Lianrui Zuo
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Lianrui Zuo has a total of 211 co-authors. Here are the top co-authors by number of shared publications:
- Bennett A. Landman
Department of Electrical and Computer Engineering Vanderbilt University Nashville Tennessee USA
14 shared publications - Aaron Carass
Johns Hopkins University
13 shared publications - Yihao Liu
Vanderbilt University
12 shared publications - Jerry L. Prince
Johns Hopkins University
10 shared publications Yihao Liu
Department of Electrical and Computer Engineering, Vanderbilt University, Nashville, 37235, TN, United States
10 shared publications- Blake E. Dewey
Department of Neurology, Johns Hopkins University School of Medicine, Baltimore, MD, USA
7 shared publications
Latest Publications
- Open Access
- 11 October 2026
- Published by Elsevier in SSRN Electronic Journal
Identity-Duplication Auditing in National-Scale Neuroimaging Repositories
Objective . National-scale magnetic resonance imaging (MRI) repositories increasingly integrate data from different studies and institutions. However, subject identifiers that are valid only within individual datasets are no longer guaranteed to remain globally unique after aggregation, making it possible for the same subject to be assigned multiple identifiers, which we define as identity duplication. Such duplication can create leakage between training and test data and inflate apparent performance in downstream biomedical studies. Existing methods do not provide an end-to-end, image-based workflow for auditing this problem at repository scale. Methods. HAPPEN is a human-in-the-loop pipeline for auditing identity duplication in T1-weighted brain MRI repositories. It combines SHA-256 fingerprinting for exact-duplicate detection with supervised contrastive retrieval of non-identical scans that may originate from the same person. Retrieved pairs are reviewed as candidates in a locally hosted interface rather than automatically classified as duplicates. We benchmarked retrieval against PCA and 3D-SIFT-Rank in an internal test set and two withheld cohorts. We deployed the workflow in a 95,129-scan aggregated repository and assessed end-to-end recovery using 54 genetic-reference pairs. Transferability was assessed by locally deploying the same workflow on 22,386 scans at an independent institution without model retraining or image transfer.Results. HAPPEN achieved near-ceiling same-subject retrieval across all three cohorts and matched or exceeded the baseline point estimates on every metric. Deployment in the study repository identified 1,316 exact-duplicate scan groups and 1,275 reviewer-supported near-duplicate subject groups. Of these groups, 56% and 82%, respectively, crossed dataset boundaries. All 54 genetic-reference pairs were recovered. The external team independently completed the full workflow using a locally selected operating threshold and review standard.Conclusion. Together, these findings indicate that identity duplication is a structural risk of MRI repository aggregation and that HAPPEN can support practical, locally governed quality assurance.
- 11 October 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessIdentity-Duplication Auditing in National-Scale Neuroimaging Repositories
Objective . National-scale magnetic resonance imaging (MRI) repositories increasingly integrate data from different studies and institutions. However, subject identifiers that are valid only within individual datasets are no longer guaranteed to remain globally unique after aggregation, making it possible for the same subject to be assigned multiple identifiers, which we define as identity duplication. Such duplication can create leakage between training and test data and inflate apparent performance in downstream biomedical studies. Existing methods do not provide an end-to-end, image-based workflow for auditing this problem at repository scale. Methods. HAPPEN is a human-in-the-loop pipeline for auditing identity duplication in T1-weighted brain MRI repositories. It combines SHA-256 fingerprinting for exact-duplicate detection with supervised contrastive retrieval of non-identical scans that may originate from the same person. Retrieved pairs are reviewed as candidates in a locally hosted interface rather than automatically classified as duplicates. We benchmarked retrieval against PCA and 3D-SIFT-Rank in an internal test set and two withheld cohorts. We deployed the workflow in a 95,129-scan aggregated repository and assessed end-to-end recovery using 54 genetic-reference pairs. Transferability was assessed by locally deploying the same workflow on 22,386 scans at an independent institution without model retraining or image transfer.Results. HAPPEN achieved near-ceiling same-subject retrieval across all three cohorts and matched or exceeded the baseline point estimates on every metric. Deployment in the study repository identified 1,316 exact-duplicate scan groups and 1,275 reviewer-supported near-duplicate subject groups. Of these groups, 56% and 82%, respectively, crossed dataset boundaries. All 54 genetic-reference pairs were recovered. The external team independently completed the full workflow using a locally selected operating threshold and review standard.Conclusion. Together, these findings indicate that identity duplication is a structural risk of MRI repository aggregation and that HAPPEN can support practical, locally governed quality assurance.
