SECONDGRAM: Self-conditioned diffusion with gradient manipulation for longitudinal MRI imputation
Summary: While individual MRI snapshots provide valuable insights, the longitudinal progression in repeated MRIs often holds more significant diagnostic and prognostic value. However, a scarcity of longitudinal datasets, comprising paired initial and follow-up scans, hinders the application of machi...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-05-01
|
| Series: | Patterns |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2666389925000601 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849729195394990080 |
|---|---|
| author | Brandon Theodorou Anant Dadu Mike Nalls Faraz Faghri Jimeng Sun |
| author_facet | Brandon Theodorou Anant Dadu Mike Nalls Faraz Faghri Jimeng Sun |
| author_sort | Brandon Theodorou |
| collection | DOAJ |
| description | Summary: While individual MRI snapshots provide valuable insights, the longitudinal progression in repeated MRIs often holds more significant diagnostic and prognostic value. However, a scarcity of longitudinal datasets, comprising paired initial and follow-up scans, hinders the application of machine learning for crucial sequential tasks. We address this gap by proposing self-conditioned diffusion with gradient manipulation (SECONDGRAM) to generate absent follow-up imaging features, enabling predictions of MRI developments over time and enriching limited datasets through imputation. SECONDGRAM builds on neural diffusion models and introduces two key contributions: self-conditioned learning to leverage much larger, unlinked datasets and gradient manipulation to combat instability and overfitting in a low-data setting. We evaluate SECONDGRAM on the UK Biobank dataset and show that it not only models MRI patterns better than existing baselines but also enhances training datasets to achieve better downstream results over naive approaches. The bigger picture: Longitudinal MRI is a type of medical imaging study where MRI scans are taken several times over a period of time to track changes. These studies play a pivotal role in understanding neurodegenerative diseases like Alzheimer’s disease and Parkinson’s disease, yet the limited availability of paired longitudinal imaging datasets significantly restricts advanced machine learning applications. Our research introduces SECONDGRAM, a robust framework utilizing neural diffusion models enhanced with self-conditioned learning and gradient manipulation techniques. These models are computer algorithms that simulate how information changes. SECONDGRAM addresses data scarcity by generating realistic follow-up MRI imaging features, thereby enriching limited datasets. This methodological advancement not only improves the accuracy and realism of imputed imaging features but also significantly boosts the performance of predictive models in critical downstream tasks. The implications extend beyond healthcare, presenting a versatile solution for data augmentation in numerous fields struggling with longitudinal data constraints, thus enhancing decision-making processes in precision medicine. |
| format | Article |
| id | doaj-art-ba43ef7b6f3642a180b1bfb5487e67c4 |
| institution | DOAJ |
| issn | 2666-3899 |
| language | English |
| publishDate | 2025-05-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Patterns |
| spelling | doaj-art-ba43ef7b6f3642a180b1bfb5487e67c42025-08-20T03:09:17ZengElsevierPatterns2666-38992025-05-016510121210.1016/j.patter.2025.101212SECONDGRAM: Self-conditioned diffusion with gradient manipulation for longitudinal MRI imputationBrandon Theodorou0Anant Dadu1Mike Nalls2Faraz Faghri3Jimeng Sun4Department of Computer Science, University of Illinois at Urbana-Champaign, Urbana, IL, USA; Center for Alzheimer’s and Related Dementias, National Institutes of Health, Bethesda, MD, USACenter for Alzheimer’s and Related Dementias, National Institutes of Health, Bethesda, MD, USA; Data Tecnica International, LLC, Washington, DC, USACenter for Alzheimer’s and Related Dementias, National Institutes of Health, Bethesda, MD, USA; Laboratory of Neurogenetics, National Institute on Aging, National Institutes of Health, Bethesda, MD, USA; Data Tecnica International, LLC, Washington, DC, USACenter for Alzheimer’s and Related Dementias, National Institutes of Health, Bethesda, MD, USA; Laboratory of Neurogenetics, National Institute