Cross-Modal Fine-Tuning of 3D Convolutional Foundation Models for ADHD Classification with Low-Rank Adaptation

TL;DR

This paper introduces a parameter-efficient transfer learning method using Low-Rank Adaptation to fine-tune 3D convolutional models for MRI-based ADHD classification, achieving state-of-the-art results with significantly fewer trainable parameters.

Contribution

It presents a novel LoRA-based fine-tuning approach for 3D convolutional models, enabling effective cross-modal adaptation from CT to MRI for ADHD diagnosis.

Findings

Achieved 71.9% accuracy in ADHD classification

Attained an AUC of 0.716 with minimal trainable parameters

First successful cross-modal adaptation of foundation models in neuroimaging

Abstract

Early diagnosis of attention-deficit/hyperactivity disorder (ADHD) in children plays a crucial role in improving outcomes in education and mental health. Diagnosing ADHD using neuroimaging data, however, remains challenging due to heterogeneous presentations and overlapping symptoms with other conditions. To address this, we propose a novel parameter-efficient transfer learning approach that adapts a large-scale 3D convolutional foundation model, pre-trained on CT images, to an MRI-based ADHD classification task. Our method introduces Low-Rank Adaptation (LoRA) in 3D by factorizing 3D convolutional kernels into 2D low-rank updates, dramatically reducing trainable parameters while achieving superior performance. In a five-fold cross-validated evaluation on a public diffusion MRI database, our 3D LoRA fine-tuning strategy achieved state-of-the-art results, with one model variant reaching 71.9% accuracy and another attaining an AUC of 0.716. Both variants use only 1.64 million trainable parameters (over 113x fewer than a fully fine-tuned foundation model). Our results represent one of the first successful cross-modal (CT-to-MRI) adaptations of a foundation model in neuroimaging, establishing a new benchmark for ADHD classification while greatly improving efficiency.