Significance Public databases are an important resource for machine learning research, but their growing availability sometimes leads to “off-label” usage, where data published for one task are used for another. This work reveals that such off-label usage could lead to biased, overly optimistic results of machine-learning algorithms. The underlying cause is that public data are processed with hidden processing pipelines that alter the data features. Here we study three well-known algorithms developed for image reconstruction from magnetic resonance imaging measurements and show they could produce biased results with up to 48% artificial improvement when applied to public databases. We relate to the publication of such results as implicit “data crimes” to raise community awareness of this growing big data problem.

This work proposes the Temporal Differences (TED) Compressed Sensing (CS) method for accelerating thermal monitoring in MR‐guided High‐Intensity Focused Ultrasound (MRgHIFU) treatments. TED combines k‐space subsampling, parallel imaging, and a unique CS recovery of the temporal differences between pre‐heating and post‐heating multi‐coil data. TED was validated through retrospective experiments with (i) two phantom datasets acquired with 1.5 T and 3 T MRgHIFU systems from different vendors, (ii) data from an in vivo animal model experiment, and (iii) four datasets from clinical in vivo MRgHIFU treatments of prostate cancer in humans. TED produced highly accurate temperature change maps from subsampled k‐space data for all datasets. For the clinical in vivo data, an analysis of 105 time frames showed that the average TED reconstruction error is 1.06‐1.67 °C. Furthermore, TED consistently outperforms two state‐of‐the‐art methods, l1‐SPIRiT and the K‐space Hybrid Method, and offers errors that are significantly lower, by 29% or more. Moreover, TED offers robust performance over a range of its tunable parameters, stability across MRgHIFU systems from different vendors, and a short runtime of 1.7 s. In summary, TED enables k‐space subsampling while retaining high‐temperature mapping accuracy.

Purpose: To develop and test a novel parameter-free non-iterative wavelet domain method for reconstruction of undersampled multicoil MR data. Theory and Methods: A linear parallel MRI method that operates in the Stationary Wavelet Transform (SWT) domain is proposed. The method is coined COnvolution-based REconstruction for Parallel MRI (CORE-PI). This method computes the SWT of the unknown MR image directly from subsampled k-space measurements, without modifying the RF excitation pulse. It then reconstructs the image using the wavelet filter bank approach, with simple linear computations. The CORE-PI implementation is demonstrated by experiments with a numeric brain phantom and in vivo brain scans data, with various wavelet types and high reduction factors. It is compared to the well-known parallel MRI methods GRAPPA and l1-SPIRiT. Results: The experimental results show that CORE-PI is suitable for different 1D Cartesian k-space undersampling schemes, including regular and irregular ones, and for wavelets of different families. CORE-PI accurately reconstructs the SWT coefficients of the unknown MR image; this waveletdomain decomposition is fully computed despite the k-space undersampling. Furthermore, CORE-PI provides high-quality final reconstructions, with an average NRMSE of 0.013, which is significantly lower than that obtained by GRAPPA and l1-SPIRiT. Moreover, CORE-PI offers significantly faster computation times: the typical CORE-PI runtime is about 60 seconds, which is about 20% shorter than that of l1-SPIRiT and 55%-75% shorter than that of GRAPPA. Conclusion: COnvolution-based REconstruction for Parallel MRI advantageously offers: (a) flexible 1D undersampling of a Cartesian k-space, (b) a parameter-free non-iterative implementation, (c) reconstruction performance comparable or better than that of GRAPPA and l1-SPIRiT, and (d) robust fast computations.

No abstract

No abstract