on Aging, National Institutes of Health, Bethesda, MD, USA; Data Tecnica International, LLC, Washington, DC, USA; Corresponding authorDepartment of Computer Science, University of Illinois at Urbana-Champaign, Urbana, IL, USA; Corresponding authorSummary: While individual MRI snapshots provide valuable insights, the longitudinal progression in repeated MRIs often holds more significant diagnostic and prognostic value. However, a scarcity of longitudinal datasets, comprising paired initial and follow-up scans, hinders the application of machine learning for crucial sequential tasks. We address this gap by proposing self-conditioned diffusion with gradient manipulation (SECONDGRAM) to generate absent follow-up imaging features, enabling predictions of MRI developments over time and enriching limited datasets through imputation. SECONDGRAM builds on neural diffusion models and introduces two key contributions: self-conditioned learning to leverage much larger, unlinked datasets and gradient manipulation to combat instability and overfitting in a low-data setting. We evaluate SECONDGRAM on the UK Biobank dataset and show that it not only models MRI patterns better than existing baselines but also enhances training datasets to achieve better downstream results over naive approaches. The bigger picture: Longitudinal MRI is a type of medical imaging study where MRI scans are taken several times over a period of time to track changes. These studies play a pivotal role in understanding neurodegenerative diseases like Alzheimer’s disease and Parkinson’s disease, yet the limited availability of paired longitudinal imaging datasets significantly restricts advanced machine learning applications. Our research introduces SECONDGRAM, a robust framework utilizing neural diffusion models enhanced with self-conditioned learning and gradient manipulation techniques. These models are computer algorithms that simulate how information changes. SECONDGRAM addresses data scarcity by generating realistic follow-up MRI imaging features, thereby enriching limited datasets. This methodological advancement not only improves the accuracy and realism of imputed imaging features but also significantly boosts the performance of predictive models in critical downstream tasks. The implications extend beyond healthcare, presenting a versatile solution for data augmentation in numerous fields struggling with longitudinal data constraints, thus enhancing decision-making processes in precision medicine.http://www.sciencedirect.com/science/article/pii/S2666389925000601machine learning in healthcaregenerative modelingdiffusion modelsimputationaugmentationlongitudinal MRI |
| spellingShingle | Brandon Theodorou Anant Dadu Mike Nalls Faraz Faghri Jimeng Sun SECONDGRAM: Self-conditioned diffusion with gradient manipulation for longitudinal MRI imputation Patterns machine learning in healthcare generative modeling diffusion models imputation augmentation longitudinal MRI |
| title | SECONDGRAM: Self-conditioned diffusion with gradient manipulation for longitudinal MRI imputation |
| title_full | SECONDGRAM: Self-conditioned diffusion with gradient manipulation for longitudinal MRI imputation |
| title_fullStr | SECONDGRAM: Self-conditioned diffusion with gradient manipulation for longitudinal MRI imputation |
| title_full_unstemmed | SECONDGRAM: Self-conditioned diffusion with gradient manipulation for longitudinal MRI imputation |
| title_short | SECONDGRAM: Self-conditioned diffusion with gradient manipulation for longitudinal MRI imputation |
| title_sort | secondgram self conditioned diffusion with gradient manipulation for longitudinal mri imputation |
| topic | machine learning in healthcare generative modeling diffusion models imputation augmentation longitudinal MRI |
| url | http://www.sciencedirect.com/science/article/pii/S2666389925000601 |
| work_keys_str_mv | AT brandontheodorou secondgramselfconditioneddiffusionwithgradientmanipulationforlongitudinalmriimputation AT anantdadu secondgramselfconditioneddiffusionwithgradientmanipulationforlongitudinalmriimputation AT mikenalls secondgramselfconditioneddiffusionwithgradientmanipulationforlongitudinalmriimputation AT farazfaghri secondgramselfconditioneddiffusionwithgradientmanipulationforlongitudinalmriimputation AT jimengsun secondgramselfconditioneddiffusionwithgradientmanipulationforlongitudinalmriimputation |