Pith. sign in

REVIEW 3 major objections 5 minor 43 references

Generalizing imaging biomarker repeatability studies using Bayesian inference: Applications in detecting heterogeneous treatment response in whole-body diffusion-weighted MRI of metastatic prostate cancer

T0 review · 3 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read The paper proposes a Bayesian framework that generalizes imaging-biomarker repeatability studies to non-normal biomarkers, and reports an ~70% individual-tumor response rate in two mCRPC cohorts.

desk verdict Genuinely new simplex-biomarker repeatability model and a clean BA-equivalence proof, but the ~70% response-rate claim leans on unvalidated transfer of measurement error across scanner protocols. read the letter →

arxiv 2505.09197 v1 pith:2OYCR4A2 submitted 2025-05-14 stat.AP eess.IV

Matthew D Blackledge , Konstantinos Zormpas-Petridis , Ricardo Donners , Antonio Candito , David J Collins , Johann de Bono , Chris Parker , Dow-Mu Koh
show 1 more author
Nina Tunariu
This is my paper · ORCID
classification stat.APeess.IV
keywords BayesianinferenceimagingbiomarkersrepeatabilityDirichlet-Multinomialhabitatmappingdiffusion-weightedMRImetastaticprostatecancerresponseheterogeneity
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper proposes a Bayesian framework for judging whether an imaging biomarker changed between two scans, replacing the usual assumption that biomarkers are single, normally distributed numbers. It frames each lesion's response as a choice between a null model (no true change beyond measurement noise) and an alternative model (change), and derives posterior odds for each lesion from a mixture model. Applied to whole-body diffusion-weighted MRI of metastatic castrate-resistant prostate cancer, the method classifies individual bone lesions and reports that roughly 70% of tumors responded, while the rest showed no evidence of change or changed in the opposite direction. The authors' aim is to make repeatability assessment, and per-lesion response classification, available for any biomarker type, including habitat proportions on the unit simplex.

What carries the argument

The load-bearing object is the posterior predictive distribution under the null model M0, written as p(y|y0, yb, M0), approximated by Hamiltonian Monte Carlo samples and turned into a credible region by a kernel-distance criterion. Posterior odds PO10 come from a Bayesian mixture model over lesion labels (change versus no change), with a mixing weight lambda that estimates the population response rate. For habitat biomarkers the likelihood is a Dirichlet-Multinomial model on unnormalised voxel counts, with baseline proportions fixed at (0.1, 0.8, 0.1) from ADC percentiles; this is what lets the framework handle bounded, compositional biomarker data.

What would settle it

Acquire true test-retest data using the exact acquisition parameters of Study 1 and Study 2 (same b-values, in-plane resolution, and slice thickness), fit the same null model, and compare the resulting posterior predictive credible regions with those derived from Study 3. If the regions differ materially, or if lesions known by an independent clinical or biological marker to be stable are classified as responders at a rate above the nominal level, the transfer of measurement error from Study 3 is the breaking point.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is that conventional repeatability analysis is a special case of Bayesian novelty detection: with normally distributed biomarkers and intraclass correlation tending to one, the posterior predictive null distribution reduces to the classical Bland-Altman result, so the Bayesian credible region and the Repeatability Coefficient coincide. Once this is in place, the same machinery extends to arbitrary biomarker likelihoods. For habitat biomarkers, which are counts of voxels falling into ADC-defined tissue classes, the paper models the null and alternative cases with Dirichlet-Multinomial distributions and estimates posterior odds for every lesion via Bayesian mixture modeling. In two mCRPC treatment studies, the posterior odds reveal inter- and intra-lesion heterogeneity and yield an approximately 70% response rate among individual tumors.

Load-bearing premise

The classifications and the ~70% response rate depend on using Study 3's double-baseline data (10 patients, 73 lesions) as the measurement-error model for Studies 1 and 2, even though the acquisition protocols differ; if the noise in the treatment studies is not the same, every posterior odds estimate is miscalibrated.

