PROVIZ

• Sunoqrot, Mohammed R. S.; Segre, Rebecca; Nketiah, Gabriel Addio; Davik, Petter; Sjøbakk, Torill Eidhammer & Langørgen, Sverre [Show all 8 contributors for this article] (2026). Prospective validation of an AI software for detecting clinically significant prostate cancer on biparametric MRI. Insights into Imaging. 17. doi: 10.1186/s13244-025-02199-9. Full text in Research Archive Show summary

  To evaluate the feasibility and safety (primary endpoints), and performance (secondary endpoint) of a new artificial intelligence (AI) software for detecting clinically significant prostate cancer (csPCa) on biparametric MRI (bpMRI) compared to an expert radiologist. In this prospective study at St. Olavs Hospital, Norway (December 2023–October 2024), 89 consecutive biopsy-naïve men underwent bpMRI for suspected PCa. Scans were interpreted by a radiologist using PI-RADS v2.1 and a radiomics-based AI software. Biopsies were obtained from all radiologist- and/or AI-identified lesions. csPCa was defined as ISUP ≥ 2. Feasibility was defined by a < 10% software-failure rate, and safety by the absence of serious adverse device effects (SADEs). Performance was evaluated with ROC, free-response ROC, and precision-recall curves. Among 89 patients eligible for primary endpoints evaluation, the software demonstrated feasibility (7% failure rate) and safety (no SADEs). Among 76 eligible for secondary endpoint evaluation (median age 68 years [IQR: 63–73]), csPCa was found in 51% (39/76). Patient-level, software achieved an area under the ROC curve [95% CI] of 0.90 [0.83, 0.96] versus 0.86 [0.76, 0.93] (p = 0.25). At a retrospectively optimized threshold matching the radiologist’s patient-level sensitivity at PI-RADS 3 (0.92), software achieved specificity of 0.68 [0.57, 0.78] versus 0.57 [0.46, 0.68] (p = 0.29). Lesion-level, software achieved higher average precision (0.61 [0.52, 0.71] vs. 0.56 [0.46, 0.67]) and lower average false-positive per patient (0.33 [0.22, 0.43] vs. 0.41 [0.30, 0.52]) at the optimized threshold. The software was feasible and safe, and diagnostic performance showed potential to reduce unnecessary biopsies. This clinically validated artificial intelligence software enables feasible and safe detection of clinically significant prostate cancer on biparametric MRI, with demonstrated potential to reduce unnecessary biopsies and improve diagnostic accuracy, indicating potential for integration into clinical prostate cancer care.

• Davik, Petter; Elschot, Mattijs; Bathen, Tone Frost & Bertilsson, Helena Maria (2024). Repeat PSA testing improves biopsy outcome predictions in men undergoing MRI and targeted biopsies. European Urology. ISSN 0302-2838. 85. doi: 10.1016/s0302-2838(24)01295-8. Full text in Research Archive
• Hesjedal, Maria Bårdsen; Lysø, Emilie Hybertsen; Solbjør, Marit & Skolbekken, John-Arne (2024). Valuing good health care: How medical doctors, scientists and patients relate ethical challenges with artificial intelligence decision-making support tools in prostate cancer diagnostics to good health care. Sociology of Health and Illness. ISSN 0141-9889. 46, p. 1808–1827. doi: 10.1111/1467-9566.13818. Full text in Research Archive Show summary

  Artificial intelligence (AI) is increasingly used in health care to improve diagnostics and treatment. Decision-making tools intended to help professionals in diagnostic processes are developed in a variety of medical fields. Despite the imagined benefits, AI in health care is contested. Scholars point to ethical and social issues related to the development, implementation, and use of AI in diagnostics. Here, we investigate how three relevant groups construct ethical challenges with AI decision-making tools in prostate cancer (PCa) diagnostics: scientists developing AI decision support tools for interpreting MRI scans for PCa, medical doctors working with PCa and PCa patients. This qualitative study is based on participant observation and interviews with the abovementioned actors. The analysis focuses on how each group draws on their understanding of ‘good health care’ when discussing ethical challenges, and how they mobilise different registers of valuing in this process. Our theoretical approach is inspired by scholarship on evaluation and justification. We demonstrate how ethical challenges in this area are conceptualised, weighted and negotiated among these participants as processes of valuing good health care and compare their perspectives.

• Davik, Petter; Remmers, Sebastiaan; Elschot, Mattijs; Roobol, Monique J.; Bathen, Tone Frost & Bertilsson, Maria Helena (2023). Performance of magnetic resonance imaging-based prostate cancer risk calculators and decision strategies in two large European medical centres. BJU International. ISSN 1464-4096. doi: 10.1111/bju.16163. Full text in Research Archive
• Meglic, Jakob; Sunoqrot, Mohammed R. S.; Bathen, Tone Frost & Elschot, Mattijs (2023). Label-set impact on deep learning-based prostate segmentation on MRI. Insights into Imaging. 14. doi: 10.1186/s13244-023-01502-w. Full text in Research Archive Show summary

  Background Prostate segmentation is an essential step in computer-aided detection and diagnosis systems for prostate cancer. Deep learning (DL)-based methods provide good performance for prostate gland and zones segmentation, but little is known about the impact of manual segmentation (that is, label) selection on their performance. In this work, we investigated these effects by obtaining two different expert label-sets for the PROSTATEx I challenge training dataset (n = 198) and using them, in addition to an in-house dataset (n = 233), to assess the effect on segmentation performance. The automatic segmentation method we used was nnU-Net. Results The selection of training/testing label-set had a significant (p < 0.001) impact on model performance. Furthermore, it was found that model performance was significantly (p < 0.001) higher when the model was trained and tested with the same label-set. Moreover, the results showed that agreement between automatic segmentations was significantly (p < 0.0001) higher than agreement between manual segmentations and that the models were able to outperform the human label-sets used to train them. Conclusions We investigated the impact of label-set selection on the performance of a DL-based prostate segmentation model. We found that the use of different sets of manual prostate gland and zone segmentations has a measurable impact on model performance. Nevertheless, DL-based segmentation appeared to have a greater inter-reader agreement than manual segmentation. More thought should be given to the label-set, with a focus on multicenter manual segmentation and agreement on common procedures.

• Patsanis, Alexandros; Sunoqrot, Mohammed R. S.; Langørgen, Sverre; Wang, Hao; Selnæs, Kirsten Margrete & Bertilsson, Maria Helena [Show all 8 contributors for this article] (2023). A comparison of Generative Adversarial Networks for automated prostate cancer detection on T2-weighted MRI. Informatics in Medicine Unlocked (IMU). 39. doi: 10.1016/j.imu.2023.101234. Full text in Research Archive Show summary

  Generative Adversarial Networks (GANs) have shown potential in medical imaging. In this study, several previously developed GANs were investigated for prostate cancer (PCa) detection on T2-weighted (T2W) magnetic resonance images (MRI). T2W MRI from an in-house collected dataset (N=961) were used to train, validate, and test an automated computer-aided detection (CAD) pipeline. The open-access PROSTATEx training dataset (N=199) was used as an external test set. The CAD pipeline consisted of normalization, prostate segmentation, quality control, prostate gland cropping, and a GAN model. Six GANs (f-AnoGAN, HealthyGAN, StarGAN, StarGAN-v2, Fixed-Point-GAN and DeScarGAN) were evaluated for PCa detection on the patient-level using the area under the receiver operating characteristic curve (AUC). The best performing GAN (validation set) was trained with five different initializations and evaluated on the internal and external test sets to assess its robustness. Fixed-Point-GAN performed best (validation, AUC 0.76) and was selected for further assessment. The highest performance on the internal and external test sets were an AUC of 0.73 (95% CI: 0.68-0.79) and 0.77 (95% CI: 0.70-0.83), respectively. The average AUCs ± standard deviation across all runs corresponded to 0.71 ± 0.01 and 0.71 ± 0.04, respectively. Fixed-Point-GAN was identified as a promising GAN for the detection of PCa on T2W MRI. This model needs to be further investigated and trained on a larger dataset of multiparametric or biparametric MR images to assess its full potential as a support tool for radiologists.

• Patsanis, Alexandros; Sunoqrot, Mohammed R. S.; Bathen, Tone Frost & Elschot, Mattijs (2023). CROPro: a tool for automated cropping of prostate magnetic resonance images. Journal of Medical Imaging. ISSN 2329-4302. 10(2), p. 1–19. doi: 10.1117/1.JMI.10.2.024004. Full text in Research Archive Show summary

  Purpose To bypass manual data preprocessing and optimize deep learning performance, we developed and evaluated CROPro, a tool to standardize automated cropping of prostate magnetic resonance (MR) images. Approach CROPro enables automatic cropping of MR images regardless of patient health status, image size, prostate volume, or pixel spacing. CROPro can crop foreground pixels from a region of interest (e.g., prostate) with different image sizes, pixel spacing, and sampling strategies. Performance was evaluated in the context of clinically significant prostate cancer (csPCa) classification. Transfer learning was used to train five convolutional neural network (CNN) and five vision transformer (ViT) models using different combinations of cropped image sizes (64 × 64, 128 × 128, and 256 × 256 pixels2), pixel spacing (0.2 × 0.2, 0.3 × 0.3, 0.4 × 0.4, and 0.5 × 0.5 mm2), and sampling strategies (center, random, and stride cropping) over the prostate. T2-weighted MR images (N = 1475) from the online available PI-CAI challenge were used to train (N = 1033), validate (N = 221), and test (N = 221) all models. Results Among CNNs, SqueezeNet with stride cropping (image size: 128 × 128, pixel spacing: 0.2 × 0.2 mm2) achieved the best classification performance (0.678 ± 0.006). Among ViTs, ViT-H/14 with random cropping (image size: 64 × 64 and pixel spacing: 0.5 × 0.5 mm2) achieved the best performance (0.756 ± 0.009). Model performance depended on the cropped area, with optimal size generally larger with center cropping (∼40 cm2) than random/stride cropping (∼10 cm2). Conclusion We found that csPCa classification performance of CNNs and ViTs depends on the cropping settings. We demonstrated that CROPro is well suited to optimize these settings in a standardized manner, which could improve the overall performance of deep learning models.

• Dewi, Dyah Ekashanti Octorina; Sunoqrot, Mohammed R. S.; Nketiah, Gabriel Addio; Sandsmark, Elise; Giskeødegård, Guro F. & Langørgen, Sverre [Show all 9 contributors for this article] (2023). The impact of pre-processing and disease characteristics on reproducibility of T2-weighted MRI radiomics features. Magnetic Resonance Materials in Physics, Biology and Medicine. ISSN 0968-5243. doi: 10.1007/s10334-023-01112-z. Full text in Research Archive
• Sørland, Kaia Ingerdatter; Sunoqrot, Mohammed R. S.; Sandsmark, Elise; Langørgen, Sverre; Bertilsson, Helena & Trimble, Christopher George [Show all 11 contributors for this article] (2022). Pseudo-T2 mapping for normalization of T2-weighted prostate MRI. Magnetic Resonance Materials in Physics, Biology and Medicine. ISSN 0968-5243. doi: 10.1007/s10334-022-01003-9. Full text in Research Archive Show summary

  Objective Signal intensity normalization is necessary to reduce heterogeneity in T2-weighted (T2W) magnetic resonance imaging (MRI) for quantitative analysis of multicenter data. AutoRef is an automated dual-reference tissue normalization method that normalizes transversal prostate T2W MRI by creating a pseudo-T2 map. The aim of this study was to evaluate the accuracy of pseudo-T2s and multicenter standardization performance for AutoRef with three pairs of reference tissues: fat/muscle (AutoRefF), femoral head/muscle (AutoRefFH) and pelvic bone/muscle (AutoRefPB). Materials and methods T2s measured by multi-echo spin echo (MESE) were compared to AutoRef pseudo-T2s in the whole prostate (WP) and zones (PZ and TZ/CZ/AFS) for seven asymptomatic volunteers with a paired Wilcoxon signed-rank test. AutoRef normalization was assessed on T2W images from a multicenter evaluation set of 1186 prostate cancer patients. Performance was measured by inter-patient histogram intersections of voxel intensities in the WP before and after normalization in a selected subset of 80 cases. Results AutoRefFH pseudo-T2s best approached MESE T2s in the volunteer study, with no significant difference shown (WP: p = 0.30, TZ/CZ/AFS: p = 0.22, PZ: p = 0.69). All three AutoRef versions increased inter-patient histogram intersections in the multicenter dataset, with median histogram intersections of 0.505 (original data), 0.738 (AutoRefFH), 0.739 (AutoRefF) and 0.726 (AutoRefPB). Discussion All AutoRef versions reduced variation in the multicenter data. AutoRefFH pseudo-T2s were closest to experimentally measured T2s.

• Sunoqrot, Mohammed R. S.; Saha, Anindo; Hosseinzadeh, Matin; Elschot, Mattijs & Huisman, Henkjan J. (2022). Artificial intelligence for prostate MRI: open datasets, available applications, and grand challenges. European Radiology Experimental. 6:35(35), p. 1–13. doi: 10.1186/s41747-022-00288-8. Full text in Research Archive Show summary

  Artificial intelligence (AI) for prostate magnetic resonance imaging (MRI) is starting to play a clinical role for prostate cancer (PCa) patients. AI-assisted reading is feasible, allowing workflow reduction. A total of 3,369 multi-vendor prostate MRI cases are available in open datasets, acquired from 2003 to 2021 in Europe or USA at 3 T (n = 3,018; 89.6%) or 1.5 T (n = 296; 8.8%), 346 cases scanned with endorectal coil (10.3%), 3,023 (89.7%) with phased-array surface coils; 412 collected for anatomical segmentation tasks, 3,096 for PCa detection/classification; for 2,240 cases lesions delineation is available and 56 cases have matching histopathologic images; for 2,620 cases the PSA level is provided; the total size of all open datasets amounts to approximately 253 GB. Of note, quality of annotations provided per dataset highly differ and attention must be paid when using these datasets (e.g., data overlap). Seven grand challenges and commercial applications from eleven vendors are here considered. Few small studies provided prospective validation. More work is needed, in particular validation on large-scale multi-institutional, well-curated public datasets to test general applicability. Moreover, AI needs to be explored for clinical stages other than detection/characterization (e.g., follow-up, prognosis, interventions, and focal treatment).

• Davik, Petter; Remmers, Sebastiaan; Elschot, Mattijs; Roobol, Monique J; Bathen, Tone Frost & Bertilsson, Helena (2022). Reducing prostate biopsies and magnetic resonance imaging with prostate cancer risk stratification. BJUI Compass. 3(5), p. 344–353. doi: 10.1002/bco2.146. Full text in Research Archive
• Sunoqrot, Mohammed R. S.; Selnæs, Kirsten Margrete; Sandsmark, Elise; Langørgen, Sverre; Bertilsson, Helena & Bathen, Tone Frost [Show all 7 contributors for this article] (2021). The reproducibility of deep learning-based segmentation of the prostate gland and zones on t2-weighted mr images. Diagnostics (Basel). 11(9), p. 1–15. doi: 10.3390/diagnostics11091690. Full text in Research Archive Show summary

  Volume of interest segmentation is an essential step in computer-aided detection and diagnosis (CAD) systems. Deep learning (DL)-based methods provide good performance for prostate segmentation, but little is known about the reproducibility of these methods. In this work, an in-house collected dataset from 244 patients was used to investigate the intra-patient reproducibility of 14 shape features for DL-based segmentation methods of the whole prostate gland (WP), peripheral zone (PZ), and the remaining prostate zones (non-PZ) on T2-weighted (T2W) magnetic resonance (MR) images compared to manual segmentations. The DL-based segmentation was performed using three different convolutional neural networks (CNNs): V-Net, nnU-Net-2D, and nnU-Net-3D. The two-way random, single score intra-class correlation coefficient (ICC) was used to measure the inter-scan reproducibility of each feature for each CNN and the manual segmentation. We found that the reproducibility of the investigated methods is comparable to manual for all CNNs (14/14 features), except for V-Net in PZ (7/14 features). The ICC score for segmentation volume was found to be 0.888, 0.607, 0.819, and 0.903 in PZ; 0.988, 0.967, 0.986, and 0.983 in non-PZ; 0.982, 0.975, 0.973, and 0.984 in WP for manual, V-Net, nnU-Net-2D, and nnU-Net-3D, respectively. The results of this work show the feasibility of embedding DL-based segmentation in CAD systems, based on multiple T2W MR scans of the prostate, which is an important step towards the clinical implementation.

• Syversen, Ingrid Framås; Elschot, Mattijs; Sandsmark, Elise; Bertilsson, Helena; Bathen, Tone Frost & Goa, Pål Erik (2021). Exploring the diagnostic potential of adding T2 dependence in diffusion-weighted MR imaging of the prostate. PLOS ONE. 16(5). doi: 10.1371/journal.pone.0252387. Full text in Research Archive Show summary

  Background: Magnetic resonance imaging (MRI) is essential in the detection and staging of prostate cancer. However, improved tools to distinguish between low-risk and high-risk cancer are needed in order to select the appropriate treatment. Purpose: To investigate the diagnostic potential of signal fractions estimated from a two-component model using combined T2- and diffusion-weighted imaging (T2-DWI). Material and methods: 62 patients with prostate cancer and 14 patients with benign prostatic hyperplasia (BPH) underwent combined T2-DWI (TE = 55 and 73 ms, b-values = 50 and 700 s/mm^2) following clinical suspicion of cancer, providing a set of 4 measurements per voxel. Cancer was confirmed in post-MRI biopsy, and regions of interest (ROIs) were delineated based on radiology reporting. Signal fractions of the slow component (SF_slow) of the proposed two-component model were calculated from a model fit with 2 free parameters, and compared to conventional bi- and mono-exponential apparent diffusion coefficient (ADC) models. Results: All three models showed a significant difference (p<0.0001) between peripheral zone (PZ) tumor and normal tissue ROIs, but not between non-PZ tumor and BPH ROIs. The area under the receiver operating characteristics curve distinguishing tumor from prostate voxels was 0.956, 0.949 and 0.949 for the two-component, bi-exponential and mono-exponential models, respectively. The corresponding Spearman correlation coefficients between tumor values and Gleason Grade Group were fair (0.370, 0.499 and -0.490), but not significant. Conclusion: Signal fraction estimates from a two-component model based on combined T2-DWI can differentiate between tumor and normal prostate tissue and show potential for prostate cancer diagnosis. The model performed similarly to conventional diffusion models.

• Nketiah, Gabriel Addio; Elschot, Mattijs; Scheenen, Tom W.J.; Maas, Marnix C.; Bathen, Tone Frost & Selnæs, Kirsten Margrete (2021). Utility of T2-weighted MRI texture analysis in assessment of peripheral zone prostate cancer aggressiveness: a single-arm, multicenter study. Scientific Reports. 11:2085, p. 1–13. doi: 10.1038/s41598-021-81272-x. Full text in Research Archive Show summary

  T2-weighted (T2W) MRI provides high spatial resolution and tissue-specifc contrast, but it is predominantly used for qualitative evaluation of prostate anatomy and anomalies. This retrospective multicenter study evaluated the potential of T2W image-derived textural features for quantitative assessment of peripheral zone prostate cancer (PCa) aggressiveness. A standardized preoperative multiparametric MRI was performed on 87 PCa patients across 6 institutions. T2W intensity and apparent difusion coefcient (ADC) histogram, and T2W textural features were computed from tumor volumes annotated based on whole-mount histology. Spearman correlations were used to evaluate association between textural features and PCa grade groups (i.e. 1–5). Feature utility in diferentiating and classifying low-(grade group 1) vs. intermediate/high-(grade group≥ 2) aggressive cancers was evaluated using Mann–Whitney U-tests, and a support vector machine classifer employing “holdone-institution-out” cross-validation scheme, respectively. Textural features indicating image homogeneity and disorder/complexity correlated signifcantly (p< 0.05) with PCa grade groups. In the intermediate/high-aggressive cancers, textural homogeneity and disorder/complexity were signifcantly lower and higher, respectively, compared to the low-aggressive cancers. The mean classifcation accuracy across the centers was highest for the combined ADC and T2W intensitytextural features (84%) compared to ADC histogram (75%), T2W histogram (72%), T2W textural (72%) features alone or T2W histogram and texture (77%), T2W and ADC histogram (79%) combined. Texture analysis of T2W images provides quantitative information or features that are associated with peripheral zone PCa aggressiveness and can augment their lassifcation.

• Sunoqrot, Mohammed R. S.; Selnæs, Kirsten Margrete; Sandsmark, Elise; Nketiah, Gabriel Addio; Zavala-Romero, Olmo & Stoyanova, Radka [Show all 8 contributors for this article] (2020). A quality control system for automated prostate segmentation on T2-weighted MRI. Diagnostics (Basel). 10(9), p. 1–16. doi: 10.3390/diagnostics10090714. Full text in Research Archive Show summary

  Computer-aided detection and diagnosis (CAD) systems have the potential to improve robustness and efficiency compared to traditional radiological reading of magnetic resonance imaging (MRI). Fully automated segmentation of the prostate is a crucial step of CAD for prostate cancer, but visual inspection is still required to detect poorly segmented cases. The aim of this work was therefore to establish a fully automated quality control (QC) system for prostate segmentation based on T2-weighted MRI. Four different deep learning-based segmentation methods were used to segment the prostate for 585 patients. First order, shape and textural radiomics features were extracted from the segmented prostate masks. A reference quality score (QS) was calculated for each automated segmentation in comparison to a manual segmentation. A least absolute shrinkage and selection operator (LASSO) was trained and optimized on a randomly assigned training dataset (N = 1756, 439 cases from each segmentation method) to build a generalizable linear regression model based on the radiomics features that best estimated the reference QS. Subsequently, the model was used to estimate the QSs for an independent testing dataset (N = 584, 146 cases from each segmentation method). The mean ± standard deviation absolute error between the estimated and reference QSs was 5.47 ± 6.33 on a scale from 0 to 100. In addition, we found a strong correlation between the estimated and reference QSs (rho = 0.70). In conclusion, we developed an automated QC system that may be helpful for evaluating the quality of automated prostate segmentations.

• Sunoqrot, Mohammed R. S.; Nketiah, Gabriel Addio; Selnæs, Kirsten Margrete; Bathen, Tone Frost & Elschot, Mattijs (2020). Automated reference tissue normalization of T2-weighted MR images of the prostate using object recognition. Magnetic Resonance Materials in Physics, Biology and Medicine. ISSN 0968-5243. 34(2), p. 309–321. doi: 10.1007/s10334-020-00871-3. Full text in Research Archive Show summary

  Objectives To develop and evaluate an automated method for prostate T2-weighted (T2W) image normalization using dual-reference (fat and muscle) tissue. Materials and methods Transverse T2W images from the publicly available PROMISE12 (N = 80) and PROSTATEx (N = 202) challenge datasets, and an in-house collected dataset (N = 60) were used. Aggregate channel features object detectors were trained to detect reference fat and muscle tissue regions, which were processed and utilized to normalize the 3D images by linear scaling. Mean prostate pseudo T2 values after normalization were compared to literature values. Inter-patient histogram intersections of voxel intensities in the prostate were compared between our approach, the original images, and other commonly used normalization methods. Healthy vs. malignant tissue classification performance was compared before and after normalization. Results The prostate pseudo T2 values of the three tested datasets (mean ± standard deviation = 78.49 ± 9.42, 79.69 ± 6.34 and 79.29 ± 6.30 ms) corresponded well to T2 values from literature (80 ± 34 ms). Our normalization approach resulted in significantly higher (p < 0.001) inter-patient histogram intersections (median = 0.746) than the original images (median = 0.417) and most other normalization methods. Healthy vs. malignant classification also improved significantly (p < 0.001) in peripheral (AUC 0.826 vs. 0.769) and transition (AUC 0.743 vs. 0.678) zones. Conclusion An automated dual-reference tissue normalization of T2W images could help improve the quantitative assessment of prostate cancer.

View all works in NVA

• Davik, Petter; Elschot, Mattijs; Bathen, Tone Frost; Remmers, Sebastiaan; Bertilsson, Helena & Roobol, Monique J. (2023). A0058 - Comparing all published MRI prostate cancer risk calculators in a large 2-centre European cohort, European Urology, European Urology. ISSN 0302-2838. Full text in Research Archive
• Skolbekken, John-Arne (2023). Kan kunstig intelligens erstatte leger? Full text in Research Archive
• Skolbekken, John-Arne (2023). Screening for kvinner, men ikke for menn? Full text in Research Archive
• Bathen, Tone Frost (2023). Developing desicion support based on AI – from idea to clinical adoption. Full text in Research Archive
• Meglic, Jakob; Sunoqrot, Mohammed R. S.; Bathen, Tone Frost & Elschot, Mattijs (2023). Impact of label-set on the performance of the deep learning-based segmentation of the prostate gland and zones on T2-weighted MR images. Full text in Research Archive
• Solbjør, Marit; Lysø, Emilie Hybertsen; Hesjedal, Maria Bårdsen & Skolbekken, John-Arne (2023). Norwegian men’s views on PSA screening for prostate cancer. Full text in Research Archive
• Bathen, Tone Frost (2022). ProstataCAG: Forbedret diagnostikk og behandling i pakkeforløpet for prostatakreft. Full text in Research Archive
• Bathen, Tone Frost (2022). Hva er klinisk forskning – og hvorfor bør kreftpasienter delta? Full text in Research Archive
• Bathen, Tone Frost (2022). My experiences in partnering on health-related EU grant applications. Full text in Research Archive
• Bathen, Tone Frost (2022). Metabolomics – a viable approach for increased understanding of breast cancer. Full text in Research Archive
• Patsanis, Alexandros; Sunoqrot, Mohammed R. S.; Langørgen, Sverre; Wang, Hao; Selnæs, Kirsten Margrete & Bertilsson, Helena [Show all 8 contributors for this article] (2022). A comparison of Generative Adversarial Networks for Automated Prostate Cancer Detection on T2-weighted Magnetic Resonance Images. Full text in Research Archive
• Bathen, Tone Frost & Solbjør, Marit (2022). Hva er klinisk forskning – og hvorfor bør kreftpasienter delta? Full text in Research Archive
• Solbjør, Marit (2022). Norwegian men’s views on PSA screening for prostate cancer. Full text in Research Archive
• Andersen, Maria Karoline; Tessem, May-Britt; Debik, Julia; Abrahamsen, Bendik Skarre; Sørland, Kaia Ingerdatter & Giskeødegård, Guro F. [Show all 8 contributors for this article] (2022). Tumorteltet - På tokt etter kreftceller. Full text in Research Archive Show summary

  Vann er vesentlig i alt som lever, og helt essensielt i MR-teknologi. Da måler vi signalet fra vannet i kroppen og cellene våre, og får klare bilder av det som skjer under huden. Med MR-bilder og ‘tumorteltet’ vil vi vise hvordan vi ser forskjell på kreft og friske celler.

• Dewi, Dyah Ekashanti Octorina; Sunoqrot, Mohammed R. S.; Nketiah, Gabriel Addio; Sandsmark, Elise; Langørgen, Sverre & Bertilsson, Maria Helena [Show all 8 contributors for this article] (2022). Developing a Delta Radiomics Framework for Prostate Cancer Progression Biomarkers in Patients under Active Surveillance: Pilot Study. Full text in Research Archive
• Sørland, Kaia Ingerdatter; Trimble, Christopher George; Sandsmark, Elise; Bathen, Tone Frost; Elschot, Mattijs & Cloos, Martijn A. (2022). Reducing streak artefacts in radial MR fingerprinting of the prostate through automated channel removal. Full text in Research Archive
• Saha, Anindo; Twilt, Jasper J.; Bosma, Joeran S.; Ginneken, Bram van; Yakar, Derya & Elschot, Mattijs [Show all 10 contributors for this article] (2022). Artificial Intelligence and Radiologists at Prostate Cancer Detection in MRI: The PI-CAI Challenge. Full text in Research Archive
• Remmers, Sebastiaan & Davik, Petter (2022). A comparison of the PLUM and RPCRC- MRI-based prostate cancer risk prediction models in two North American and European cohorts. Full text in Research Archive Show summary

  Introduction & Objectives: Multiparametric magnetic resonance imaging (mpMRI)-based risk calculators (RC) have emerged to augment traditional risk prostate cancer (PCa) prediction models. However, few data are available comparing performance of different mpMRI-based RCs in external cohorts or between different countries. We aimed to compare the Prospective Loyola University MRI Prostate Biopsy (PLUM) RC to the Rotterdam Prostate Cancer Risk Calculator (RPCRC)-MRI (fully CE marked) in men clinical suspicion of PCa in both an American and European center. Materials & Methods: Men who were biopsy-naïve or had a prior negative biopsy receiving MRI/transrectal ultrasound fusion-guided prostate biopsy with targeted and systematic cores at centers external to both models (UAB, NTNU) were included. Primary outcomes considered were diagnosis of any PCa (≥ grade group (GG) 1) and clinically significant (cs)PCa (≥GG2). Performance of the two models was evaluated by discrimination, calibration and decision curve analysis (DCA). Results: Of 674 patients, 209/371 (56.3%) and 127/371 (34.2%) at UAB and 239/303 (55%) and 125/303 (41%) at NTNU were diagnosed with any PCa and csPCa, respectively. Discrimination of the PLUM and RPCRC-MRI was similar for any PCa (84%, 95% CI 81-87 vs. 81%, 95% CI 77-84 respectively) and csPCa (87%, 95% CI 85-90 vs. 86%, 95% CI 84-89 respectively). Both the PLUM and RPCRC-MRI were well calibrated (Fig). The DCA showed as well comparable results between the RCs above the clinical relevant thresholds (Fig). The RPCRC-MRI could avoided, with a threshold of 20% for any PCa or 4% for csPCa, 18% of all biopsies, missing 2% of all csPCa. The PLUM could avoid, with a threshold of 15% for csPCa, 35% of all biopsies, missing 6% of all csPCa https://doi.org/10.1016/S0302-2838(22)00818-1

• Tessem, May-Britt (2022). Tissue multi-omics of prostate cancer. Full text in Research Archive
• Hesjedal, Maria Bårdsen (2022). Tverrfaglighet som argument for samlokalisering. Full text in Research Archive
• Lysø, Emilie Hybertsen; Hesjedal, Maria Bårdsen; Skolbekken, John-Arne & Solbjør, Marit (2022). Norwegian men's views of AI in prostate cancer diagnostics. Full text in Research Archive
• Hesjedal, Maria Bårdsen (2022). Artificial Intelligence in Prostate Cancer Diagnostics: Transdisciplinary Collaboration in a Norwegian Biotechnology Project. Full text in Research Archive
• Hesjedal, Maria Bårdsen (2022). Hvordan tilrettelegge for tverrfaglig forskningssamarbeid? Full text in Research Archive
• Lysø, Emilie Hybertsen & Hesjedal, Maria Bårdsen (2022). AI in prostate cancer diagnostics. Full text in Research Archive
• Lysø, Emilie Hybertsen; Hesjedal, Maria Bårdsen; Skolbekken, John-Arne & Solbjør, Marit (2022). Norwegian men's views of AI in prostate cancer diagnostics. Full text in Research Archive
• Davik, Petter (2022). Reducing Prostate MRIs and biopsies with risk stratification before and after MRI. Full text in Research Archive
• Sunoqrot, Mohammed R. S.; Selnæs, Kirsten Margrete; Abrahamsen, Bendik Skarre; Patsanis, Alexandros; Nketiah, Gabriel Addio & Bathen, Tone Frost [Show all 7 contributors for this article] (2022). A deep learning-based quality control system for co-registration of prostate MR images. Full text in Research Archive
• Davik, Petter (2021). Reduksjon av MR prostata og prostatabiopsier med risikostratifisering. Full text in Research Archive Show summary

  Reduksjon av MR prostata og prostatabiopsier ved risikostratifisering: validering av ERSPC risikokalkulatorer for norsk populasjon. Davik P a, b, Remmers S c, Elschot M d, e, Roobol M b, Bathen TF d, Bertilsson H a, b a. Institutt for klinisk og molekylær medisin (IKOM), NTNU, Trondheim; b. Urologisk avdeling, St Olavs Hospital, Trondheim; c. Department of Urology, Erasmus University Medical Center, Rotterdam, The Netherlands; d. Institutt for sirkulasjon og bildediagnostikk (ISB), NTNU, Trondheim; e. Klinikk for bildediagnostikk, St Olavs Hospital, Trondheim. Bakgrunn Prostatabiopsi innebærer risiko for infeksjon samt for livstruende sepsis og død mens MR prostata er en kostbar og tidkrevende undersøkelse. Risiko-kalkulatorer basert på kliniske variabler kan predikere sannsynligheten for å finne prostatacancer ved biopsi og kan brukes til å utelate MR og/eller biopsi når sannsynligheten for cancer er lav. Kalkulatorene fra European Randomized Study of Screening for Prostate Cancer (ERSPC RC3/4) bruker Prostata Spesifikt Antigen (PSA), prostatavolum, funn fra Digital Rektal Eksaminasjon (DRE) og tidligere prostatabiopsihistorikk som uavhengige variabler i logistiske regresjonsmodeller. Disse gir sannsynligheten for klinisk signifikant cancer (csPCa) (Gleason ≥ 3+4) og all prostatakreft (PCa) og kan bidra til at 20 – 33% av biopsier unngås. Risikokalkulatorer som inkluderer MR-funn (bl.a. MRI-ERSPC-RC) bidrar til ytterligere reduksjon i biopsier, men krever MR-undersøkelse. Vi kalibrerte og re-kalibrerte ERSPC-RC 3/4 og MRI-ERSPC-RC til en norsk universitetssykehuskohort for å identifisere pasienter hvor MR og prostatabiopsier trygt kan utelates. Metode Alle menn som ble henvist med mistanke om PCa og gjennomgikk MR prostata med påfølgende prostatabiopsi mellom januar 2016 og mars 2017 ble retrospektivt identifisert og registrert i en database. Sannsynligheten for PCa og csPCa ble kalkulert med ukalibrerte kalkulatorer og sammenlignet med biopsifunn. RC-ene ble deretter rekalibrert til vår kohort og vurdert ved kalibrering, diskriminasjon med area under the curve (AUC) av receiver operating curves (ROC) og klinisk nytteverdi med Decision Curve Analyses (DCA). Resultat 303 menn ble inkludert. 79% var ikke tidligere biopsert. 55% hadde positive biopsier, 41% csPCa. AUC for ERSPC-RC 3/4 var 0.82 for csPCa og 0.77 for PCa. For MRI-ERSPC-RC var AUC 0.89 for csPCa of 0.85 for PCa. DCA viste klar nettogevinst for bruk av ERSPC RC 3/4 ved > 2 % terskelverdi for biopsi og for MRI-ERSPC-RC > 1 % terskelverdi. Ved > 10% sjanse for csPCa eller 20% sjanse for PCa terskelverdi ved bruk av ERSPC RC kunne 15.5% av MR prostata vært utelatt, 0.8% av csPCa ville da misses. Ved samme terskelverdi for bruk av MRI-ERSPC-RC kunne 24.4% av prostatabiopsier vært utelatt, 1.6% av csPCa ville da misses. Konklusjon Bruk av europeiske risikokalkulatorer rekalibrert til norske forhold kan gi betydelig reduksjon i antall MR prostata og prostatabiopsier med liten kostnad.

