Abstract Details

Title Using Large-scale Contrastive Language-Image Pre-training to Maximize Brain MRI-Based Headache Classification

Topic Headache

Presentation(s) P4 - Poster Session 4 (5:00 PM-6:00 PM)

Poster/Presentation
Number 12-007

Objective

To optimize headache detection accuracy and biomarker extraction from relatively small brain MRI dataset.

Background

Deep learning methods showed promise for classifying headache disorders and extracting biomarkers from MRI data. Pre-trained models perform better for specialized tasks when fine-tuned on small, domain-specific datasets. Our model is fine-tuned on a pre-trained model to optimize headache classification and biomarker extraction.

Design/Methods We leveraged BioMedCLIP, a contrastive language-image model, pre-trained on PMC-15M (dataset of 15 million biomedical image-text pairs), to maximize headache classification. We fine-tuned BioMedCLIP using relatively small MRI dataset from 528 Healthy Controls (HC, including 424 from public IXI dataset), 96 participants with migraine (Mig), 49 with persistent post-traumatic headache (PPTH), and 48 with acute post-traumatic headache (APTH). All T1-weighted images were registered to MNI-152 1mm template. Six participants from each cohort were used for validation and testing, rest were used for fine-tuning. To enhance classification, we employed a novel evaluation method aggregating slice data and performing patient-level predictions using multi-instance learning, capturing the likelihood of disease manifestation from slices indicating headache biomarkers. For biomarker extraction, we utilized GradCAM, a deep learning explainability technique, to identify the brain regions significantly associated with each headache phenotype.

Results We evaluated three models on three classification tasks: HC vs. Mig, HC vs. APTH, and HC vs. PPTH, achieving accuracies of 91.67%, 83.33%, and 91.67% on held-out test sets, respectively. Key regions included superior frontal cortex, middle temporal white matter, and rostral middle frontal areas for Mig; inferior parietal, cerebellar cortex, and superior parietal cortex for APTH; and left cerebellar cortex, pars triangularis cortex, and middle temporal cortex for PPTH.

Conclusions Fine-tuning BioMedCLIP model on relatively small neuroimaging dataset maximizes headache classification and biomarker extraction. This approach provides a robust framework for classifying headache disorders and identifying relevant biomarkers from limited MRI data.

Authors/Disclosures
Fazle Rafsani PRESENTER	Mr. Rafsani has nothing to disclose.
Devam Sheth, MS	Mr. Sheth has nothing to disclose.
Yiming Che, PhD	Dr. Che has nothing to disclose.
Jay Shah, MD (Arizona State University)	Mr. Shah has nothing to disclose.
Md Mahfuzur Rahman Siddiquee (Arizona State University)	Md Mahfuzur Rahman Siddiquee has received personal compensation for serving as an employee of Meta.
Catherine D. Chong, PhD, FAAN	Dr. Chong has received personal compensation in the range of $500-$4,999 for serving on a Speakers Bureau for HCOP.
Simona Nikolova, PhD (FSU)	Dr. Nikolova has nothing to disclose.
Gina Dumkrieger, PhD (Mayo Clinic)	Gina Dumkrieger has received personal compensation for serving as an employee of Mayo Clinic. The institution of Gina Dumkrieger has received research support from NIH. The institution of Gina Dumkrieger has received research support from DOD. The institution of Gina Dumkrieger has received research support from AMGEN.
Baoxin Li (Arizona State University, School of Computing and Augmented Intelligence)	Baoxin Li has nothing to disclose.
Teresa Wu (Arizona State University)	Teresa Wu has nothing to disclose.
Todd J. Schwedt, MD, FAAN (Mayo Clinic)	Dr. Schwedt has received personal compensation in the range of $500-$4,999 for serving as a Consultant for Lundbeck. The institution of Dr. Schwedt has received personal compensation in the range of $10,000-$49,999 for serving as a Consultant for Abbvie. Dr. Schwedt has received personal compensation in the range of $500-$4,999 for serving as a Consultant for Salvia. Dr. Schwedt has or had stock in Nocira.Dr. Schwedt has or had stock in Allevalux. The institution of Dr. Schwedt has received research support from National Institutes of Health. The institution of Dr. Schwedt has received research support from United States Department of Defense. The institution of Dr. Schwedt has received research support from Patient Centered Outcomes Research Institute. The institution of Dr. Schwedt has received research support from Henry Jackson Foundation. The institution of Dr. Schwedt has received research support from Pfizer. The institution of Dr. Schwedt has received research support from National Headache Foundation. The institution of Dr. Schwedt has received research support from American Heart Association. The institution of Dr. Schwedt has received research support from AbbVie. The institution of Dr. Schwedt has received research support from Flinn Foundation. Dr. Schwedt has received intellectual property interests from a discovery or technology relating to health care. Dr. Schwedt has received intellectual property interests from a discovery or technology relating to health care. Dr. Schwedt has received intellectual property interests from a discovery or technology relating to health care. Dr. Schwedt has received intellectual property interests from a discovery or technology relating to health care. Dr. Schwedt has received publishing royalties from a publication relating to health care.

��ɫ��

��ɫ��