Editorial extensions

If this is right

  • Repeatability studies no longer need to assume real-valued normal biomarkers; any biomarker with a defined likelihood, such as counts, proportions, or bounded values, can get a credible region and a per-subject response decision.
  • Classical Bland-Altman repeatability coefficients and the Student-t form with Nb degrees of freedom emerge as limiting cases, so the framework is backward-compatible with existing precision results.
  • Posterior odds computed from the mixture model give a per-lesion classifier and a population-level response-rate estimate; in the two mCRPC studies this rate is about 70%.
  • Because posterior odds can exceed the 95% specificity ceiling of p-value-based testing, lesion classifications can in principle be more specific than conventional thresholding.
  • Habitat-style simplex biomarkers, previously without a repeatability method, can be monitored in the same decision framework.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • An implication the paper leaves implicit is that the 70% figure is a posterior summary of the two cohorts, not a universal response rate; applying it to a new trial requires re-estimating the measurement-error and mixing parameters for that trial's protocol.
  • Because the paper notes that including post-treatment data changed the habitat measurement-precision estimate in Study 1, a simulation study varying the baseline proportions and precision priors would show how sensitive the 70% and per-lesion labels are to model misspecification.
  • A natural next step, which the paper lists as future work, is to add per-patient response-rate parameters lambda_i; one could then correlate those per-patient posterior distributions with survival endpoints such as progression-free or overall survival.
  • The same anomaly-detection machinery could be applied to other constrained biomarkers, such as fat fraction bounded on [0,1] or count-based radiomic features, by substituting the appropriate likelihood and rerunning the repeatability study.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. The paper proposes a Bayesian framework for imaging-biomarker repeatability that extends conventional Bland-Altman-style analysis to biomarkers that are not univariate normal, in particular to Dirichlet-Multinomial distributed 'habitat' proportions on the simplex. The authors derive an analytical link between Bayesian anomaly detection and the conventional repeatability coefficient, propose posterior odds (PO10) via Bayesian mixture modeling to classify individual lesions as responding or not, and apply the framework to whole-body diffusion-weighted MRI data from two mCRPC treatment cohorts (Studies 1 and 2) using a separate double-baseline cohort (Study 3) to estimate measurement error. They report an approximately 70% response rate among individual tumors across both studies, based on the posterior median of the mixing weight λ. The manuscript includes simulation studies, a prior sensitivity analysis, and Stan code in the supplementary material.

Significance. If the methodological and clinical claims are valid, the paper provides a genuinely useful generalization of repeatability assessment to non-normal, simplex-valued imaging biomarkers, with a principled Bayesian alternative to Bland-Altman for such data. The theoretical derivations in the supplement are internally consistent, and the simulation study demonstrates low bias and good coverage in most parameter regions, which supports the computational implementation (modulo the Stan-code issues noted below). The framework also gives a concrete, interpretable way to quantify inter-lesion response heterogeneity, which is clinically relevant to mCRPC. However, the key clinical quantity -- the 70% response rate -- rests on an untested assumption about the transferability of measurement error across cohorts, and the printed Stan code for the habitat model is not syntactically executable as written; these issues currently limit the reliability and reproducibility of the central claims.

major comments (3)
  1. [Sections 3.2 and 2.5; Figure 11] The transferability of the double-baseline measurement error from Study 3 to Studies 1 and 2 is a load-bearing premise for the per-lesion posterior odds and the reported ~70% response rate. The three studies differ in slice thickness (6 mm vs 5 mm), in-plane resolution, b-value sampling, and lesion volume (median 5.1 ml vs 12.5 and 22.0 ml), yet Section 2.5 only asserts that the double-baseline data 'should be representative' without providing evidence or a sensitivity analysis. Moreover, the paper's own prior-sensitivity analysis in Section 4.3 (Figure 11) shows that for habitat biomarkers in Study 1, the posterior of the measurement precision τ shifts when post-treatment data are included, which the authors attribute to 'some misalignment between the assumed model and the acquired post-treatment data.' The recommendation to rely on repeat-baseline-only estimates does not resolve the issue because those estimates come from a different cohort. I ask the authors to provide a direct assessment of how the posterior odds and λ estimates would change under plausible perturbations of τ (e.g., a sensitivity interval over τ), or to present an empirical demonstration that measurement repeatability is similar across the three acquisition protocols.
  2. [Section 7.4 (Supplementary Stan code)] The Stan code for the Dirichlet-Multinomial habitat model is syntactically malformed and, as printed, cannot be executed. The custom function `dir_mult` is declared with a 2D array argument (`array[,] int y`) and separate `vector mu` and `real prec` arguments, but in both the model block and the generated-quantities block it is called with `yp[n]`, which is a 1D array of length K, and with a single expression `mu0 * prec` (a vector) rather than separate `mu` and `prec` arguments. Additionally, the likelihood statement `yb ~ dirichlet_multinomial(mu0, prec);` does not reference the custom `dir_mult_lpmf`; if it relies on Stan's built-in `dirichlet_multinomial` distribution, the custom function is redundant and the inconsistency still prevents the code from running as written. Because the habitat results are a central contribution, the authors must correct the code and confirm that the reported results are reproducible from the provided model code.
  3. [Abstract and Section 4.2] The headline 'approximately 70% response rate among individual tumors across both studies' is stated without any uncertainty quantification. The posterior distribution of λ is shown in Figure 9, but the text reports no numerical estimate, credible interval, or posterior predictive check for this quantity. Given that the 70% figure is a posterior median based on the mixture model, the authors should report the posterior median and 95% credible interval for λ for each study and each biomarker type, and should moderate the claim if the posterior is wide or if the credible interval includes values far from 0.7.
minor comments (5)
  1. [Throughout] The manuscript contains numerous typographical and spelling errors, including 'Hamiltonean', 'disitributed', 'paramters', 'lections', 'seperated' (Figure 7 caption), 'respoonse', and 'depcted'. A careful proofread is needed before resubmission.
  2. [Section 4.1 / Figure 6] The caption of Figure 6 states 'median ADC in this instance' even though the figure describes the Dirichlet-Multinomial habitat model; this is misleading and should be corrected.
  3. [Section 4.2 / Figure 7] The color-coding description in the text refers to bins defined by posterior odds values, but the last bin is labeled 'BF10 > 10 (dark green)' while the rest are labeled with PO10; for consistency, all bins should be labeled as PO10.
  4. [Section 2.6] The assumption that the baseline habitat proportion is known and fixed at μ0 = (0.1, 0.8, 0.1) is ad hoc; please add a sentence acknowledging this as a modeling assumption and, if possible, report a sensitivity analysis or at least describe how deviations from this prior value would affect inference.
  5. [Table 5] In the simulation study, the measurement precision τ is fixed to 11.54 and the voxel-size parameters are fixed to values derived from the clinical studies; because the transferability of τ across cohorts is a central concern (see Major Comment 1), the simulation design should include a grid over τ or otherwise acknowledge this limitation.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: theoretical derivations are self-contained, simulation studies are computational self-consistency checks, and the cited self-works are contextual rather than load-bearing.