- 22 September 2026
- Published by Springer Nature
- p. 3-13
An Artifact-Based Agent Framework for Adaptive and Reproducible Medical Image Processing
- 22 September 2026
- Published by Springer Nature
- p. 3-13
An Artifact-Based Agent Framework for Adaptive and Reproducible Medical Image Processing
- Open Access
- 1 September 2026
- Published by SPIE-Intl Soc Optical Eng in Journal of Medical Imaging
- Vol. 13 (05) , 1-19
Contrastive learning and advanced bundle metrics outperform fractional anisotropy for Alzheimer’s disease diffusion MRI prediction: findings from the Beyond FA challenge
PurposeThe inclusion of diffusion-weighted magnetic resonance imaging (DW-MRI) in major national studies has allowed for quantification and characterization of white matter macrostructure and microstructure in cases of disease and disorder. Diffusion tensor imaging (DTI) yields 3D scalar maps to summarize the DW-MRI signal with quantitative metrics. Fractional anisotropy (FA) is a DTI metric that describes the degree of anisotropy within diffusion and is frequently used to interpret white matter integrity, but it exhibits high sensitivity in pathological interpretation. These properties of FA make it difficult to use as a consistent candidate imaging biomarker, as it makes different ages, cognitive statuses, and pathologies difficult to discern. With the development of other white matter models and metrics, such as structural connectivity network measures and tractography bundle analyses, there are many combinations to analyze for white matter integrity analysis, especially in the context of lower-quality clinical data that FA’s voxel-wise computation can be sensitive to. The widespread collection of DW-MRI data necessitates an analysis of these metrics to interpret white matter integrity beyond what FA can capture.ApproachWe designed a challenge for the medical imaging community to compare crowdsourced DW-MRI candidate imaging biomarkers to a baseline relying on FA in an age, sex, and cognitive status classification task and variables unknown to teams during the challenge. After the challenge submission period ended, we evaluated methods based on sensitivity to environmental variables (humidity and temperature on the day of the scan), and we correlated features with known Alzheimer’s disease (AD) pathology (amyloid beta 42, tau, and ptau cerebrospinal fluid/CSF readings) to evaluate if the methods collected in this study are robust to the limitations of FA. The key innovations are that we crowdsource DW-MRI feature representations for AD, we develop a framework to compare imaging feature representations in this context, we evaluate the sensitivity of DW-MRI imaging candidate biomarkers to environmental variables, and we correlate the novel features to known AD pathological measures.ResultsWe find that bundle-based methods outperform whole-brain methods in age classification tasks, but deep learning, computer vision-based methods show a new avenue for cognitive status prediction tasks.ConclusionThe submissions to our challenge show that there are many approaches to consider when developing new AD feature representations.