• Elschot, Mattijs (2021). PET/MRI in Prostate Cancer: AI for improved diagnostics. Full text in Research Archive
• Patsanis, Alexandros; Sunoqrot, Mohammed R. S.; Sandsmark, Elise; Langørgen, Sverre; Bertilsson, Helena & Selnæs, Kirsten Margrete [Show all 10 contributors for this article] (2021). Improving Prostate Cancer Detection Using Bi-parametric MRI with Conditional Generative Adversarial Networks. Full text in Research Archive
• Bathen, Tone Frost & Ola, Kindseth (2021). Brukermedvirkning i forskning. Full text in Research Archive
• Bathen, Tone Frost (2021). Kunstig intelligens i bildediagnostikk. Full text in Research Archive
• Sunoqrot, Mohammed R. S. (2021). Normalization, Segmentation, and Quality Control in Radiomics. Full text in Research Archive
• Sunoqrot, Mohammed R. S.; Selnæs, Kirsten Margrete; Sandsmark, Elise; Langørgen, Sverre; Bertilsson, Helena & Bathen, Tone Frost [Show all 7 contributors for this article] (2021). The repeatability of deep learning based segmentation of the prostate, peripheral and transition zones on T2-weighted MR images. Full text in Research Archive
• Sunoqrot, Mohammed R. S.; Selnæs, Kirsten Margrete; Abrahamsen, Bendik Skarre; Patsanis, Alexandros; Nketiah, Gabriel Addio & Bathen, Tone Frost [Show all 7 contributors for this article] (2021). A deep learning-based quality control system for co-registration of prostate MR images. Full text in Research Archive
• Patsanis, Alexandros; Sunoqrot, Mohammed R. S.; Sandsmark, Elise; Langørgen, Sverre; Bertilsson, Helena & Selnæs, Kirsten Margrete [Show all 9 contributors for this article] (2021). Prostate Cancer Detection on T2-weighted MR images with Generative Adversarial Networks. Full text in Research Archive
• Bathen, Tone Frost (2021). MRI and artificial intelligence: Can we improve prostate cancer diagnostics? Full text in Research Archive
• Bathen, Tone Frost & Nketiah, Gabriel Addio (2021). Editorial for "MRI Radiomics-Based Machine Learning for Predict of Clinically Significant Prostate Cancer in Equivocal PI-RADS 3 Lesions". JMRI. 54(5), p. 1474–1475. doi: 10.1002/jmri.27752. Full text in Research Archive
• Bathen, Tone Frost & Tessem, May-Britt (2021). Nye metoder og nye markører for sikrere diagnostisering og bedre behandling. Full text in Research Archive
• Dewi, Dyah Ekashanti Octorina; Sunoqrot, Mohammed R. S.; Nketiah, Gabriel Addio; Sandsmark, Elise; Langørgen, Sverre & Bertilsson, Helena [Show all 8 contributors for this article] (2021). Repeatability of Radiomic Features in T2-Weighted Prostate MRI: Impact of Pre-processing Configurations. Full text in Research Archive
• Dewi, Dyah Ekashanti Octorina (2021). Radiomics: An Introduction. Full text in Research Archive
• Sørland, Kaia Ingerdatter; Sunoqrot, Mohammed R. S.; Goa, Pål Erik; Sandsmark, Elise; Langørgen, Sverre & Bertilsson, Helena [Show all 9 contributors for this article] (2021). Automated reference tissue normalization of prostate T2-weighted MRI on a large, multicenter dataset. Full text in Research Archive
• Sørland, Kaia Ingerdatter; Goa, Pål Erik; Selnæs, Kirsten Margrete; Sandsmark, Elise; Trimble, Christopher George & Sunoqrot, Mohammed R. S. [Show all 8 contributors for this article] (2021). Pseudo-T2 mapping of T2-weighted MRI of the prostate: Comparison to gold standard. Full text in Research Archive
• Nketiah, Gabriel Addio & Bathen, Tone Frost (2021). Editorial for “MRI Radiomics-Based Machine Learning for Predict of Clinically Significant Prostate Cancer in Equivocal PI-RADS 3 Lesions”. Journal of Magnetic Resonance Imaging. ISSN 1053-1807. 54(5), p. 1474–1475. doi: 10.1002/jmri.27752. Full text in Research Archive
• Patsanis, Alexandros (2021). Deep Learning and Feature extraction. Full text in Research Archive
• Solbjør, Marit; Bathen, Tone Frost; Brøgger, Helga Maria & Nydal, Rune (2020). KUNSTIG INTELLIGENS I KREFTDIAGNOSTIKK - MELLOM HÅP, REALISME OG ETISKE UTFORDRINGER. Full text in Research Archive Show summary

  Åpent møte på Litteraturhuset i Trondheim. Tre populærvitenskapelige foredrag samt paneldebatt. Foredragsholdere: Tone F. Bathen, NTNU: PROVIZ – et forskningsprosjekt om MR-visualisering av prostatakreft Helga Brøgger, Leder for Norsk radiologisk forening: Vidunderlig og vanskelig- AI i radiologi Rune Nydal, NTNU og NENT: Etiske utfordringer med AI i diagnostikk Møteleder: Marit Solbjør Møtet ble gjennomført med 10 publikum i salen og 55 publikum fulgte møtet digitalt pga covid-19-situasjonen.

• Sunoqrot, Mohammed R. S.; Kucharczak, Sandra; Grajek, Magdalena; Selnæs, Kirsten Margrete; Bathen, Tone Frost & Elschot, Mattijs (2020). The repeatability of deep learning-based segmentation of the prostate on T2-weighted MR images. Full text in Research Archive
• Bathen, Tone Frost (2019). Prostate MR – improved diagnostics and treatment by use of artificial intelligence? The opportunities and challenges. Full text in Research Archive
• Bathen, Tone Frost (2019). Prostata MR: I hvilken grad kan teknologien hjelpe oss med å ta kloke valg? Full text in Research Archive

View all works in NVA