full rationale

The paper's derivation chain is internally coherent. The equivalence between the Bayesian posterior predictive and Bland-Altman analysis in Section 2.3.1 is obtained from an explicit normal-normal model with stated priors, and the Bayes-factor expression in Section 2.4.1 is derived from the assumed generative distributions, not assumed from the result. The mixture-model posterior odds in Section 2.5 are estimated from the likelihoods of the fitted models; neither the per-lesion classifications nor the approximately 70% response rate is a parameter fitted to one subset and then reported as a prediction of a closely related quantity. The simulation studies generate data from the same models and assess parameter recovery; this is a computational self-consistency check, and the paper explicitly frames it as demonstrating that inference is achievable for well-defined models, not as external validation. Self-citations, such as Blackledge et al. (2017) for habitat mapping and Thrussell et al. (2022) for the IMS naming, provide context or terminology rather than carrying the central statistical argument, and no uniqueness theorem is imported to force the chosen model. The internal caveat in Section 4.3 about possible misalignment of the habitat measurement precision and the use of Study 3 double-baseline data for Studies 1 and 2 is a measurement-assumption or external-validity concern, not a circular step; it does not reduce any derived quantity to its own input. Accordingly, no circular step meeting the required evidentiary standard was found.

Assumptions & free parameters 5 free parameters · 5 assumptions · 0 invented entities

The central claim rests on standard Bayesian modeling assumptions plus two domain-specific assumptions: the transferability of the repeat-baseline noise estimate across studies with different acquisition parameters, and the Dirichlet-Multinomial form for habitat counts. The prior widths are hand-chosen and not fully reported, which affects reproducibility but not the theoretical derivations.

free parameters (5)
  • Prior width sigma_mu for median ADC model = not stated in main text
    The Stan code takes mu_prior as data, but the actual value used in the clinical analyses is not reported. Prior sensitivity analysis varies it (sigma_mu <= 0.1 mm2/s), implying a specific fixed value for the main fit.
  • Prior width gamma_sigma for median ADC model = not stated in main text
    The Stan code takes sd_prior as data, but the value is not given in the paper. The posterior estimates of sigma0, sigmaDelta, and sigma depend on this choice.
  • Prior width gamma_tau for habitat model = not stated in main text
    The DM Stan code takes prec_prior as data, but the value is not reported. The posterior of precision tau may depend on this choice, as shown in the prior sensitivity analysis.
  • Baseline habitat proportions mu0 = (0.1, 0.8, 0.1)
    Chosen by fixing 10th and 90th ADC percentiles as thresholds. This defines the habitat biomarker and is treated as a known quantity in the Dirichlet-Multinomial model.
  • Number of components in Dirichlet mixture density approximation = 5
    Used to approximate the posterior predictive density for novelty detection credible regions. Chosen by hand without sensitivity analysis.
assumptions (5)
  • domain assumption Measurement error for real biomarkers is zero-mean normal and homoskedastic
    Used throughout Section 2, especially in Theorem 7.1 and the Bland-Altman equivalence result. The paper notes this assumption may fail for bounded biomarkers but uses it for the working example.
  • domain assumption Change in a real biomarker under M1 is independent of baseline value and normally distributed
    Table 1 and the Stan mixture model in Section 7.3 use the difference dp = yp2 - yp1 with mean mud and variance sdd^2 + 2*sdr^2, ignoring baseline value. The paper mentions Markovian extensions but does not implement them.
  • ad hoc to paper Habitat counts follow a Dirichlet-Multinomial distribution with a simplex baseline mean mu0 = (0.1, 0.8, 0.1)
    This is the core model for intra-tumoral heterogeneity (Section 2.6, Table 3). No empirical validation is provided that voxel counts in ADC percentile bins follow this distribution.
  • domain assumption The double-baseline cohort is representative of measurement error in the treatment cohorts
    Section 2.5 states this requirement and Section 3.2 applies Study 3 data to Studies 1 and 2. The studies differ in acquisition parameters, so this is a strong, untested premise.
  • domain assumption Lesions are statistically independent
    The likelihood in Eq. 8 and the Stan code assume independent lesions. The authors acknowledge this in Section 5 and propose future hierarchical models, but the reported response rate and posterior odds rely on it.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Generalizing imaging biomarker repeatability studies using Bayesian inference: Applications in detecting heterogeneous treatment response in whole-body diffusion-weighted MRI of metastatic prostate cancer." pith.science (2026). https://pith.science/paper/2OYCR4A2

