Learning from scanners: Bias reduction and feature correction in radiomics
Open accessAbstract
PURPOSE: Radiomics are quantitative features extracted from medical images. Many radiomic features depend not only on tumor properties, but also on non-tumor related factors such as scanner signal-to-noise ratio (SNR), reconstruction kernel and other image acquisition settings. This causes undesirable value variations in the features and reduces the performance of prediction models. In this paper, we investigate whether we can use phantom measurements to characterize and correct for the scanner SNR dependence. METHODS: We used a phantom with 17 regions of interest (ROI) to investigate the influence of different SNR values. CT scans were acquired with 9 different exposure settings. We developed an additive correction model to reduce scanner SNR influence. RESULTS: Sixty-two of 92 radiomic features showed high variance due to the scanner SNR. Of these 62 features, 47 showed at least a factor 2 significant standard deviation reduction by using the additive correction model. We assessed the clinical relevance of radiomics instability by using a 221 NSCLC patient cohort measured with the same scanner. CONCLUSIONS: Phantom measurements show that roughly two third of the radiomic features depend on the exposure setting of the scanner. The dependence can be modeled and corrected significantly reducing the variation in feature values with at least a factor of 2. More complex models will likely increase the correctability. Scanner SNR correction will result in more reliable radiomics predictions in NSCLC.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Features with high cross-scanner discrepancies (adjusted p < 0.05) were removed.
“Features with high cross-scanner discrepancies, defined as an adjusted p value < 0.05, were removed.”
From Radiomics-based prediction of FIGO grade for placenta accreta spectrum (2023) · cites this paper as [28]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
Scikit-learn: Machine Learning in Python
Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Müller, A., Nothman, J., Louppe, G., et al. (2012) · arXiv · doi:10.48550/arxiv.1201.0490 · Cited by 71,016 · see citing works
Radiomics: Images Are More than Pictures, They Are Data
Gillies, R. J., Kinahan, P. E., & Hricak, H. (2016) · Radiology · doi:10.1148/radiol.2015151169 · Cited by 8,421 · see citing works
Computational Radiomics System to Decode the Radiographic Phenotype
(2017) · doi:10.1158/0008-5472.can-17-0339 · Cited by 6,593 · see citing works
Decoding tumour phenotype by noninvasive imaging using a quantitative radiomics approach
Aerts, H. J. W. L., Velazquez, E. R., Leijenaar, R. T. H., Parmar, C., Grossmann, P., Carvalho, S., Bussink, J., Monshouwer, R., Haibe-Kains, B., Rietveld, D., et al. (2014) · Nature Communications · doi:10.1038/ncomms5006 · Cited by 5,132 · see citing works
Self-citationMeasuring Computed Tomography Scanner Variability of Radiomics Features
Mackin, D., Fave, X. J., Zhang, L., Fried, D., Yang, J., Taylor, B. A., Rodriguez-Rivera, E., Dodge, C. T., Jones, A. K., & Court, L. E. (2015) · Investigative Radiology · doi:10.1097/rli.0000000000000180 · Cited by 649 · see citing works
Intrinsic dependencies of CT radiomic features on voxel size and number of gray levels
Shafiq‐ul‐Hassan, M., Zhang, G. G., Latifi, K., Ullah, G., Hunt, D., Balagurunathan, Y., Abdalah, M. A., Schabath, M. B., Goldgof, D. B., Mackin, D., et al. (2017) · Medical Physics · doi:10.1002/mp.12123 · Cited by 587 · see citing works
Deep learning for lung cancer prognostication: A retrospective multi-cohort radiomics study
Hosny, A., Parmar, C., Coroller, T. P., Grossmann, P., Zeleznik, R., Kumar, A., Bussink, J., Gillies, R. J., Mak, R. H., & Aerts, H. J. W. L. (2018) · PLOS Medicine · doi:10.1371/journal.pmed.1002711 · Cited by 579 · see citing works
