Saved in:
| Main Authors: | , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2508.11864 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908496148037632 |
|---|---|
| author | Tang, Yucheng Rajwa, Pawel Ng, Alexander Wang, Yipei Yan, Wen Thorley, Natasha Asif, Aqua Allen, Clare Dickinson, Louise Giganti, Francesco Punwani, Shonit Alexander, Daniel C. Kasivisvanathan, Veeru Hu, Yipeng |
| author_facet | Tang, Yucheng Rajwa, Pawel Ng, Alexander Wang, Yipei Yan, Wen Thorley, Natasha Asif, Aqua Allen, Clare Dickinson, Louise Giganti, Francesco Punwani, Shonit Alexander, Daniel C. Kasivisvanathan, Veeru Hu, Yipeng |
| contents | Foundation models in medical imaging have shown promising label efficiency, achieving high performance on downstream tasks using only a fraction of the annotated data otherwise required. In this study, we evaluate this potential in the context of prostate multiparametric MRI using ProFound, a recently developed domain-specific vision foundation model pretrained on large-scale prostate MRI datasets. We investigate the impact of variable image quality on the label-efficient finetuning, by quantifying the generalisability of the finetuned models. We conduct a comprehensive set of experiments by systematically varying the ratios of high- and low-quality images in the 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. Importantly, experimental results also show that, although finetuning requires significantly less labeled data compared to training from scratch when the quality ratio is consistent, this label efficiency is not independent of the image quality distribution. For example, we show cases that, without sufficient high-quality images in finetuning, finetuned models may fail to outperform those without pretraining. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2508_11864 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Impact of Clinical Image Quality on Efficient Foundation Model Finetuning Tang, Yucheng Rajwa, Pawel Ng, Alexander Wang, Yipei Yan, Wen Thorley, Natasha Asif, Aqua Allen, Clare Dickinson, Louise Giganti, Francesco Punwani, Shonit Alexander, Daniel C. Kasivisvanathan, Veeru Hu, Yipeng Computer Vision and Pattern Recognition Foundation models in medical imaging have shown promising label efficiency, achieving high performance on downstream tasks using only a fraction of the annotated data otherwise required. In this study, we evaluate this potential in the context of prostate multiparametric MRI using ProFound, a recently developed domain-specific vision foundation model pretrained on large-scale prostate MRI datasets. We investigate the impact of variable image quality on the label-efficient finetuning, by quantifying the generalisability of the finetuned models. We conduct a comprehensive set of experiments by systematically varying the ratios of high- and low-quality images in the 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. Importantly, experimental results also show that, although finetuning requires significantly less labeled data compared to training from scratch when the quality ratio is consistent, this label efficiency is not independent of the image quality distribution. For example, we show cases that, without sufficient high-quality images in finetuning, finetuned models may fail to outperform those without pretraining. |
| title | Impact of Clinical Image Quality on Efficient Foundation Model Finetuning |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2508.11864 |