Metrics that matter: Evaluating image quality metrics for medical image generation

Evaluating generative models for synthetic medical imaging is crucial yet challenging, especially given the high standards of fidelity, anatomical accuracy, and safety required for clinical applications. Standard evaluation of generated images often relies on no-reference image quality metrics when ground truth images are unavailable, but their reliability in this complex domain is not well established. This study comprehensively assesses commonly used no-reference image quality metrics using brain MRI data, including tumour and vascular images, providing a representative exemplar for the field. We systematically evaluate metric sensitivity to a range of challenges, including noise, distribution shifts, and, critically, localised morphological alterations designed to mimic clinically relevant inaccuracies. We then compare these metric scores against model performance on a relevant downstream segmentation task, analysing results across both controlled image perturbations and outputs from different generative model architectures. Our findings reveal significant limitations: many widely-used no-reference image quality metrics correlate poorly with downstream task suitability and exhibit a profound insensitivity to localised anatomical details crucial for clinical validity. Furthermore, these metrics can yield misleading scores regarding distribution shifts, e.g. data memorisation. This reveals the risk of misjudging model readiness, potentially leading to the deployment of flawed tools that could compromise patient safety. We conclude that ensuring generative models are truly fit for clinical purpose requires a multifaceted validation framework, integrating performance on relevant downstream tasks with the cautious interpretation of carefully selected no-reference image quality metrics.

Unsupervised Domain Adaptation for Brain Vessel Segmentation through Transwarp Contrastive Learning

Unsupervised domain adaptation (UDA) aims to align the labelled source distribution with the unlabelled target distribution to obtain domain-invariant predictive models. Since cross-modality medical data exhibit significant intra and inter-domain shifts and most are unlabelled, UDA is more important while challenging in medical image analysis. This paper proposes a simple yet potent contrastive learning framework for UDA to narrow the inter-domain gap between labelled source and unlabelled target distribution. Our method is validated on cerebral vessel datasets. Experimental results show that our approach can learn latent features from labelled 3DRA modality data and improve vessel segmentation performance in unlabelled MRA modality data.

GS-EMA: Integrating Gradient Surgery Exponential Moving Average with Boundary-Aware Contrastive Learning for Enhanced Domain Generalization in Aneurysm Segmentation

The automated segmentation of cerebral aneurysms is pivotal for accurate diagnosis and treatment planning. Confronted with significant domain shifts and class imbalance in 3D Rotational Angiography (3DRA) data from various medical institutions, the task becomes challenging. These shifts include differences in image appearance, intensity distribution, resolution, and aneurysm size, all of which complicate the segmentation process. To tackle these issues, we propose a novel domain generalization strategy that employs gradient surgery exponential moving average (GS-EMA) optimization technique coupled with boundary-aware contrastive learning (BACL). Our approach is distinct in its ability to adapt to new, unseen domains by learning domain-invariant features, thereby improving the robustness and accuracy of aneurysm segmentation across diverse clinical datasets. The results demonstrate that our proposed approach can extract more domain-invariant features, minimizing over-segmentation and capturing more complete aneurysm structures.

Learned Local Attention Maps for Synthesising Vessel Segmentations

Magnetic resonance angiography (MRA) is an imaging modality for visualising blood vessels. It is useful for several diagnostic applications and for assessing the risk of adverse events such as haemorrhagic stroke (resulting from the rupture of aneurysms in blood vessels). However, MRAs are not acquired routinely, hence, an approach to synthesise blood vessel segmentations from more routinely acquired MR contrasts such as T1 and T2, would be useful. We present an encoder-decoder model for synthesising segmentations of the main cerebral arteries in the circle of Willis (CoW) from only T2 MRI. We propose a two-phase multi-objective learning approach, which captures both global and local features. It uses learned local attention maps generated by dilating the segmentation labels, which forces the network to only extract information from the T2 MRI relevant to synthesising the CoW. Our synthetic vessel segmentations generated from only T2 MRI achieved a mean Dice score of $0.79 \pm 0.03$ in testing, compared to state-of-the-art segmentation networks such as transformer U-Net ($0.71 \pm 0.04$) and nnU-net($0.68 \pm 0.05$), while using only a fraction of the parameters. The main qualitative difference between our synthetic vessel segmentations and the comparative models was in the sharper resolution of the CoW vessel segments, especially in the posterior circulation.

Shape-guided Conditional Latent Diffusion Models for Synthesising Brain Vasculature

The Circle of Willis (CoW) is the part of cerebral vasculature responsible for delivering blood to the brain. Understanding the diverse anatomical variations and configurations of the CoW is paramount to advance research on cerebrovascular diseases and refine clinical interventions. However, comprehensive investigation of less prevalent CoW variations remains challenging because of the dominance of a few commonly occurring configurations. We propose a novel generative approach utilising a conditional latent diffusion model with shape and anatomical guidance to generate realistic 3D CoW segmentations, including different phenotypical variations. Our conditional latent diffusion model incorporates shape guidance to better preserve vessel continuity and demonstrates superior performance when compared to alternative generative models, including conditional variants of 3D GAN and 3D VAE. We observed that our model generated CoW variants that are more realistic and demonstrate higher visual fidelity than competing approaches with an FID score 53\% better than the best-performing GAN-based model.