Accurate and Interpretable Postmenstrual Age Prediction via Multimodal Large Language Model

TL;DR

This paper presents a multimodal large language model fine-tuned for accurate and interpretable postmenstrual age prediction from neonatal MRI scans, combining high accuracy with clinical explanation generation.

Contribution

It introduces a parameter-efficient fine-tuning approach for a large language model to perform both precise PMA prediction and generate clinically relevant explanations from neonatal brain MRI data.

Findings

Achieved low prediction error with 95% CI of 0.78 to 1.52 weeks.

Enabled the model to generate interpretable, developmentally grounded explanations.

Demonstrated a method for combining regression and explanation tasks in a single model.

Abstract

Accurate estimation of postmenstrual age (PMA) at scan is crucial for assessing neonatal development and health. While deep learning models have achieved high accuracy in predicting PMA from brain MRI, they often function as black boxes, offering limited transparency and interpretability in clinical decision support. In this work, we address the dual challenge of accuracy and interpretability by adapting a multimodal large language model (MLLM) to perform both precise PMA prediction and clinically relevant explanation generation. We introduce a parameter-efficient fine-tuning (PEFT) strategy using instruction tuning and Low-Rank Adaptation (LoRA) applied to the Qwen2.5-VL-7B model. The model is trained on four 2D cortical surface projection maps derived from neonatal MRI scans. By employing distinct prompts for training and inference, our approach enables the MLLM to handle a regression task during training and generate clinically relevant explanations during inference. The fine-tuned model achieves a low prediction error with a 95 percent confidence interval of 0.78 to 1.52 weeks, while producing interpretable outputs grounded in developmental features, marking a significant step toward transparent and trustworthy AI systems in perinatal neuroscience.