargescaleempiricalstudy |