REVIEW 3 major objections 5 minor 5 references
Optimal Windowing of MR Images using Deep Learning: An Enabler for Enhanced Visualization
T0 review · 3 major / 5 minor · reviewed 2026-08-14 · deepseek-v4-flash
Pith's one-line read Removing background with a deep neural net improves automatic MR windowing.
desk verdict Good practical idea, but the reported evidence does not isolate windowing improvement from background masking, so the central claim is unproven as presented. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The load-bearing object is the U-net, a convolutional neural network trained to output an anatomy mask and thereby suppress background pixels. It carries the argument because the foreground histogram it produces is the sole input to the new WW/WL computation; every downstream gain in windowing inherits the segmentation's accuracy. The method's other component is the standard WW/WL calculation applied only to those foreground pixels.
What would settle it
Run the trained U-net on a set of MR images from an anatomy or protocol not in the training set (for example, cardiac or spine), compute WW/WL from the masked foreground, and blind-compare with the default windowing; if radiologists prefer the default, or if the mask excludes diagnostically relevant tissue, on a substantial fraction of cases, the claimed generic improvement fails.
Extended reading notes
Core claim
The central claim is that WW/WL computed exclusively from foreground pixels of a background-suppressed MR image are superior to those computed from the full image, because background noise otherwise forces a trade-off between foreground contrast and background suppression. The paper proposes to obtain the foreground by applying a U-net trained on 2700 brain and abdomen images, then compute WW/WL from the resulting mask, and display the segmented image with the new settings. The authors assert that this improves display quality and user control for both original and derived images without compromising anatomical detail.
Load-bearing premise
The method's benefit rests on the U-net generalizing from its 2700 brain and abdomen training images to all other anatomies and protocols with the same segmentation quality; if it mislabels foreground on an untested anatomy, the window settings are computed from a corrupted histogram and can be worse than the default.
Editorial extensions
If this is right
- Default window settings in MR viewers could be computed from foreground-only histograms, removing the background-noise trade-off that forces manual correction.
- Derived images such as functional maps, not just source anatomies, would inherit better automatic windowing from the same background-suppression step.
- Manual post-processing becomes easier because background pixels are already eliminated, so user adjustments affect only the anatomy of interest.
- The same pipeline can be applied without per-image tuning, relying on the trained U-net rather than hand-crafted thresholds or level sets.
- If the U-net generalizes as claimed, the approach would replace thresholding-based auto-windowing methods in clinical display pipelines.
Reading between the lines
- A natural extension would measure the windowing improvement directly (for instance, contrast-to-noise or reader agreement) rather than via DICE alone, since segmentation accuracy is only a proxy for display quality.
- The background-suppression step could be reused as a preprocessing module for other visualization and analysis tasks, such as volume rendering, automated measurements, or AI-based triage, wherever the background histogram distorts the signal.
- A testable cross-domain claim follows: the same U-net architecture should improve windowing on other anatomies and sequences, but only if its mask quality transfers; cardiac, spine, or diffusion-weighted images would be a straightforward check.
- One could also test whether suppressing background before manual adjustment reduces radiologist interaction time, a consequence the paper implies but does not quantify.
Signed reviews
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper proposes a two-stage pipeline for improving window width (WW) and window level (WL) in MR images: first, a U-net removes background/anatomy-of-no-interest pixels; second, WW/WL are computed from the remaining foreground pixels. The method is intended to improve automatic windowing and to give users better control when manually adjusting windows, for both original and derived MR images. The authors report an average DICE score of 0.94 on 1130 test images covering brain and abdomen, and state that clinical application specialists confirmed that background-suppressed images had better WW/WL and better dynamic adjustability.
Significance. If validated, the proposed pipeline would be a simple and potentially generic preprocessing step for MR display that could reduce the need for manual WW/WL adjustment and improve image readability. The segmentation component is evaluated on a reasonably sized test set with a strong average DICE score (0.94), and the non-circular segmentation evaluation is a strength. However, the central claim of improved windowing is supported only by an undescribed qualitative review, and the evaluation is confounded because the displayed images are background-suppressed rather than the original images rendered with the new WW/WL. The claimed generality to all anatomies, protocols, and derived images is not tested. The significance of the paper as presented is therefore limited, though the idea is plausible and the gap is closable with additional experiments.
major comments (3)
- [Section 3, Methodology; Section 4, Result 2] The evaluation does not isolate the effect of the newly computed WW/WL from the effect of background masking. In Section 3 the displayed image is the segmented (background-suppressed) image, and in Section 4 the reviewers were shown background-suppressed images with the new WW/WL; the observed improvement could therefore be caused entirely by hiding the background. A control condition is required: the original, unmasked image rendered with the foreground-derived WW/WL should be compared with the default WW/WL, and the masked image should also be rendered with the default WW/WL. Without such comparisons, the conclusion in Section 5 that the proposed method significantly improves WW/WL is not supported.
- [Section 4, Result 1] DICE is a segmentation metric, not a windowing metric. The reported average DICE of 0.94 with a minimum of 0.713 implies that some test images have substantial segmentation errors, which can directly corrupt the foreground histogram used for WW/WL; no analysis shows how segmentation error propagates to WW/WL. The paper should report a quantitative windowing-quality measure, such as agreement with expert-chosen WW/WL or an objective contrast metric, with error bars and a baseline comparison.
