# Open World MRI Reconstruction with Bias-Calibrated Adaptation Source: [Open World MRI Reconstruction with Bias-Calibrated Adaptation]() Collection index: [llms.txt]() Data updated: 2026-03-18T05:00:04.466Z Authors: Jiyao Liu, Shangqi Gao, Lihao Liu, Junzhi Ning, Jinjie Wei, Junjun He, Xiahai Zhuang, Ningsheng Xu Published: 2026-03-13T17:49:08.000Z ## Abstract Real-world MRI reconstruction systems face the open-world challenge: test data from unseen imaging centers, anatomical structures, or acquisition protocols can differ drastically from training data, causing severe performance degradation. Existing methods struggle with this challenge. To address this, we propose BiasRecon, a bias-calibrated adaptation framework grounded in the minimal intervention principle: preserve what transfers, calibrate what does not. Concretely, BiasRecon formulates open-world adaptation as an alternating optimization framework that jointly optimizes three components: (1) frequency-guided prior calibration that introduces layer-wise calibration variables to selectively modulate frequency-specific features of the pre-trained score network via self-supervised k-space signals, (2) score-based denoising that leverages the calibrated generative prior for high-fidelity image reconstruction, and (3) adaptive regularization that employs Stein's Unbiased Risk Estimator to dynamically balance the prior-measurement trade-off, matching test-time noise characteristics without requiring ground truth. By intervening minimally and precisely through this alternating scheme, BiasRecon achieves robust adaptation with fewer than 100 tunable parameters. Extensive experiments across four datasets demonstrate state-of-the-art performance on open-world reconstruction tasks. ## Recorded benchmark results A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. - [MRI reconstruction on FastMRI Knee In-Distribution (test)](): MRI Reconstruction; FastMRI Knee In-Distribution (test); PSNR: 31.76. - [MRI reconstruction on FastMRI-Brain Cross-Anatomy (test)](): MRI Reconstruction; FastMRI-Brain Cross-Anatomy (test); PSNR: 33.65. - [MRI Reconstruction on Stanford Knee Cross-Center (test)](): MRI Reconstruction; Stanford Knee Cross-Center (test); PSNR: 31.29. - [MRI reconstruction on CMRxRecon Combined shift (test)](): MRI Reconstruction; CMRxRecon Combined shift (test); PSNR: 35.32. - [MRI Reconstruction on FastMRI-brain open-world (Gaussian 1D sampling, 4x acceleration)](): MRI Reconstruction; FastMRI-brain open-world (Gaussian 1D sampling, 4x acceleration); PSNR: 35.03. - [MRI Reconstruction on FastMRI Brain Gaussian 1D 8x Acceleration](): MRI Reconstruction; FastMRI-brain open-world Gaussian 1D sampling 8x acceleration; PSNR: 33.65. - [MRI Reconstruction on open-world FastMRI-brain (Uniform 1D sampling, 4x acceleration)](): MRI Reconstruction; open-world FastMRI-brain Uniform 1D sampling, 4x acceleration; PSNR: 32.72. - [MRI Reconstruction on FastMRI-brain open-world Uniform 1D sampling 8x acceleration](): MRI Reconstruction; FastMRI-brain open-world Uniform 1D sampling 8x acceleration; PSNR: 28.4. ## Original sources and implementation [Original paper]() [PDF]()