Assessing the clinical performance of artificial intelligence software for prostate cancer detection on MRI

Penzkofer, Tobias; Padhani, Anwar R.; Türkbey, Barış; Ahmed, Hashim U.

doi:10.1007/s00330-022-08609-6

Cited by 6 publications

(2 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the case of multiple lesions, the non-index lesions are more likely to be less apparent at institution B that thus harbor sPC less frequently and are often replaceable by the index-lesion with regard to the overall exam-level diagnosis. Thus, a lower fraction of the individual lesions require to be correctly identified by DL to achieve the same exam-level sensitivity as in cohort A. Lesion-level analysis is important for assessing the performance of AI systems [ 28 ], which however requires to be considered in the context of the exam-level diagnosis. The Dice-score itself does not necessarily determine successful lesion detection, as different raters often agree on at least a portion of a lesion, as reported with Dice-scores of 0.48–0.52 [ 29 ].…”

Section: Discussionmentioning

confidence: 99%

Application of a validated prostate MRI deep learning system to independent same-vendor multi-institutional data: demonstration of transferability

Netzer,

Eith,

Bethge

et al. 2023

Eur Radiol

View full text Add to dashboard Cite

Objectives To evaluate a fully automatic deep learning system to detect and segment clinically significant prostate cancer (csPCa) on same-vendor prostate MRI from two different institutions not contributing to training of the system. Materials and methods In this retrospective study, a previously bi-institutionally validated deep learning system (UNETM) was applied to bi-parametric prostate MRI data from one external institution (A), a PI-RADS distribution-matched internal cohort (B), and a csPCa stratified subset of single-institution external public challenge data (C). csPCa was defined as ISUP Grade Group ≥ 2 determined from combined targeted and extended systematic MRI/transrectal US-fusion biopsy. Performance of UNETM was evaluated by comparing ROC AUC and specificity at typical PI-RADS sensitivity levels. Lesion-level analysis between UNETM segmentations and radiologist-delineated segmentations was performed using Dice coefficient, free-response operating characteristic (FROC), and weighted alternative (waFROC). The influence of using different diffusion sequences was analyzed in cohort A. Results In 250/250/140 exams in cohorts A/B/C, differences in ROC AUC were insignificant with 0.80 (95% CI: 0.74–0.85)/0.87 (95% CI: 0.83–0.92)/0.82 (95% CI: 0.75–0.89). At sensitivities of 95% and 90%, UNETM achieved specificity of 30%/50% in A, 44%/71% in B, and 43%/49% in C, respectively. Dice coefficient of UNETM and radiologist-delineated lesions was 0.36 in A and 0.49 in B. The waFROC AUC was 0.67 (95% CI: 0.60–0.83) in A and 0.7 (95% CI: 0.64–0.78) in B. UNETM performed marginally better on readout-segmented than on single-shot echo-planar-imaging. Conclusion For same-vendor examinations, deep learning provided comparable discrimination of csPCa and non-csPCa lesions and examinations between local and two independent external data sets, demonstrating the applicability of the system to institutions not participating in model training. Clinical relevance statement A previously bi-institutionally validated fully automatic deep learning system maintained acceptable exam-level diagnostic performance in two independent external data sets, indicating the potential of deploying AI models without retraining or fine-tuning, and corroborating evidence that AI models extract a substantial amount of transferable domain knowledge about MRI-based prostate cancer assessment. Key Points • A previously bi-institutionally validated fully automatic deep learning system maintained acceptable exam-level diagnostic performance in two independent external data sets. • Lesion detection performance and segmentation congruence was similar on the institutional and an external data set, as measured by the weighted alternative FROC AUC and Dice coefficient. • Although the system generalized to two external institutions without re-training, achieving expected sensitivity and specificity levels using the deep learning system requires probability thresholds to be adjusted, underlining the importance of institution-specific calibration and quality control.

show abstract

Section: Discussionmentioning

confidence: 99%

Application of a validated prostate MRI deep learning system to independent same-vendor multi-institutional data: demonstration of transferability

Netzer,

Eith,

Bethge

et al. 2023

Eur Radiol

View full text Add to dashboard Cite

show abstract

“…Although experience and education are invaluable in interpreting the findings, it is still not always possible to distinguish benign from malignant changes with absolute certainty. Therefore, numerous studies are focused on finding a non-invasive method of distinguishing prostate cancer from benign causes of false positive MRI findings [68,[72][73][74]118,[124][125][126][127].…”

Section: Biparametric or Multipametric Mri-future Perspectivementioning

confidence: 99%

Dynamic Contrast-Enhanced Study in the mpMRI of the Prostate—Unnecessary or Underutilised? A Narrative Review

Guljaš,

Dupan Krivdić,

Drežnjak Madunić

et al. 2023

Diagnostics

View full text Add to dashboard Cite

The aim of this review is to summarise recent scientific literature regarding the clinical use of DCE-MRI as a component of multiparametric resonance imaging of the prostate. This review presents the principles of DCE-MRI acquisition and analysis, the current role of DCE-MRI in clinical practice with special regard to its role in presently available categorisation systems, and an overview of the advantages and disadvantages of DCE-MRI described in the current literature. DCE-MRI is an important functional sequence that requires intravenous administration of a gadolinium-based contrast agent and gives information regarding the vascularity and capillary permeability of the lesion. Although numerous studies have confirmed that DCE-MRI has great potential in the diagnosis and monitoring of prostate cancer, its role is still inadequate in the PI-RADS categorisation. Moreover, there have been numerous scientific discussions about abandoning the intravenous application of gadolinium-based contrast as a routine part of MRI examination of the prostate. In this review, we summarised the recent literature on the advantages and disadvantages of DCE-MRI, focusing on an overview of currently available data on bpMRI and mpMRI, as well as on studies providing information on the potential better usability of DCE-MRI in improving the sensitivity and specificity of mpMRI examinations of the prostate.

show abstract

Artificial intelligence algorithms aimed at characterizing or detecting prostate cancer on MRI: How accurate are they when tested on independent cohorts? – A systematic review

Rouvière

Jaouen

Baseilhac

et al. 2023

Diagnostic and Interventional Imaging

View full text Add to dashboard Cite

Assessing the clinical performance of artificial intelligence software for prostate cancer detection on MRI

Cited by 6 publications

References 13 publications

Application of a validated prostate MRI deep learning system to independent same-vendor multi-institutional data: demonstration of transferability

Application of a validated prostate MRI deep learning system to independent same-vendor multi-institutional data: demonstration of transferability

Dynamic Contrast-Enhanced Study in the mpMRI of the Prostate—Unnecessary or Underutilised? A Narrative Review

Artificial intelligence algorithms aimed at characterizing or detecting prostate cancer on MRI: How accurate are they when tested on independent cohorts? – A systematic review

Contact Info

Product

Resources

About