- [Sections 2 and 5] The paper claims the DL approach is generic and performs equally well for all applications, anatomies, and protocols, and that WW/WL is improved for derived images, but the experiments cover only brain and abdomen images and no derived images. These claims should be removed or explicitly restricted to the tested domains, or supported by additional experiments.
minor comments (5)
- [Section 4, Figure 1] The manuscript references 'Figure 1' but the full text contains no figure, so the reader cannot inspect the claimed visual improvement.
- [Section 4, Result 2] The image review by clinical application specialists is not described; the manuscript does not report the number of reviewers, the viewing protocol, the rating scale, or the baseline images, so the qualitative result cannot be interpreted.
- [Section 1] The term 'derived images (e.g., functional maps)' is not defined, and no examples or results for such images are provided.
- [Section 3] The U-net architecture, training hyperparameters, data split, and preprocessing steps are not described, which limits reproducibility; please add a brief description or a reference.
- [References] The reference 'Kaushik et al.' is incomplete; it lacks a year and publication venue.
Circularity Check
Windowing benefit is evaluated on the same background-suppressed display that defines the new WW/WL, so the improvement claim is not isolated from the masking step.
-
other
[Section 3 Methodology steps 1-3; Section 4 Results item 2; Section 5 Conclusion]
"Use the foreground pixels to compute the WW/WL. Display the segmented image with the newly computed WW/WL. ... The background suppressed images were presented with a better WW/WL than the original images ... The WW/WL of both original as well as derived images can be significantly improved by pre-processing using proposed method."
The new WW/WL in step 2 is computed from the foreground pixels selected by the U-net mask, and in step 3 the same mask is used to produce the displayed 'segmented image.' The Section 4 evaluation asks reviewers to compare these background-suppressed images with the original images, so the perceived 'better WW/WL' is inseparable from the masking operation. Because the original image is never rendered with the foreground-derived WW/WL and compared with the default window, the conclusion that WW/WL itself is improved is not independently tested: the improvement is attributed to the pipeline's combined output, making it partly true by construction.
full rationale
The segmentation leg is non-circular: the U-net is trained on 2700 brain/abdomen images with masks and evaluated on 1130 held-out images with DICE (average 0.94, min 0.713, max 0.982), which is an external ground-truth metric. There are no load-bearing self-citations and no fitted parameters renamed as predictions. However, the windowing claim, which is the paper's central contribution, is validated on the same background-suppressed display that is used to compute the new WW/WL, so the 'better window' verdict cannot be separated from the background-masking step. No comparison with the original image rendered at the foreground-derived WW/WL is provided, and the minimum DICE of 0.713 shows segmentation errors that could corrupt the foreground histogram; no objective windowing baseline is given. This is a construction-borne evaluation confound rather than a full derivation-equivalent circularity, so the score is moderate rather than extreme.
Assumptions & free parameters
free parameters (1)
- U-net model parameters =
not reported
assumptions (3)
- domain assumption Background noise in MR images biases default WW/WL calculations, so removing background improves windowing.
- domain assumption A generic deep learning segmentation method works equally well for all applications, anatomies, and protocols.
- domain assumption Clinical application specialist review is a valid measure of WW/WL quality.
Cite this review
Pith. "Pith review of Optimal Windowing of MR Images using Deep Learning: An Enabler for Enhanced Visualization." pith.science (2026). https://pith.science/paper/UXZTZACR
@misc{pith2026190800822,
author = {Pith},
title = {Pith review of: Optimal Windowing of MR Images using Deep Learning: An Enabler for Enhanced Visualization},
year = {2026},
howpublished = {\url{https://pith.science/paper/UXZTZACR}},
note = {Machine review of arXiv:1908.00822}
}
read the original abstract
Window width (WW) and window level (WL) adjustments aid in visualizing anatomies with a suitable contrast. However, the presence of background noise in MR images biases the calculation of default WW/WL values since it necessitates a trade-off between enhancing contrast of foreground/anatomy of interest vs suppressing background/ outside the anatomy of interest. This paper proposes an intelligent algorithm to improve the automatic computation of WW/WL and provide better control for user defined windowing.This is achieved by first eliminating the background pixels using a Deep Neural network and then computing WW/WL.
Figures
Reference graph
Works this paper leans on
-
[1]
Method for automated window-level settings for magnetic resonance images, May 15 2007
Igor Belykh and Craig W Cornelius. Method for automated window-level settings for magnetic resonance images, May 15 2007. US Patent 7,218,763
work page 2007
-
[2]
Automatic windowing method for mr images, May 4 1999
Joel P Felmlee, William Ryan, Ramesh Avula, and Bradley J Erickson. Automatic windowing method for mr images, May 4 1999. US Patent 5,900,732
work page 1999
-
[3]
Adaptive intensity windowing for mr images
Sandeep S Kaushik, Shashank Mujumdar, and Jayanthi Sivaswamy. Adaptive intensity windowing for mr images
-
[4]
An adaptive window width/center adjustment system with online training capabilities for mr images
Shang-Hong Lai and Ming Fang. An adaptive window width/center adjustment system with online training capabilities for mr images. Artificial intelligence in medicine, 33 0 (1): 0 89--101, 2005
work page 2005
-
[5]
Automated medical image segmentation techniques
Neeraj Sharma and Lalit M Aggarwal. Automated medical image segmentation techniques. Journal of medical physics/Association of Medical Physicists of India, 35 0 (1): 0 3, 2010
work page 2010
Reviewed August 14, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.