Self-citationSala, E., Mema, E., Himoto, Y., Veeraraghavan, H., Brenton, J. D., Snyder, A., Weigelt, B., & Vargas, H. A. (2016) · Clinical Radiology · doi:10.1016/j.crad.2016.09.013 · Cited by 369 · see citing works
Vulnerabilities of radiomic signature development: The need for safeguards
Welch, M. L., McIntosh, C., Haibe‐Kains, B., Milosevic, M., Wee, L., Dekker, A. L. A. J., Huang, S. H., Purdie, T. G., O’Sullivan, B., Aerts, H. J. W. L., et al. (2018) · Radiotherapy and Oncology · doi:10.1016/j.radonc.2018.10.027 · Cited by 320 · see citing works
Self-citationHarmonizing the pixel size in retrospective computed tomography radiomics studies
Mackin, D., Fave, X. J., Zhang, L., Yang, J., Jones, A. K., Ng, C. S., & Court, L. E. (2017) · PLoS ONE · doi:10.1371/journal.pone.0178524 · Cited by 164 · see citing works
Image biomarker standardisation initiativePreprint
Zwanenburg, A., Leger, S., Vallières, M., & Löck, S. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1612.07003 · Cited by 146 · see citing works
Effect of tube current on computed tomography radiomic features
Mackin, D., Ger, R. B., Dodge, C. T., Fave, X. J., Chi, P.-C. M., Zhang, L., Yang, J., Bache, S. T., Dodge, C., Jones, A. K., et al. (2018) · Scientific Reports · doi:10.1038/s41598-018-20713-6 · Cited by 110 · see citing works
Preliminary investigation into sources of uncertainty in quantitative imaging features
Fave, X. J., Cook, M. C., Frederick, A., Zhang, L., Yang, J., Fried, D., Stingo, F. C., & Court, L. E. (2015) · Computerized Medical Imaging and Graphics · doi:10.1016/j.compmedimag.2015.04.006 · Cited by 95 · see citing works
Abdollahi, H., Mahdavi, S. R., Mofid, B., Bakhshandeh, M., Razzaghdoust, A., Saadipoor, A., & Tanha, K. (2018) · International Journal of Radiation Biology · doi:10.1080/09553002.2018.1492756 · Cited by 51 · see citing works
Shafiq-ul-Hassan, M., Zhang, G. G., Hunt, D., Latifi, K., Ullah, G., Gillies, R. J., & Moros, E. G. (2017) · Journal of medical imaging · doi:10.1117/1.jmi.5.1.011013 · Cited by 43 · see citing works
Flagged first, then by citations
Cited by
Sorted by citations
Radiomics in medical imaging—“how-to” guide and critical reflection
van Timmeren, J. E., Cester, D., Tanadini-Lang, S., Alkadhi, H., & Baessler, B. (2020) · Insights into Imaging · doi:10.1186/s13244-020-00887-2 · Cited by 1,316 · see citing works
Ibrahim, A., Primakov, S. P., Beuque, M. P. L., Woodruff, H. C., Halilaj, I., Wu, G., Refaee, T., Granzier, R. W. Y., Widaatalla, Y., Hustinx, R., et al. (2020) · Methods · doi:10.1016/j.ymeth.2020.05.022 · Cited by 266 · see citing works
Papadimitroulas, P., Brocki, L., Christopher Chung, N., Marchadour, W., Vermet, F., Gaubert, L., Eleftheriadis, V., Plachouris, D., Visvikis, D., Kagadis, G. C., et al. (2021) · Physica Medica · doi:10.1016/j.ejmp.2021.03.009 · Cited by 226 · see citing works
“…to difference in exposure in the CT acquisition was proposed, by learning from phantom and clinical data how to model the differences, and then applying that learned correction to features values, thereby demonstrating at least 2 times standard deviation reduction for 47 out of 62 features [154].”
MethodMali, S. A., Ibrahim, A., Woodruff, H. C., Andrearczyk, V., Müller, H., Primakov, S. P., Salahuddin, Z., Chatterjee, A., & Lambin, P. (2021) · Journal of Personalized Medicine · doi:10.3390/jpm11090842 · Cited by 218 · see citing works
“Similar studies have been conducted in the past where the reproducibility of CT radiomic features was investigated using phantom data [25–27, 146 Considering MRI, a recent study using radiomics to investigate the reproducibility of features across several MRI scanners and scanning protocol…”
BackgroundMethodGao, Q., Li, Z., Zhang, J., Zhang, Y., & Shan, H. (2023) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2023.3320812 · Cited by 172 · see citing works
“This dataset contains 9 different dose scans (from 33 to 499mAs) using the Thorax protocol [48].”