- 1 September 2026
- Published by SPIE-Intl Soc Optical Eng in Journal of Medical Imaging
- Vol. 13 (05) , 1-19
Open AccessContrastive learning and advanced bundle metrics outperform fractional anisotropy for Alzheimer’s disease diffusion MRI prediction: findings from the Beyond FA challenge
PurposeThe inclusion of diffusion-weighted magnetic resonance imaging (DW-MRI) in major national studies has allowed for quantification and characterization of white matter macrostructure and microstructure in cases of disease and disorder. Diffusion tensor imaging (DTI) yields 3D scalar maps to summarize the DW-MRI signal with quantitative metrics. Fractional anisotropy (FA) is a DTI metric that describes the degree of anisotropy within diffusion and is frequently used to interpret white matter integrity, but it exhibits high sensitivity in pathological interpretation. These properties of FA make it difficult to use as a consistent candidate imaging biomarker, as it makes different ages, cognitive statuses, and pathologies difficult to discern. With the development of other white matter models and metrics, such as structural connectivity network measures and tractography bundle analyses, there are many combinations to analyze for white matter integrity analysis, especially in the context of lower-quality clinical data that FA’s voxel-wise computation can be sensitive to. The widespread collection of DW-MRI data necessitates an analysis of these metrics to interpret white matter integrity beyond what FA can capture.ApproachWe designed a challenge for the medical imaging community to compare crowdsourced DW-MRI candidate imaging biomarkers to a baseline relying on FA in an age, sex, and cognitive status classification task and variables unknown to teams during the challenge. After the challenge submission period ended, we evaluated methods based on sensitivity to environmental variables (humidity and temperature on the day of the scan), and we correlated features with known Alzheimer’s disease (AD) pathology (amyloid beta 42, tau, and ptau cerebrospinal fluid/CSF readings) to evaluate if the methods collected in this study are robust to the limitations of FA. The key innovations are that we crowdsource DW-MRI feature representations for AD, we develop a framework to compare imaging feature representations in this context, we evaluate the sensitivity of DW-MRI imaging candidate biomarkers to environmental variables, and we correlate the novel features to known AD pathological measures.ResultsWe find that bundle-based methods outperform whole-brain methods in age classification tasks, but deep learning, computer vision-based methods show a new avenue for cognitive status prediction tasks.ConclusionThe submissions to our challenge show that there are many approaches to consider when developing new AD feature representations.
- 1 September 2026
- Published by SPIE-Intl Soc Optical Eng in Journal of Medical Imaging
- Vol. 13 (05) , 1-20
Charting confidence in white matter brain charts: enabling study planning through stability validation
PurposeNormative modeling of quantitative brain measurements is being widely adopted as a promising method for charting population-level developmental trajectories and identifying abnormalities in groups and individuals. Generalized additive models for location, scale, and shape (GAMLSS) are a statistical framework that has been identified by the World Health Organization as a robust method for large-scale population modeling. Validating the stability of normative models is essential for appropriate benchmarking and detecting anomalies, especially for patients with neurodegenerative or developmental disorders. As these models continue to expand in the number of features and amount of data, it becomes increasingly important to assess the stability of the models and their predictions.ApproachAn analytic approach is extended from previous work to assess variability of the GAMLSS framework for normative modeling of white matter measurements across the lifespan. The analytic variance and bias are compared with empirical bootstrapping estimates and those of an analogous model that is a nonparametric, data-driven analog.ResultsAcross all models, the analytic approach showed low model variability (<0.5% Coefficient of Variation) for lifespan trajectories that increased with age. Empirical variability of trajectories was slightly elevated by comparison, especially at the beginning of the lifespan. The trajectory variability for the data-driven analog was even larger, showing a ten-fold increase from the empirical GAMLSS variability. For individual centile prediction, the majority of data points for the analytic assessment had a mean absolute difference (MAD) in centile score prediction as z-scores that was less than 0.003 with the standard deviation of these errors below 0.06. The empirical assessment showed an elevated standard deviation (0.15) and MAD of z-score centile predictions (0.06) for the majority of datapoints. By contrast, when compared with the empirical assessment, the nonparametric, data-driven analog showed over a two-fold increase in the MAD of centile z-score predictions (0.14) and a three-fold increase in standard deviation (0.45) for the majority of data points. The divergences in variability across assessments suggest that individual centile predictions are more sensitive to random fluctuations in the data, which the nonparametric data-driven approach exacerbates.ConclusionsResults suggest that the GAMLSS brain charts are a stable method of defining and charting normative white matter development as well as planning future clinical or research studies. Overall, this study provides an assessment of GAMLSS-derived brain chart variance in a large-scale population.
- 1 September 2026
- Published by SPIE-Intl Soc Optical Eng in Journal of Medical Imaging
- Vol. 13 (05) , 1-20
Charting confidence in white matter brain charts: enabling study planning through stability validation
PurposeNormative modeling of quantitative brain measurements is being widely adopted as a promising method for charting population-level developmental trajectories and identifying abnormalities in groups and individuals. Generalized additive models for location, scale, and shape (GAMLSS) are a statistical framework that has been identified by the World Health Organization as a robust method for large-scale population modeling. Validating the stability of normative models is essential for appropriate benchmarking and detecting anomalies, especially for patients with neurodegenerative or developmental disorders. As these models continue to expand in the number of features and amount of data, it becomes increasingly important to assess the stability of the models and their predictions.ApproachAn analytic approach is extended from previous work to assess variability of the GAMLSS framework for normative modeling of white matter measurements across the lifespan. The analytic variance and bias are compared with empirical bootstrapping estimates and those of an analogous model that is a nonparametric, data-driven analog.ResultsAcross all models, the analytic approach showed low model variability (<0.5% Coefficient of Variation) for lifespan trajectories that increased with age. Empirical variability of trajectories was slightly elevated by comparison, especially at the beginning of the lifespan. The trajectory variability for the data-driven analog was even larger, showing a ten-fold increase from the empirical GAMLSS variability. For individual centile prediction, the majority of data points for the analytic assessment had a mean absolute difference (MAD) in centile score prediction as z-scores that was less than 0.003 with the standard deviation of these errors below 0.06. The empirical assessment showed an elevated standard deviation (0.15) and MAD of z-score centile predictions (0.06) for the majority of datapoints. By contrast, when compared with the empirical assessment, the nonparametric, data-driven analog showed over a two-fold increase in the MAD of centile z-score predictions (0.14) and a three-fold increase in standard deviation (0.45) for the majority of data points. The divergences in variability across assessments suggest that individual centile predictions are more sensitive to random fluctuations in the data, which the nonparametric data-driven approach exacerbates.ConclusionsResults suggest that the GAMLSS brain charts are a stable method of defining and charting normative white matter development as well as planning future clinical or research studies. Overall, this study provides an assessment of GAMLSS-derived brain chart variance in a large-scale population.
- 23 June 2026
- Published by Elsevier in Medical Image Analysis
- Vol. 113, 104175
Beyond the LUMIR challenge: The pathway to foundational registration models
- 23 June 2026
- Published by Elsevier in Medical Image Analysis
- Vol. 113, 104175
Beyond the LUMIR challenge: The pathway to foundational registration models
- Open Access
- 23 June 2026
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 21 (6) , e0341522
Longitudinal analysis of CYFRA 21-1 levels in patients with pulmonary nodules: Differential trajectories between benign and malignant cases
Background: CYFRA 21−1, a cytokeratin-19 fragment, is a validated serum biomarker for non-small cell lung cancer (NSCLC). However, most studies rely on single time-point measurements, limiting its specificity in differentiating malignancy from benign pulmonary conditions. Inspired by the clinical utility of serial PSA measurements in prostate cancer, we investigated whether longitudinal trends in CYFRA 21−1 could enhance diagnostic and monitoring capabilities in patients with pulmonary nodules. Methods and Findings: We analyzed 132 patients with pulmonary nodules from the Vanderbilt Thoracic Biorepository. For the primary analysis, patients who underwent treatment prior to biomarker assessment were excluded, resulting in an