@misc{pith2026250509197,
  author       = {Pith},
  title        = {Pith review of: Generalizing imaging biomarker repeatability studies using Bayesian inference: Applications in detecting heterogeneous treatment response in whole-body diffusion-weighted MRI of metastatic prostate cancer},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/2OYCR4A2}},
  note         = {Machine review of arXiv:2505.09197}
}
read the original abstract

The assessment of imaging biomarkers is critical for advancing precision medicine and improving disease characterization. Despite the availability of methods to derive disease heterogeneity metrics in imaging studies, a robust framework for evaluating measurement uncertainty remains underdeveloped. To address this gap, we propose a novel Bayesian framework to assess the precision of disease heterogeneity measures in biomarker studies. Our approach extends traditional methods for evaluating biomarker precision by providing greater flexibility in statistical assumptions and enabling the analysis of biomarkers beyond univariate or multivariate normally-distributed variables. Using Hamiltonian Monte Carlo sampling, the framework supports both, for example, normally-distributed and Dirichlet-Multinomial distributed variables, enabling the derivation of posterior distributions for biomarker parameters under diverse model assumptions. Designed to be broadly applicable across various imaging modalities and biomarker types, the framework builds a foundation for generalizing reproducible and objective biomarker evaluation. To demonstrate utility, we apply the framework to whole-body diffusion-weighted MRI (WBDWI) to assess heterogeneous therapeutic responses in metastatic bone disease. Specifically, we analyze data from two patient studies investigating treatments for metastatic castrate-resistant prostate cancer (mCRPC). Our results reveal an approximately 70% response rate among individual tumors across both studies, objectively characterizing differential responses to systemic therapies and validating the clinical relevance of the proposed methodology. This Bayesian framework provides a powerful tool for advancing biomarker research across diverse imaging-based studies while offering valuable insights into specific clinical applications, such as mCRPC treatment response.

Figures

Figures reproduced from arXiv: 2505.09197 by the authors.

Figure 1
Figure 1. Bayesian networks of two competing models for biomarker measurements made and before, [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Example 95% credible region (red contour) for 5000 samples from a 2-dimensional normal distribution, computed using the kernel distance, D (blue surface). MCMC samples that are found to be outside the credible regions are color-coded blue, whilst those inside are color-coded black. Triangles represent example data from a post-treatment measurement; the green triangle shows a case that would be considered a significa… view at source ↗
Figure 3
Figure 3. Variation in the expected Bayes factor for real, normally-distributed biomarker measurements. [PITH_FULL_IMAGE:figures/full_fig_p009_3.png] view at source ↗
Figures from the paper (9 more)
Figure 4
Figure 4. Figure 4: Our approach to habitat mapping for post-treatment assessment consists (from left to right) of: (i) Delineating [PITH_FULL_IMAGE:figures/full_fig_p011_4.png]
Figure 5
Figure 5. Figure 5: Simulation results for real-valued biomarkers (median ADC in this instance). The [PITH_FULL_IMAGE:figures/full_fig_p014_5.png]
Figure 6
Figure 6. Figure 6: Simulation results for dirichlet-multinomial (habitat) biomarkers (median ADC in this instance). All results [PITH_FULL_IMAGE:figures/full_fig_p015_6.png]
Figure 7
Figure 7. Figure 7: Example results for five patient evaluated in this study. We have presented a Maximum Intensity Projection [PITH_FULL_IMAGE:figures/full_fig_p016_7.png]
Figure 8
Figure 8. Figure 8: Barycenric coordinate plots of all lesions within both studies. For each lesion, its derived posterior odds is [PITH_FULL_IMAGE:figures/full_fig_p017_8.png]
Figure 9
Figure 9. Figure 9: Kernel density plots of the posterior distributions for the mixing weight [PITH_FULL_IMAGE:figures/full_fig_p018_9.png]
Figure 10
Figure 10. Figure 10: A comparison of log-P O10 derived for median ADC within each lesion and for habitats defined for each region. Note that our estimation method for P O10 can lead to inifite values. These were capped to 80,000 and not included in estimation of r 2 or the p-value. Vertic…
Figure 11
Figure 11. Figure 11: Prior sensitivity analysis for Study 1. The primary prior parameter influencing the posterior distributions [PITH_FULL_IMAGE:figures/full_fig_p019_11.png]
Figure 12
Figure 12. Figure 12: Prior sensitivity analysis for Study 2. The only prior parameter that had significant impact on posterior [PITH_FULL_IMAGE:figures/full_fig_p031_12.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

