Impact of Clinical Image Quality on Efficient Foundation Model Finetuning

2025-08-15

AI Summary

Clinical image quality distribution significantly impacts the label efficiency and generalization of medical vision foundation models during fine-tuning. Method: Building upon ProFound—a prostate MRI–specific foundation model—we systematically vary the proportions of high- and low-quality images in both fine-tuning and test sets to quantify their effects on label-efficient fine-tuning and downstream task performance. Contribution/Results: Matching image quality distributions between fine-tuning and test sets substantially improves label efficiency. Sufficient high-quality samples are essential to maintain model performance; when scarce, pretrained models underperform even random initialization. Downstream tasks exhibit heterogeneous sensitivity to distributional shifts. This work establishes, for the first time, the critical role of clinical image quality distribution alignment in adapting medical foundation models—providing theoretical foundations and practical guidance for standardized acquisition and annotation protocols in medical imaging AI.

Technology Category

• Computer Vision: Other Foundations of Computer Vision
• Machine Learning: Other Foundations of Machine Learning
• Search and Optimization: Other Foundations of Search & Optimization

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Security and Privacy: Data transparency and provenance

Abstract

Foundation models in medical imaging have shown promising label efficiency, achieving high downstream performance with only a fraction of annotated data. Here, we evaluate this in prostate multiparametric MRI using ProFound, a domain-specific vision foundation model pretrained on large-scale prostate MRI datasets. We investigate how variable image quality affects label-efficient finetuning by measuring the generalisability of finetuned models. Experiments systematically vary high-/low-quality image ratios in finetuning and evaluation sets. Our findings indicate that image quality distribution and its finetune-and-test mismatch significantly affect model performance. In particular: a) Varying the ratio of high- to low-quality images between finetuning and test sets leads to notable differences in downstream performance; and b) The presence of sufficient high-quality images in the finetuning set is critical for maintaining strong performance, whilst the importance of matched finetuning and testing distribution varies between different downstream tasks, such as automated radiology reporting and prostate cancer detection.When quality ratios are consistent, finetuning needs far less labeled data than training from scratch, but label efficiency depends on image quality distribution. Without enough high-quality finetuning data, pretrained models may fail to outperform those trained without pretraining. This highlights the importance of assessing and aligning quality distributions between finetuning and deployment, and the need for quality standards in finetuning data for specific downstream tasks. Using ProFound, we show the value of quantifying image quality in both finetuning and deployment to fully realise the data and compute efficiency benefits of foundation models.

Problem

• Evaluating how clinical image quality affects label-efficient finetuning of foundation models
• Assessing impact of quality mismatch between finetuning and test sets on performance
• Determining sufficient high-quality image requirements for specific downstream tasks

Innovation

• Systematically varied high/low quality image ratios
• Evaluated finetuning-test distribution mismatch impact
• Quantified image quality importance for label efficiency