untreated cohort of 121 patients (91 benign and 30 malignant nodules). CYFRA 21−1 levels were measured serially using electrochemiluminescence assays. Longitudinal trends were assessed using linear mixed-effects models to estimate biomarker trajectories. Primary analyses compared benign vs. malignant nodules using longitudinal modeling of log-transformed CYFRA 21−1 values. At baseline, CYFRA 21−1 levels were significantly higher in malignant versus benign nodules. Longitudinal mixed-effects modeling did not demonstrate statistically significant differences in trajectories between benign and malignant nodules. Benign nodules showed a small positive trend in log(CYFRA 21−1) whereas malignant nodules showed greater longitudinal variability. The magnitude of change assessed using the absolute slope of log(CYFRA 21−1) was significantly greater in malignant nodules compared with benign nodules (p < 0.05). Exploratory diagnostic analysis showed that baseline log(CYFRA 21−1) achieved and AUC of 0.68 (95% CI 0.56-0.79) with sensitivity 0.63 and specificity 0.71. The absolute slope of log(CYFRA 21−1) yielded an AUC of 0.67 (95% CI 0.48-0.87) with sensitivity 0.39 and specificity 0.97. Conclusions: CYFRA 21−1 exhibits substantial within-patient variability over time, with trajectories that reflect disease state and treatment. These findings suggest that longitudinal monitoring of CYFRA 21−1 may provide additional information beyond single time-point measurements in the evaluation of pulmonary nodules. Further studies in large prospective cohorts are warranted to validate these findings before clinical implementation.
- 23 June 2026
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 21 (6) , e0341522
Open AccessLongitudinal analysis of CYFRA 21-1 levels in patients with pulmonary nodules: Differential trajectories between benign and malignant cases
Background: CYFRA 21−1, a cytokeratin-19 fragment, is a validated serum biomarker for non-small cell lung cancer (NSCLC). However, most studies rely on single time-point measurements, limiting its specificity in differentiating malignancy from benign pulmonary conditions. Inspired by the clinical utility of serial PSA measurements in prostate cancer, we investigated whether longitudinal trends in CYFRA 21−1 could enhance diagnostic and monitoring capabilities in patients with pulmonary nodules. Methods and Findings: We analyzed 132 patients with pulmonary nodules from the Vanderbilt Thoracic Biorepository. For the primary analysis, patients who underwent treatment prior to biomarker assessment were excluded, resulting in an untreated cohort of 121 patients (91 benign and 30 malignant nodules). CYFRA 21−1 levels were measured serially using electrochemiluminescence assays. Longitudinal trends were assessed using linear mixed-effects models to estimate biomarker trajectories. Primary analyses compared benign vs. malignant nodules using longitudinal modeling of log-transformed CYFRA 21−1 values. At baseline, CYFRA 21−1 levels were significantly higher in malignant versus benign nodules. Longitudinal mixed-effects modeling did not demonstrate statistically significant differences in trajectories between benign and malignant nodules. Benign nodules showed a small positive trend in log(CYFRA 21−1) whereas malignant nodules showed greater longitudinal variability. The magnitude of change assessed using the absolute slope of log(CYFRA 21−1) was significantly greater in malignant nodules compared with benign nodules (p < 0.05). Exploratory diagnostic analysis showed that baseline log(CYFRA 21−1) achieved and AUC of 0.68 (95% CI 0.56-0.79) with sensitivity 0.63 and specificity 0.71. The absolute slope of log(CYFRA 21−1) yielded an AUC of 0.67 (95% CI 0.48-0.87) with sensitivity 0.39 and specificity 0.97. Conclusions: CYFRA 21−1 exhibits substantial within-patient variability over time, with trajectories that reflect disease state and treatment. These findings suggest that longitudinal monitoring of CYFRA 21−1 may provide additional information beyond single time-point measurements in the evaluation of pulmonary nodules. Further studies in large prospective cohorts are warranted to validate these findings before clinical implementation.