43 extracted references · 33 canonical work pages

  1. [1]

    Theorem 7.2

    B(A+1) where the necessary conditionsA> −1 and Re(B)> 0 are guaranteed by the data, we arrive at p(d|db)∝  d2 + NbX j=1 d2 bj   − Nb+1 2 = d2 +Nbd2 b − Nb+1 2 ∝ d2/d2 b Nb + 1 !− Nb+1 2 where d2 b = 1 Nb NbX i=1 d2 bj We notice from its functional form that this is a Student t distribution St(t;ν) = Γ ν+1 2 √νπΓ ν 2 t2 ν + 1 −ν+1 2 witht2 =d2/d2 b and...

    work page 2010

  2. [6]

    Armando Manduca, Philip J Bayly, Richard L Ehman, Arunark Kolipaka, Thomas J Royston, Ingolf Sack, Ralph Sinkus, and Bernard E Van Beers

    doi:10.18632/oncotarget.17752. Armando Manduca, Philip J Bayly, Richard L Ehman, Arunark Kolipaka, Thomas J Royston, Ingolf Sack, Ralph Sinkus, and Bernard E Van Beers. Mr elastography: Principles, guidelines, and terminology. Magn Reson Med, 85 (5):2377–2390, May

    work page doi:10.18632/oncotarget.17752

  3. [13]

    Sarah R Amend and Kenneth J Pienta

    doi:10.1038/nrc.2017.69. Sarah R Amend and Kenneth J Pienta. Ecology meets cancer biology: the cancer swamp promotes the lethal cancer phenotype. Oncotarget, 6(12):9669–78,

    work page doi:10.1038/nrc.2017.69 2017

  4. [14]

    E Sala, E Mema, Y Himoto, H Veeraraghavan, J D Brenton, A Snyder, B Weigelt, and H A Vargas

    doi:10.18632/oncotarget.3430. E Sala, E Mema, Y Himoto, H Veeraraghavan, J D Brenton, A Snyder, B Weigelt, and H A Vargas. Unravelling tumour heterogeneity using next-generation imaging: radiomics, radiogenomics, and habitat imaging. Clin Radiol, 72(1): 3–10, Jan

    work page doi:10.18632/oncotarget.3430

  5. [15]

    doi:10.1016/j.crad.2016.09.013. Mireia Crispin-Ortuzar, Marcel Gehrung, Stephan Ursprung, Andrew B Gill, Anne Y Warren, Lucian Beer, Ferdia A Gallagher, Thomas J Mitchell, Iosif A Mendichovszky, Andrew N Priest, Grant D Stewart, Evis Sala, and Florian Markowetz. Three-dimensional printed molds for image-guided surgical biopsies: An open source computation...

    work page doi:10.1016/j.crad.2016.09.013 2016

  6. [17]

    doi:10.1007/s00330-020-07560-8. Konstantinos Zormpas-Petridis, Evon Poon, Matthew Clarke, Neil P Jerome, Jessica K R Boult, Matthew D Blackledge, Fernando Carceller, Alexander Koers, Giuseppe Barone, Andrew D J Pearson, Lucas Moreno, John Anderson, Neil Sebire, Kieran McHugh, Dow-Mu Koh, Louis Chesler, Yinyin Yuan, Simon P Robinson, and Yann Jamin. Noninv...

    doi:10.1007/s00330-020-07560-8

  7. [18]

    Camilla Panico, Giacomo Avesani, Konstantinos Zormpas-Petridis, Leonardo Rundo, Camilla Nero, and Evis Sala

    doi:10.1158/0008-5472.CAN-20-0133. Camilla Panico, Giacomo Avesani, Konstantinos Zormpas-Petridis, Leonardo Rundo, Camilla Nero, and Evis Sala. Radiomics and radiogenomics of ovarian cancer: Implications for treatment monitoring and clinical management. Radiol Clin North Am , 61(4):749–760, Jul

    work page doi:10.1158/0008-5472.can-20-0133

  8. [19]

