A study of why we need to reassess full reference image quality assessment with medical images
arXiv:2405.19097 · doi:10.1007/s10278-025-01462-1
Abstract
Image quality assessment (IQA) is indispensable in clinical practice to ensure high standards, as well as in the development stage of machine learning algorithms that operate on medical images. The popular full reference (FR) IQA measures PSNR and SSIM are known and tested for working successfully in many natural imaging tasks, but discrepancies in medical scenarios have been reported in the literature, highlighting the gap between development and actual clinical application. Such inconsistencies are not surprising, as medical images have very different properties than natural images, and PSNR and SSIM have neither been targeted nor properly tested for medical images. This may cause unforeseen problems in clinical applications due to wrong judgment of novel methods. This paper provides a structured and comprehensive overview of examples where PSNR and SSIM prove to be unsuitable for the assessment of novel algorithms using different kinds of medical images, including real-world MRI, CT, OCT, X-Ray, digital pathology and photoacoustic imaging data. Therefore, improvement is urgently needed in particular in this era of AI to increase reliability and explainability in machine learning for medical imaging and beyond. Lastly, we will provide ideas for future research as well as suggesting guidelines for the usage of FR-IQA measures applied to medical images.
References in corpus (20)
- Deep Convolutional Neural Network for Inverse Problems in Imaging
- Learned Primal-dual Reconstruction
- Metrics reloaded: Recommendations for image analysis validation
- Convex optimization problem prototyping for image reconstruction in computed tomography with the Chambolle-Pock algorithm
- A Haar Wavelet-Based Perceptual Similarity Index for Image Quality Assessment
- Results of the 2020 fastMRI Challenge for Machine Learning MR Image Reconstruction
- Advancing machine learning for MR image reconstruction with an open competition: Overview of the 2019 fastMRI challenge
- Deep learning for biomedical photoacoustic imaging: A review
- Noise2Inverse: Self-supervised deep convolutional denoising for tomography
- Estimation of blood oxygenation with learned spectral decoloring for quantitative photoacoustic imaging (LSD-qPAI)
- A Hitchhiker's Guide to Structural Similarity
- Task adapted reconstruction for inverse problems
- Artifact- and content-specific quality assessment for MRI with image rulers
- XPDNet for MRI Reconstruction: an application to the 2020 fastMRI challenge
- Learned convex regularizers for inverse problems
- Self-Supervised Training For Low Dose CT Reconstruction
- GMRES Methods for Tomographic Reconstruction with an Unmatched Back Projector
- On Krylov Methods for Large Scale CBCT Reconstruction
- Self-Supervised MRI Reconstruction with Unrolled Diffusion Models
- On the reconstruction accuracy of multi-coil MRI with orthogonal projections