MethodHarmonization strategies for multicenter radiomics investigations
Da‐ano, R., Visvikis, D., & Hatt, M. (2020) · Physics in Medicine and Biology · doi:10.1088/1361-6560/aba798 · Cited by 161 · see citing works
“…a correctability score in order to identify radiomic features biased and correctable in terms of scanner signal to noise ratio, using a commercial phantom along with the use of images of 221 NSCLC patients imaged with the same scanner as the phantom set, has been introduced (Zhovannik et al 2019).”
MethodRadiomics and artificial intelligence in lung cancer screening
Binczyk, F., Prazuch, W., Bozek, P., & Polańska, J. (2021) · Translational Lung Cancer Research · doi:10.21037/tlcr-20-708 · Cited by 150 · see citing works
“(63) have shown that more than 60 radiomic features are sensitive to the Table 2 Review of lung nodule candidate detection algorithms”
BackgroundStructural and functional radiomics for lung cancer
Wu, G., Jochems, A., Refaee, T., Ibrahim, A., Yan, C., Sanduleanu, S., Woodruff, H. C., & Lambin, P. (2021) · European Journal of Nuclear Medicine and Molecular Imaging · doi:10.1007/s00259-021-05242-1 · Cited by 138 · see citing works
Radiomics as a personalized medicine tool in lung cancer: Separating the hope from the hype
Fornacon-Wood, I., Finn, C. F., O’Connor, J. P. B., & Price, G. J. (2020) · Lung Cancer · doi:10.1016/j.lungcan.2020.05.028 · Cited by 115 · see citing works
Pfaehler, E. A. G., Zhovannik, I., Wei, L., Boellaard, R., Dekker, A. L. A. J., Monshouwer, R., Naqa, I. E., Bussink, J., Gillies, R. J., Wee, L., et al. (2021) · Physics and Imaging in Radiation Oncology · doi:10.1016/j.phro.2021.10.007 · Cited by 104 · see citing works
“Five clinical studies [14–17] and two phantom studies [13] made their image dataset publicly available or used a publicly available dataset.”
BackgroundSelf-citationRobustness of radiomic features in magnetic resonance imaging: review and a phantom study
Cattell, R. F., Chen, S., & Huang, C. (2019) · Visual Computing for Industry Biomedicine and Art · doi:10.1186/s42492-019-0025-6 · Cited by 97 · see citing works
Machine learning applications in radiation oncology
Field, M., Hardcastle, N., Jameson, M., Aherne, N. J., & Holloway, L. C. (2021) · Physics and Imaging in Radiation Oncology · doi:10.1016/j.phro.2021.05.007 · Cited by 89 · see citing works
“Physics and imaging specialists can standardise community-wide imaging acquisition protocols such that radiomic or deep learning models can have wider applicability [171,172].”
BackgroundReiazi, R., Abbas, E., Famiyeh, P., Rezaie, A., Kwan, J. Y. Y., Patel, T., Bratman, S. V., Tadic, T., Liu, F., & Haibe‐Kains, B. (2021) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2021.104400 · Cited by 85 · see citing works
Koçak, B., Kuş, E. A., & Kılıçkesmez, Ö. (2020) · European Radiology · doi:10.1007/s00330-020-07324-4 · Cited by 55 · see citing works
Ibrahim, A., Refaee, T., Primakov, S. P., Barufaldi, B., Acciavatti, R. J., Granzier, R. W. Y., Hustinx, R., Mottaghy, F. M., Woodruff, H. C., Wildberger, J. E., et al. (2021) · Cancers · doi:10.3390/cancers13081848 · Cited by 52 · see citing works
“Many studies have already reported on the sensitivity of HRFs to different factors, including (i) temporal variability, or test–retest [16,17], in which two scans of a patient (or a phantom) are taken after a time interval using the exact scanning parameters; and (ii) scanning parameters variability [11,18,19], in whi…”
BackgroundThe Emerging Role of Radiomics in COPD and Lung Cancer
Refaee, T., Wu, G., Ibrahim, A., Halilaj, I., Leijenaar, R. T. H., Rogers, W., Gietema, H. A., Hendriks, L. E. L., Lambin, P., & Woodruff, H. C. (2020) · Respiration · doi:10.1159/000505429 · Cited by 50 · see citing works
Impact of Preprocessing Parameters in Medical Imaging-Based Radiomic Studies: A Systematic Review
Trojani, V., Bassi, M. C., Verzellesi, L., & Bertolini, M. (2024) · Cancers · doi:10.3390/cancers16152668 · Cited by 47 · see citing works
Ibrahim, A., Refaee, T., Leijenaar, R. T. H., Primakov, S. P., Hustinx, R., Mottaghy, F. M., Woodruff, H. C., Maidment, A. D. A., & Lambin, P. (2021) · PLoS ONE · doi:10.1371/journal.pone.0251147 · Cited by 41 · see citing works
Zhong, J., Xia, Y., Chen, Y., Li, J., Lu, W., Shi, X., Feng, J., Yan, F., Yao, W., & Zhang, H. (2022) · European Radiology · doi:10.1007/s00330-022-09119-1 · Cited by 40 · see citing works
“Less texture features were found to be robust, since texture features relied on interpixel relationships [9, 12], which might be changed by different reconstruction algorithms.”
BackgroundRinaldi, L. M., Angelis, S. P. D., Raimondi, S., Rizzo, S., Fanciullo, C., Rampinelli, C., Mariani, M., Lascialfari, A., Cremonesi, M., Orecchia, R., et al. (2022) · European Radiology Experimental · doi:10.1186/s41747-021-00258-6 · Cited by 38 · see citing works
“According to IBSI recommendations [35], before feature computation image resampling in the axial plane (Pyradiomics B-Spline interpolator, ‘sitkBSpline’ [36, 37]) and voxel intensity discretisation (25 HU fixed bin [7, 34, 38, 39]) were applied.”
MethodLi, Y., Reyhan, M. L., Zhang⋆, Y., Wang, X., Zhou, J., Zhang, Y., Yue, N. J., & Nie, K. (2022) · Medical Physics · doi:10.1002/mp.15491 · Cited by 36 · see citing works
Kalendralis, P., Shi, Z., Traverso, A., Choudhury, A., Sloep, M., Zhovannik, I., Starmans, M. P. A., Grittner, D., Feltens, P., Monshouwer, R., et al. (2020) · Medical Physics · doi:10.1002/mp.14322 · Cited by 32 · see citing works
Self-citationLee, S. B., Cho, Y. J., Hong, Y., Jeong, D., Lee, J., Kim, S., Lee, S., & Choi, Y. H. (2021) · Investigative Radiology · doi:10.1097/rli.0000000000000839 · Cited by 32 · see citing works
“8,11,32,34,35 In our study, the texture and wavelet features showed poor reproducibility across different CT protocols.”
BackgroundImpact of image quality on radiomics applications
Cui, Y., & Yin, F.-F. (2022) · Physics in Medicine and Biology · doi:10.1088/1361-6560/ac7fd7 · Cited by 30 · see citing works
“Besides image normalization, some image preprocessing techniques such as image denoising and resampling also proved their usefulness in improving the robustness of radiomics analysis ( Sha fi q-ul-Hassan et al 2018, Zhovannik et al 2019 ) .”
Koçak, B., Yüzkan, S., Mutlu, S., Karagülle, M., Kala, A., Kadıoğlu, M., Solak, S., Sunman, Ş., Temiz, Z. H., & Ganiyusufoğlu, A. K. (2023) · Diagnostic and Interventional Radiology · doi:10.4274/dir.2023.232543 · Cited by 29 · see citing works
“For the bin width method, the following preprocessing settings were used for T1ce: 6, 13, 25, 50, and 100; for T2: 5, 11, 21, 42, and 84.”
MethodDreher, C., Linde, P., Boda‐Heggemann, J., & Baeßler, B. (2020) · Strahlentherapie und Onkologie · doi:10.1007/s00066-020-01615-x · Cited by 26 · see citing works
Chen, X., Li, Y., Zhou, Y., Yang, Y., Yang, J., Pang, P., Wang, Y., Cheng, J., Chen, H., & Guo, Y. (2022) · European Radiology · doi:10.1007/s00330-022-08541-9 · Cited by 26 · see citing works
Adversarial radiomics: the rising of potential risks in medical imaging from adversarial learning
Barucci, A., & Neri, E. (2020) · European Journal of Nuclear Medicine and Molecular Imaging · doi:10.1007/s00259-020-04879-8 · Cited by 28 · see citing works
Generative models improve radiomics reproducibility in low dose CTs: a simulation study
Chen, J., Zhang, C., Traverso, A., Zhovannik, I., Dekker, A. L. A. J., Wee, L., & Bermejo, I. (2021) · Physics in Medicine and Biology · doi:10.1088/1361-6560/ac16c0 · Cited by 21 · see citing works
Self-citationRadiomics-based prediction of FIGO grade for placenta accreta spectrum
Bartels, H. C., O’Doherty, J., Wolsztynski, E., Brophy, D. P., MacDermott, R., Atallah, D., Saliba, S., Young, C., Downey, P., Donnelly, J., et al. (2023) · European Radiology Experimental · doi:10.1186/s41747-023-00369-2 · Cited by 19 · see citing works
“As radiomic features are susceptible to variation between different MRI scanners [28], the distributions of the remaining features from each of the two MRI scanners used in this study were compared by way of a two-sample, two-sided Mann–Whitney U test [29] at the 5% significance level; p values…”
MethodSchenone, D., Dominietto, A., Campi, C., Frassoni, F., Cea, M., Aquino, S., Angelucci, E., Rossi, F., Torri, L., Bignotti, B., et al. (2021) · Diagnostics · doi:10.3390/diagnostics11101759 · Cited by 18 · see citing works
“Recent studies [35] have shown that the characteristics of the extracted features may depend on non-tumor related factors like the signal-to-noise ratio of the experimental data.”
BackgroundLysdahlgaard, S. (2022) · Radiography · doi:10.1016/j.radi.2022.03.015 · Cited by 18 · see citing works
Chu, L. C. H., Solmaz, B., Park, S., Kawamoto, S., Yuille, A., Hruban, R. H., & Fishman, E. K. (2020) · Abdominal Radiology · doi:10.1007/s00261-020-02556-w · Cited by 18 · see citing works
Gang, G. J., Deshpande, R., & Stayman, J. W. (2021) · Physics in Medicine and Biology · doi:10.1088/1361-6560/abeea5 · Cited by 18 · see citing works
“…typically involves empirically fitted relationships for individual features based on certain image properties, e.g. signal-to-noise ratio from a calibration phantom (Zhovannik et al 2019), peak frequencies andmaximum intensities of the noise power spectrum (NPS) (Shafiq-ulHassan et al 2017a).”
BackgroundZhu, L., Dong, H., Sun, J., Wang, L., Xing, Y., Hu, Y., Lu, J. A., Yang, J., Chu, J., Yan, C., et al. (2024) · European Radiology · doi:10.1007/s00330-024-10976-1 · Cited by 17 · see citing works
Resampling and harmonization for mitigation of heterogeneity in image parameters of baseline scans
Singh, A., Horng, H., Chitalia, R., Roshkovan, L., Katz, S. I., Noël, P. B., Shinohara, R. T., & Despina Kontos, B. (2022) · Scientific Reports · doi:10.1038/s41598-022-26083-4 · Cited by 16 · see citing works
Hope, A. J., Verduin, M., Dilling, T. J., Choudhury, A., Fijten, R., Wee, L., Aerts, H. J. W. L., Naqa, I. E., Mitchell, R., Vooijs, M., et al. (2021) · Cancers · doi:10.3390/cancers13102382 · Cited by 16 · see citing works
“, where radiomic features were corrected against changes in exposure values in CT phantom images [36], while Ligero et al.”
BackgroundSelf-citationImproving reproducibility and performance of radiomics in low‐dose CT using cycle GANs
Chen, J., Wee, L., Dekker, A. L. A. J., & Bermejo, I. (2022) · Journal of Applied Clinical Medical Physics · doi:10.1002/acm2.13739 · Cited by 16 · see citing works
“44 This phantom dataset is a collection of phantoms CTs by scanning a Gammex 467 CT phantom (Middleton, WI, USA) using a Philips Brilliance Big Bore CT with different doses (50 mAs, 400 mAs).”
Self-citationLinsalata, S., Borgheresi, R., Marfisi, D., Barca, P., Sainato, A., Paiar, F., Neri, E., Traino, C., & Giannelli, M. (2022) · BioMed Research International · doi:10.1155/2022/2003286 · Cited by 16 · see citing works
Chen, X., Feng, B., Xu, K., Chen, Y., Duan, X., Jin, Z., Li, K., Li, R., Long, W., & Liu, X. (2023) · European Radiology · doi:10.1007/s00330-023-09690-1 · Cited by 15 · see citing works
“Meanwhile, the calculation of manually predefined features required accurate segmentation results, and the eigenvalues maybe affected by the segmentation results [17, 18].”
Refaee, T., Salahuddin, Z., Widaatalla, Y., Primakov, S. P., Woodruff, H. C., Hustinx, R., Mottaghy, F. M., Ibrahim, A., & Lambin, P. (2022) · Journal of Personalized Medicine · doi:10.3390/jpm12040553 · Cited by 15 · see citing works
Ibrahim, A., Lu, L., Yang, H., Akın, O., Schwartz, L. H., & Zhao, B. (2022) · Applied Sciences · doi:10.3390/app12199824 · Cited by 14 · see citing works
Corso, F., Tini, G., Presti, G. L., Garau, N., Angelis, S. P. D., Bellerba, F., Rinaldi, L. M., Botta, F., Rizzo, S., Origgi, D., et al. (2021) · Cancers · doi:10.3390/cancers13123088 · Cited by 14 · see citing works
“[43], radiomic features may depend on image acquisition settings, and therefore variations in the features may cause a reduction in prediction performance.”
BackgroundAdachi, T., Nagasawa, R., Nakamura, M., Kakino, R., & Mizowaki, T. (2022) · Journal of Applied Clinical Medical Physics · doi:10.1002/acm2.13498 · Cited by 13 · see citing works
Louis, T., Lucia, F., Cousin, F., Mievis, C., Jansen, N., Duysinx, B., Pennec, R. L., Visvikis, D., Nebbache, M., Rehn, M., et al. (2024) · Scientific Reports · doi:10.1038/s41598-024-58551-4 · Cited by 12 · see citing works
Longitudinal Image Data for Outcome Modeling.
van Timmeren, J. E., Bussink, J., Koopmans, P. J., Smeenk, R., & Monshouwer, R. (2024) · Clinics in oncology · doi:10.1016/j.clon.2024.06.053 · Cited by 13 · see citing works
“Differences in acquisition and reconstruction settings can impact quantitative features derived from medical images [30,67].”
Self-citationIbrahim, A., Barufaldi, B., Refaee, T., Filho, T. M. S., Acciavatti, R. J., Salahuddin, Z., Hustinx, R., Mottaghy, F. M., Maidment, A. D. A., & Lambin, P. (2022) · Cancers · doi:10.3390/cancers14071599 · Cited by 11 · see citing works
“Nevertheless, previous studies reported large similarities between phantom and patient studies [22,39,40].”
BackgroundResultRadiomics integration into a picture archiving and communication system
Zhovannik, I., Pai, S., Santos, T. A. D. S., van Driel, L. L. G., Dekker, A. L. A. J., Fijten, R., Traverso, A., Bussink, J., & Monshouwer, R. (2021) · Physics and Imaging in Radiation Oncology · doi:10.1016/j.phro.2021.09.007 · Cited by 10 · see citing works
“However, many studies have shown that radiomic values can be influenced by specific acquisition settings [3], leading to a poor generalizability of radiomic signatures on multi-center data.”
BackgroundMethodSelf-citationBartholomeus, G. A., van Amsterdam, W. A. C., den Harder, A. M., Willemink, M. J., van Hamersvelt, R. W., de Jong, P. A., & Leiner, T. (2023) · European Radiology · doi:10.1007/s00330-023-09643-8 · Cited by 8 · see citing works
“Zhovannik et al used an additive correction model to decrease error in 47 out of 62 feature values with at least a factor of 2 [7].”
BackgroundMethodRobustness of [18F]FDG PET/CT radiomic analysis in the setting of drug-induced cardiotoxicity
Palomino-Fernández, D., Seiffert, A. P., Gómez‐Grande, A., López-Guarch, C. J., Moreno, G., Bueno, H., Aguilera, E. J. G., & González, P. S. (2023) · Computer Methods and Programs in Biomedicine · doi:10.1016/j.cmpb.2023.107981 · Cited by 7 · see citing works
Intra-scan inter-tissue variability can help harmonize radiomics features in CT
Beaumont, H., Iannessi, A., Cucchi, J., Bertrand, A., & Lucidarme, O. (2021) · European Radiology · doi:10.1007/s00330-021-08154-8 · Cited by 6 · see citing works
“Another approach by Chloe et al [21] and Zhovannik et al [22] was to identify relevant acquisition parameters, to induce reproducibility issues and build a corrective model.”
BackgroundMao, H., Feng, T., Lv, S., Shu, D., Zhu, F., Yang, J., & Zhao, Z. (2025) · European Journal of Radiology · doi:10.1016/j.ejrad.2025.112257 · Cited by 5 · see citing works
Artificial intelligence (AI) will enable improved diagnosis and treatment outcomes
Holloway, L. C., Bezak, E., & Baldock, C. (2021) · Physical and Engineering Sciences in Medicine · doi:10.1007/s13246-021-01034-x · Cited by 5 · see citing works
“AI can also be used to account for the lack of standardisation, adapting imaging data to overcome these challenges and ensuring applicability of developed AI tools to a broad range of centres and patients [19, 20].”
BackgroundRole of correlated noise in textural features extraction
Huerga, C., Morcillo, A. B., Alejo, L., Marín, A., Obesso, A., Travaglio, D., Bayón, J., Rodriguez, D., & Coronado, M. (2021) · Physica Medica · doi:10.1016/j.ejmp.2021.10.015 · Cited by 5 · see citing works
Segmentation Uncertainty Estimation as a Sanity Check for Image Biomarker Studies
Zhovannik, I., Bontempi, D., Romita, A., Pfaehler, E. A. G., Primakov, S. P., Dekker, A. L. A. J., Bussink, J., Traverso, A., & Monshouwer, R. (2022) · Cancers · doi:10.3390/cancers14051288 · Cited by 4 · see citing works
“However, a phantom set does not necessarily represent the complexity and textural patterns of biological tissues, therefore, phantom-estimated image acquisition-uncertainty should be used with precautions [7].”
BackgroundSelf-citationRadiomics in liver and pancreatic disorders: a review
Zamyatina, K. A., Godzenko, M. V., Karmazanovsky, G. G., & Revishvili, A. S. (2022) · Annaly khirurgicheskoy gepatologii = Annals of HPB surgery · doi:10.16931/1995-5464.2022-1-40-47 · Cited by 4 · see citing works
Zhang, Y., Wan, Z., Wang, D., Meng, J., Ma, F., Guo, Y., Liu, J., Li, G., & Liu, Y. (2024) · Physics in Medicine and Biology · doi:10.1088/1361-6560/ad3c91 · Cited by 4 · see citing works
Jurišica, I. (2022) · Springer eBooks · doi:10.1007/978-3-031-00119-2_13 · Cited by 3 · see citing works
Are all shortcuts in encoder–decoder networks beneficial for CT denoising?
Chen, J., Zhang, C., Wee, L., Dekker, A. L. A. J., & Bermejo, I. (2022) · Computer Methods in Biomechanics and Biomedical Engineering Imaging & Visualization · doi:10.1080/21681163.2022.2044908 · Cited by 3 · see citing works
“…200 mAs and 50 mAs from those scanned with 400 mAs, we first measured the noise intensity introduced in images with lower doses by scanning a Gammex 467 CT phantom (Middleton, WI, USA) using a Philips Brilliance Big Bore CT with different doses (400 mAs, 200mAs and 50 mAs) (Zhovannik et al. 2019).”
MethodSelf-citationKalendralis, P., Shi, Z., Traverso, A., Choudhury, A., Sloep, M., Zhovannik, I., Starmans, M. P. A., Grittner, D., Feltens, P., Monshouwer, R., et al. (2020) · Medical Physics · doi:10.1002/mp.14805 · Cited by 3 · see citing works
Self-citationZhang, J., Xu, W., Lu, W., Lin, Y., He, L., Lv, Z., Nuerrula, Y., Zouting, Zhang, Z., Zhang, H., et al. (2025) · Med Research · doi:10.1002/mdr2.70051 · Cited by 2 · see citing works
Chen, J., Zeng, H., Cheng, Y., & Yang, B. (2024) · Medical Physics · doi:10.1002/mp.17266 · Cited by 2 · see citing works
“43 However,a thorough examination revealed that the images in the I-SPY 1 and I-SPY 2 datasets were scanned from different manufacturers.”