    Nancy A Obuchowski

    doi:10.1016/j.rcl.2023.02.006. Nancy A Obuchowski. Interpreting change in quantitative imaging biomarkers. Acad Radiol, 25(3):372–379, Mar

    work page doi:10.1016/j.rcl.2023.02.006 2023

Show all 43 references
  1. [20]

    doi:10.1016/j.acra.2017.09.023

    ISSN 1878-4046 (Electronic); 1076-6332 (Print); 1076-6332 (Linking). doi:10.1016/j.acra.2017.09.023. Nancy A. Obuchowski and Andrew J. Buckler. Estimating the precision of quantitative imaging biomarkers without test-retest studies. Academic Radiology , 29(4):543–549,

    2017 doi

  2. [22]

    doi:10.1148/radiol.2017161965. Jessica M Winfield, Aisha B Miah, Dirk Strauss, Khin Thway, David J Collins, Nandita M deSouza, Martin O Leach, Veronica A Morgan, Sharon L Giles, Eleanor Moskovic, Andrew Hayes, Myles Smith, Shane H Zaidi, Daniel Henderson, and Christina Messiou...

    doi

  3. [25]

    doi:10.1200/JCO.2015.64.2702

    ISSN 15277755. doi:10.1200/JCO.2015.64.2702. Matthew D. Blackledge, David J. Collins, Nina Tunariu, Matthew R. Orton, Anwar R. Padhani, Martin O. Leach, and Dow Mu Koh. Assessment of treatment response by total tumor volume and global apparent diffusion coefficient using diffu...

    2015 doi

  4. [27]

    doi:10.1016/j.eururo.2016.05.033

    ISSN 18737560. doi:10.1016/j.eururo.2016.05.033. Anne M. Euser, Friedo W. Dekker, and Saskia le Cessie. A practical approach to bland-altman plots and variation coefficients for log transformed variables. Journal of Clinical Epidemiology, 61(10):978–982,

    2016 doi

  5. [29]

    23 Generalizing imaging biomarker repeatability studies using Bayesian inference Andrew Gelman and Donald B Rubin

    doi:10.1016/j.acra.2022.08.031. 23 Generalizing imaging biomarker repeatability studies using Bayesian inference Andrew Gelman and Donald B Rubin. Inference from iterative simulation using multiple sequences. Statistical science, 7(4):457–472,

    2022 doi

  6. [30]

    Michael D Lee and Eric-Jan Wagenmakers

    doi:10.1016/j.jmp.2017.09.005. Michael D Lee and Eric-Jan Wagenmakers. Bayesian cognitive modeling: A practical course . Cambridge university press,

    2017 doi

  7. [31]

    doi:10.3389/fonc.2022.899180. Dow-Mu Koh, Matthew Blackledge, David J Collins, Anwar R Padhani, Toni Wallace, Benjamin Wilton, N Jane Taylor, J James Stirling, Rajesh Sinha, Pat Walicke, Martin O Leach, Ian Judson, and Paul Nathan. Reproducibility and changes in the apparent d...

    2022

  8. [32]

    Matthew David Blackledge, DM Koh, David J Collins, Erica Scurr, Julie Hughes, M Leach, et al

    doi:10.1007/s00330-009- 1469-4. Matthew David Blackledge, DM Koh, David J Collins, Erica Scurr, Julie Hughes, M Leach, et al. Assessing response heterogeneity following radium 223 administration using whole body diffusion weighted mri. In International Society for Magnetic Res...

    doi

  9. [34]

    Computer-aided detection of focal bone metastases from whole-body multi-modal mri

    Jakub Ceranka, Frédéric Lecouvet, Johan De Mey, and Jef Vandemeulebroucke. Computer-aided detection of focal bone metastases from whole-body multi-modal mri. In Medical Imaging 2020: Computer-Aided Diagnosis , volume 11314, pages 174–180. SPIE,

    2020

  10. [35]

    doi:10.1007/s00330-022-08536-6. Lianghui Zhu, Huijuan Shi, Huiting Wei, Chengjiang Wang, Shanshan Shi, Fenfen Zhang, Renao Yan, Yiqing Liu, Tingting He, Liyuan Wang, Junru Cheng, Hufei Duan, Hong Du, Fengjiao Meng, Wenli Zhao, Xia Gu, Linlang Guo, Yingpeng Ni, Yonghong He, Tia...

    doi

  11. [36]

    doi:10.1016/j.ebiom.2022.104426. Lauren Brady, Michelle Kriner, Ilsa Coleman, Colm Morrissey, Martine Roudier, Lawrence D True, Roman Gulati, Stephen R Plymate, Zoey Zhou, Brian Birditt, Rhonda Meredith, Gary Geiss, Margaret Hoang, Joseph Beechem, and Peter S Nelson. Inter- an...

    2022

  12. [37]

    doi:10.1038/s41467-021-21615-4. 24 Generalizing imaging biomarker repeatability studies using Bayesian inference Raquel Perez-Lopez, Joaquin Mateo, Helen Mossop, Matthew D Blackledge, David J Collins, Mihaela Rata, Veronica A Morgan, Alison Macdonald, Shahneen Sandhu, David Lo...

    doi

  13. [38]

    doi:10.1148/radiol.2016160646. Raquel Perez-Lopez, David Lorente, Matthew D Blackledge, David J Collins, Joaquin Mateo, Diletta Bianchini, Aurelius Omlin, Andrea Zivi, Martin O Leach, Johann S de Bono, Dow-Mu Koh, and Nina Tunariu. V olume of bone metastasis assessed with whol...

    doi

  14. [39]

    Fei Tony Liu, Kai Ming Ting, and Zhi-Hua Zhou

    doi:10.1148/radiol.2015150799. Fei Tony Liu, Kai Ming Ting, and Zhi-Hua Zhou. Isolation forest. In 2008 eighth ieee international conference on data mining, pages 413–422. IEEE,

    2008 doi

  15. [41]

    Hania Paverd, Konstantinos Zormpas-Petridis, Hannah Clayton, Sarah Burge, and Mireia Crispin-Ortuzar

    doi:10.1038/s41598-022-18986-z. Hania Paverd, Konstantinos Zormpas-Petridis, Hannah Clayton, Sarah Burge, and Mireia Crispin-Ortuzar. Radiology and multi-scale data integration for precision oncology. NPJ Precision Oncology, 8(1):158,

    doi

  16. [1986]

    doi:10.1016/S0140-6736(86)90837-8

    ISSN 0140-6736. doi:10.1016/S0140-6736(86)90837-8. Howard I. Scher, Michael J. Morris, Walter M. Stadler, Celestia Higano, Ethan Basch, Karim Fizazi, Emmanuel S. Antonarakis, Tomasz M. Beer, Michael A. Carducci, Kim N. Chi, Paul G. Corn, Johann S. De Bono, Robert Dreicer, Dani...

    doi

  17. [2001]

    Stanislav Pidhorskyi, Ranya Almohsen, and Gianfranco Doretto

    doi:10.1162/089976601750264965. Stanislav Pidhorskyi, Ranya Almohsen, and Gianfranco Doretto. Generative probabilistic novelty detection with adversarial autoencoders. Advances in neural information processing systems , 31,

    doi

  18. [2004]

    Matthew D Blackledge, David J Collins, Dow-Mu Koh, and Martin O Leach

    doi:10.1007/s10278-004-1014-6. Matthew D Blackledge, David J Collins, Dow-Mu Koh, and Martin O Leach. Rapid development of image analysis research tools: bridging the gap between researcher and clinician with pyosirix. Computers in biology and medicine , 69:203–212,

    doi

  19. [2006]

    Dow-Mu Koh and David J Collins

    doi:10.1102/1470-7330.2006.0021. Dow-Mu Koh and David J Collins. Diffusion-weighted mri in the body: applications and challenges in oncology. AJR Am J Roentgenol, 188(6):1622–35, Jun

    2006

  20. [2007]

    Alexey Surov, Hans Jonas Meyer, and Andreas Wienke

    doi:10.2214/AJR.06.1403. Alexey Surov, Hans Jonas Meyer, and Andreas Wienke. Correlation between apparent diffusion coefficient (adc) and cellularity is different in several tumors: a meta-analysis. Oncotarget, 8(35):59492–59499, Aug

    doi

  21. [2009]

    doi:10.1016/J.EJCA.2008.10.026

    ISSN 1879-0852. doi:10.1016/J.EJCA.2008.10.026. URL https://pubmed.ncbi.nlm.nih. gov/19097774/. Elizabeth M Charles-Edwards and Nandita M deSouza. Diffusion-weighted magnetic resonance imaging and its application to cancer. Cancer Imaging, 6(1):135–43, Sep

    2008

  22. [2010]

    25 Generalizing imaging biomarker repeatability studies using Bayesian inference Supplementary Material 7.1 Mathematical derivations Theorem 7.1

    doi:10.1016/j.cogpsych.2009.12.001. 25 Generalizing imaging biomarker repeatability studies using Bayesian inference Supplementary Material 7.1 Mathematical derivations Theorem 7.1. If y∼N (x0,σ ) y0∼N (x0,σ ) x0∼N (µ0,σ

    2009 doi

  23. [2012]

    Parvinder Sujlana, Jan Skrok, and Laura M Fayad

    doi:10.1007/s00330-012-2446-x. Parvinder Sujlana, Jan Skrok, and Laura M Fayad. Review of dynamic contrast-enhanced mri: Technical aspects and applications in the musculoskeletal system. J Magn Reson Imaging , 47(4):875–890, Apr

    doi

  24. [2014]

    doi:10.1371/journal.pone.0091779

    ISSN 19326203. doi:10.1371/journal.pone.0091779. Anwar R. Padhani, Frederic E. Lecouvet, Nina Tunariu, Dow Mu Koh, Frederik De Keyzer, David J. Collins, Evis Sala, Heinz Peter Schlemmer, Giuseppe Petralia, H. Alberto Vargas, Stefano Fanti, H. Bertrand Tombal, and Johann de Bon...