- Open Access
- 12 May 2026
- Published by Elsevier in NeuroImage
- Vol. 335, 121979
Editorial for the Special Issue on Harmonization Techniques for MRI
- 12 May 2026
- Published by Elsevier in NeuroImage
- Vol. 335, 121979
Open AccessEditorial for the Special Issue on Harmonization Techniques for MRI
- Top Cited PapersOpen Access
- 20 January 2026
- Published by Elsevier in Medical Image Analysis
- Vol. 109, 103954
UNISELF: A unified network with instance normalization and self-ensembled lesion fusion for multiple sclerosis lesion segmentation
- 20 January 2026
- Published by Elsevier in Medical Image Analysis
- Vol. 109, 103954
Top Cited PapersOpen AccessUNISELF: A unified network with instance normalization and self-ensembled lesion fusion for multiple sclerosis lesion segmentation
- Open Access
Longitudinal Analysis of CYFRA 21-1 Levels in Patients with Pulmonary Nodules: Differential Trajectories Between Benign and Malignant Cases and Impact of Tumor Resection
Background CYFRA 21-1, a cytokeratin-19 fragment, is a validated serum biomarker for non-small cell lung cancer (NSCLC). However, most studies rely on single time-point measurements, limiting its specificity in differentiating malignancy from benign pulmonary conditions. Inspired by the clinical utility of serial PSA measurements in prostate cancer, we investigated whether longitudinal trends in CYFRA 21-1 could enhance diagnostic and monitoring capabilities in patients with pulmonary nodules Methods and Findings We analyzed 132 patients with pulmonary nodules, including 41 with lung cancer and 91 with benign diagnoses. CYFRA 21-1 levels were measured serially using electrochemiluminescence assays. Longitudinal trends were assessed using linear mixed-effects models to estimate biomarker trajectories. Subgroup analyses examined differences between benign, untreated cancer, and post-treatment cancer groups, as well as within-patient changes in a subset of 16 cancer patients with both pre- and post-surgical measurements. Log-transformed data were used for the analysis. At baseline, CYFRA 21-1 levels were significantly higher in malignant versus benign nodules. Over time, CYFRA trajectories diverged: benign cases showed slight increases, whereas cancer patients exhibited greater biomarker volatility. In treated cancer patients, trend of CYFRA levels on the natural log scale decline from –0.00137 pre-surgery to –0.00263 to post-surgery, and both cancer groups showed significantly higher absolute slopes than the benign group (p < 0.05). While pre- vs post-treatment slope differences did not reach significance (p = 0.211), the general pattern indicated that CYFRA 21-1 is a dynamic marker responsive to tumor presence and removal. Conclusions CYFRA 21-1 exhibits substantial within-patient variability over time, with trajectories that reflect disease state and treatment. These findings suggest that longitudinal monitoring of CYFRA 21-1—analogous to PSA velocity in prostate cancer—may offer improved diagnostic and prognostic insight in the evaluation of pulmonary nodules. Further studies in larger cohorts are warranted to validate these findings and explore clinical implementation of CYFRA trajectory analysis.