Radiomics in clinical trials: perspectives on standardization
Nie, K., & Xiao, Y. (2022) · Physics in Medicine and Biology · doi:10.1088/1361-6560/aca388 · Cited by 2 · see citing works
Felicetti, F., Costanzo, S., Carnì, D. L., & Lamonaca, F. (2025) · Measurement Sensors · doi:10.1016/j.measen.2024.101656 · Cited by 1 · see citing works
Singh, A., Horng, H., Roshkovan, L., Hershman, M., Shinohara, R. T., Katz, S. I., & Despina Kontos, B. (2022) · SPIE Medical Imaging: Computer-Aided Diagnosis · doi:10.1117/12.2611467
Artificial intelligence applications in radiotherapy
Kalendralis, P. (2022) · doi:10.26481/dis.20221010pk · Cited by 1 · see citing works
Self-citationRefaee, T. (2022) · doi:10.26481/dis.20220712tr
Generative models improve radiomics
Chen, J. (2023) · doi:10.26481/dis.20230703jc
Radiomics: “unlocking the potential of medical images for precision radiation oncology”
Kalendralis, P., Vallières, M., Kann, B. H., Aneja, S., Rashid, A. S., Dekker, A. L. A. J., & Fijten, R. (2024) · Elsevier eBooks · doi:10.1016/b978-0-12-822000-9.00009-4
Self-citationScapicchio, C., Imbriani, M., Lizzi, F., Quattrocchi, M., Retico, A., Saponaro, S., Tenerani, M. I., Tofani, A., Zafaranchi, A., & Fantacci, M. E. (2024) · Biomedical Physics & Engineering Express · doi:10.1088/2057-1976/ad41e7
Sweetland, K., Duddalwar, V. A., Varghese, B. A., & Hwang, D. (2024) · International Symposium on Medical Information Processing and Analysis (SIPAIM) · doi:10.1109/sipaim62974.2024.10783514
Chen, J., Cheng, Y., Chen, L., & Yang, B. (2025) · Journal of Applied Clinical Medical Physics · doi:10.1002/acm2.70061
“36 Consequently, it is imperative to mitigate the influence of scanner bias on predictions.”
Rinaldi, L. M., Angelis, S. P. D., Raimondi, S., Rizzo, S., Fanciullo, C., Rampinelli, C., Mariani, M., Lascialfari, A., Cremonesi, M., Orecchia, R., et al. (2022) · Figshare · doi:10.6084/m9.figshare.19028904
Additional file 1 of Radiomics-based prediction of FIGO grade for placenta accreta spectrum
Bartels, H. C., O’Doherty, J., Wolsztynski, E., Brophy, D. P., MacDermott, R., Atallah, D., Saliba, S., Young, C., Downey, P., Donnelly, J. C., et al. (2024) · Figshare · doi:10.6084/m9.figshare.26615047
- Authors
- Zhovannik, I., Bussink, J., Traverso, A., Shi, Z., Kalendralis, P., Wee, L., Dekker, A., Fijten, R., & Monshouwer, R.
- Affiliations
- Maastricht University Medical Centre (NL)
- Radboud University Medical Center (NL)
- Corresponding author
- Zhovannik, Ivan
- ORCID
- Venue
- Clinical and Translational Radiation Oncology
- Publisher
- Elsevier BV
- Published
- 2019-11
- Volume
- 19
- Pages
- 33-38
- ISSN
- 2405-6308 (ISSN-L)
- PMC
- PMC6690665
- OpenAlex
- W2958287596
- Cited by
- 75
- References
- 14
- Citations by year
- 2025: 4 · 2024: 10 · 2023: 7 · 2022: 22 · 2021: 19 · 2020: 12 · 2019: 1
- FWCI
- 8.91
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- Clinical and Translational Radiation Oncology (publishedVersion, cc-by-nc-nd)
- PubMed Central (publishedVersion, cc-by-nc-nd)
- PubMed (publishedVersion)
- Research Publications (Maastricht University) (publishedVersion, cc-by-nc-nd)
- Radboud Repository (Radboud University) (submittedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion)
- Archivio Istituzionale della Ricerca - Università Vita-Salute San Raffaele (submittedVersion)
- Radboud Repository (Radboud University) (submittedVersion, other-oa)
- Research Publications (Maastricht University) (submittedVersion, other-oa)
- Funders
- Nederlandse Organisatie voor Wetenschappelijk Onderzoek
- Topics
- Health Sciences › Medicine › Radiology, Nuclear Medicine and Imaging › Radiomics and Machine Learning in Medical Imaging
- Health Sciences › Medicine › Radiology, Nuclear Medicine and Imaging › Medical Imaging Techniques and Applications
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Image Retrieval and Classification Techniques
- Keywords
- radiomics, bias correction, phantom measurements, non-small cell lung cancer, reconstruction kernels
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