    doi

  25. [2015]

    doi:10.1016/j.ccell.2014.12.001. Carlo C Maley, Athena Aktipis, Trevor A Graham, Andrea Sottoriva, Amy M Boddy, Michalina Janiszewska, Ariosto S Silva, Marco Gerlinger, Yinyin Yuan, Kenneth J Pienta, Karen S Anderson, Robert Gatenby, Charles Swanton, David Posada, Chung-I Wu, ...

    2014 doi

  26. [2016]

    doi:10.1148/radiol.2015151169. James P B O’Connor, Eric O Aboagye, Judith E Adams, Hugo J W L Aerts, Sally F Barrington, Ambros J Beer, Ronald Boellaard, Sarah E Bohndiek, Michael Brady, Gina Brown, David L Buckley, Thomas L Chenevert, Laurence P Clarke, Sandra Collette, Gary ...

    doi

  27. [2017]

    doi:10.1038/nrclinonc.2016.162. E. A. Eisenhauer, P. Therasse, J. Bogaerts, L. H. Schwartz, D. Sargent, R. Ford, J. Dancey, S. Arbuck, S. Gwyther, M. Mooney, L. Rubinstein, L. Shankar, L. Dodd, R. Kaplan, D. Lacombe, and J. Verweij. New response evaluation criteria in solid tu...

    2016 doi

  28. [2018]

    doi:10.1002/jmri.25810

    ISSN 1522-2586 (Electronic); 1053-1807 (Linking). doi:10.1002/jmri.25810. Emily Hoffmann, Max Masthoff, Wolfgang G Kunz, Max Seidensticker, Stefanie Bobe, Mirjam Gerwing, Wolfgang E Berdel, Christoph Schliemann, Cornelius Faber, and Moritz Wildgruber. Multiparametric mri for c...

    doi

  29. [2019]

    doi:10.3389/fonc.2019.00280. J. Martin Bland and Douglas G. Altman. Statistical methods for assessing agreement between two methods of clinical measurement. The Lancet, 327:307–310, 2

    2019

  30. [2020]

    doi:10.1200/CCI.20.00026. Lucian Beer, Paula Martin-Gonzalez, Maria Delgado-Ortet, Marika Reinius, Leonardo Rundo, Ramona Woitek, Stephan Ursprung, Lorena Escudero, Hilal Sahin, Ionut-Gabriel Funingana, Joo-Ern Ang, Mercedes Jimenez-Linan, Tristan Lawton, Gaurav Phadke, Sally ...

    doi

  31. [2021]

    Ingolf Sack

    doi:10.1002/mrm.28627. Ingolf Sack. Magnetic resonance elastography from fundamental soft-tissue mechanics to diagnostic imaging. Na- ture Reviews Physics, 5(1):25–42,

    doi

  32. [2022]

    doi:https://doi.org/10.1016/j.acra.2021.06.009

    ISSN 1076-6332. doi:https://doi.org/10.1016/j.acra.2021.06.009. URL https://www.sciencedirect.com/science/article/ pii/S1076633221002798. Jessica M Winfield, Nina Tunariu, Mihaela Rata, Keiko Miyazaki, Neil P Jerome, Michael Germuska, Matthew D Blackledge, David J Collins, Joh...

    2021 doi

  33. [2023]

    URL https://doi.org/10.1038/ s42254-022-00543-2

    doi:10.1038/s42254-022-00543-2. URL https://doi.org/10.1038/ s42254-022-00543-2 . M O Leach, B Morgan, P S Tofts, D L Buckley, W Huang, M A Horsfield, T L Chenevert, D J Collins, A Jackson, D Lomas, B Whitcher, L Clarke, R Plummer, I Judson, R Jones, R Alonzi, T Brunner, D M K...

    doi

  34. [2024]

    Nicholas McGranahan and Charles Swanton

    doi:10.1038/s41571-024-00891-1. Nicholas McGranahan and Charles Swanton. Biological and therapeutic impact of intratumor heterogeneity in cancer evolution. Cancer Cell, 27(1):15–26, Jan

    doi

  35. [4356]

    URL https://www.sciencedirect.com/science/ article/pii/S0895435607004131

    doi:https://doi.org/10.1016/j.jclinepi.2007.11.003. URL https://www.sciencedirect.com/science/ article/pii/S0895435607004131. Nancy A Obuchowski, Erich Huang, Nandita M deSouza, David Raunig, Jana Delfino, Andrew Buckler, Charles Hatt, Xiaofeng Wang, Chaya Moskowitz, Alexander...

    2007 doi

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.