Open AccessLongitudinal Analysis of CYFRA 21-1 Levels in Patients with Pulmonary Nodules: Differential Trajectories Between Benign and Malignant Cases and Impact of Tumor Resection
Background CYFRA 21-1, a cytokeratin-19 fragment, is a validated serum biomarker for non-small cell lung cancer (NSCLC). However, most studies rely on single time-point measurements, limiting its specificity in differentiating malignancy from benign pulmonary conditions. Inspired by the clinical utility of serial PSA measurements in prostate cancer, we investigated whether longitudinal trends in CYFRA 21-1 could enhance diagnostic and monitoring capabilities in patients with pulmonary nodules Methods and Findings We analyzed 132 patients with pulmonary nodules, including 41 with lung cancer and 91 with benign diagnoses. CYFRA 21-1 levels were measured serially using electrochemiluminescence assays. Longitudinal trends were assessed using linear mixed-effects models to estimate biomarker trajectories. Subgroup analyses examined differences between benign, untreated cancer, and post-treatment cancer groups, as well as within-patient changes in a subset of 16 cancer patients with both pre- and post-surgical measurements. Log-transformed data were used for the analysis. At baseline, CYFRA 21-1 levels were significantly higher in malignant versus benign nodules. Over time, CYFRA trajectories diverged: benign cases showed slight increases, whereas cancer patients exhibited greater biomarker volatility. In treated cancer patients, trend of CYFRA levels on the natural log scale decline from –0.00137 pre-surgery to –0.00263 to post-surgery, and both cancer groups showed significantly higher absolute slopes than the benign group (p < 0.05). While pre- vs post-treatment slope differences did not reach significance (p = 0.211), the general pattern indicated that CYFRA 21-1 is a dynamic marker responsive to tumor presence and removal. Conclusions CYFRA 21-1 exhibits substantial within-patient variability over time, with trajectories that reflect disease state and treatment. These findings suggest that longitudinal monitoring of CYFRA 21-1—analogous to PSA velocity in prostate cancer—may offer improved diagnostic and prognostic insight in the evaluation of pulmonary nodules. Further studies in larger cohorts are warranted to validate these findings and explore clinical implementation of CYFRA trajectory analysis.
- Open Access
- 1 January 2026
- Vol. 301, 604-614
An Unsupervised Approach for Artifact Severity Scoring in Multi-Contrast MR Images
Quality assurance (QA) in magnetic resonance (MR) imaging is critical but remains a challenging and time-intensive process, particularly when working with large-scale, multi-site imaging datasets. Manual QA methods are subjective, prone to inter-rater variability, and impractical for high-throughput workflows. Existing automated QA methods often lack generalizability to diverse datasets or fail to provide interpretable insights into the causes of poor image quality. To address these limitations, we introduce an unsupervised and interpretable QA framework for multi-contrast MR images that quantifies artifact severity. By assigning a numerical score to each image, our method enables objective, consistent evaluation of image quality and highlights specific levels of artifact presence that can impair downstream analysis. Our framework employs an unsupervised contrastive learning approach, leveraging simulated artifact transformations, including random bias, noise, anisotropy, and ghosting, to train the model without requiring manual labels or preprocessing. A margin-based contrastive loss further enables differentiation between varying levels of artifact severity. We validate our framework using simulated artifacts on a public dataset and real artifacts on a private clinical dataset, demonstrating its robustness and generalizability for automatic MR image QA. By efficiently evaluating image quality and identifying artifacts prior to data processing, our approach streamlines QA workflows and enhances the reliability of subsequent analyses in both research and clinical settings.
- 1 January 2026
- Vol. 301, 604-614
Open AccessAn Unsupervised Approach for Artifact Severity Scoring in Multi-Contrast MR Images
Quality assurance (QA) in magnetic resonance (MR) imaging is critical but remains a challenging and time-intensive process, particularly when working with large-scale, multi-site imaging datasets. Manual QA methods are subjective, prone to inter-rater variability, and impractical for high-throughput workflows. Existing automated QA methods often lack generalizability to diverse datasets or fail to provide interpretable insights into the causes of poor image quality. To address these limitations, we introduce an unsupervised and interpretable QA framework for multi-contrast MR images that quantifies artifact severity. By assigning a numerical score to each image, our method enables objective, consistent evaluation of image quality and highlights specific levels of artifact presence that can impair downstream analysis. Our framework employs an unsupervised contrastive learning approach, leveraging simulated artifact transformations, including random bias, noise, anisotropy, and ghosting, to train the model without requiring manual labels or preprocessing. A margin-based contrastive loss further enables differentiation between varying levels of artifact severity. We validate our framework using simulated artifacts on a public dataset and real artifacts on a private clinical dataset, demonstrating its robustness and generalizability for automatic MR image QA. By efficiently evaluating image quality and identifying artifacts prior to data processing, our approach streamlines QA workflows and enhances the reliability of subsequent analyses in both research and clinical settings.