Comparison of PI-RADS Versions 2.0 and 2.1 for MRI-based Calculation of the Prostate Volume

Deep Learning Prostate MRI Segmentation Accuracy and Robustness: A Systematic Review

Purpose To investigate the accuracy and robustness of prostate segmentation using deep learning across various training data sizes, MRI vendors, prostate zones, and testing methods relative to fellowship-trained diagnostic radiologists. Materials and Methods In this systematic review, Embase, PubMed, Scopus, and Web of Science databases were queried for English-language articles using keywords and related terms for prostate MRI segmentation and deep learning algorithms dated to July 31, 2022. A total of 691 articles from the search query were collected and subsequently filtered to 48 on the basis of predefined inclusion and exclusion criteria. Multiple characteristics were extracted from selected studies, such as deep learning algorithm performance, MRI vendor, and training dataset features. The primary outcome was comparison of mean Dice similarity coefficient (DSC) for prostate segmentation for deep learning algorithms versus diagnostic radiologists. Results Forty-eight studies were included. Most published deep learning algorithms for whole prostate gland segmentation (39 of 42 [93%]) had a DSC at or above expert level (DSC ≥ 0.86). The mean DSC was 0.79 ± 0.06 (SD) for peripheral zone, 0.87 ± 0.05 for transition zone, and 0.90 ± 0.04 for whole prostate gland segmentation. For selected studies that used one major MRI vendor, the mean DSCs of each were as follows: General Electric (three of 48 studies), 0.92 ± 0.03; Philips (four of 48 studies), 0.92 ± 0.02; and Siemens (six of 48 studies), 0.91 ± 0.03. Conclusion Deep learning algorithms for prostate MRI segmentation demonstrated accuracy similar to that of expert radiologists despite varying parameters; therefore, future research should shift toward evaluating segmentation robustness and patient outcomes across diverse clinical settings. Keywords: MRI, Genital/Reproductive, Prostate Segmentation, Deep Learning Systematic review registration link: osf.io/nxaev © RSNA, 2024.

PI-RADS v2.1 evaluation of prostate "nodule in nodule" variants: clinical, imaging, and pathological features

OBJECTIVES

To analyze the correlation among the imaging features of prostate "nodule in nodule," clinical prostate indices, and pathology results.

METHODS

We retrospectively analyzed the prostate images from 47 male patients who underwent MRI scans and pathological biopsy from January 2022 to July 2023. Two radiologists (R1/R2) evaluated the morphology and signal intensity of the "nodule in nodule" in a double-blind manner and calculated the PI-RADS v2.1 score, which was compared with clinical prostate indices and pathological results.

RESULTS

34.04% (16/47) of patients were pathologically diagnosed with clinically significant prostate cancer (csPCa). Total prostate-specific antigen (tPSA), free/t PSA, PSA density (PSAD), and prostate gland volume (PGV) were significantly different between csPCa patients and benign prostatic hyperplasia (BPH) patients with prostate "nodule in nodule". R1/R2 detected 17/17 prostate "nodule in nodule" pathologically confirmed as csPCa on MRI; 10.60% (16/151) (R1) and 11.11% (17/153) (R2) had diffusion-weighted imaging (DWI) PI-RADS v2.1 score of 4, and 0.66% (1/151) (R1) had a score of 3. The percentages of encapsulated, circumscribed, and atypical nodules and obscured margins were 0.00% (0/151), 0.00% (0/151), 5.96% (9/151), and 5.30% (8/151), respectively, for R1, and 0.00% (0/153), 0.00% (0/153), 5.88% (9/153), and 4.58% (7/153) for R2.

CONCLUSION

When the inner nodules of "nodule in nodule" lesions in PI-RADS v2.1 category 1 in the TZ show incomplete capsulation or obscured margins, they are considered atypical nodules and might be upgraded to PI-RADS v2.1 category 3 if they exhibit marked diffusion restriction. However, further validation is needed.

CRITICAL RELEVANCE STATEMENT

This study first analyzed the relationship between clinical and pathological findings and the size, margin, and multimodal MRI manifestations of the prostate "nodule in nodule." These findings could improve the diagnostic accuracy of PI-RADS v2.1 for prostate lesions.

KEY POINTS

• The margin of the prostate inner nodules affects the PI-RADS v2.1 score. • The morphology of prostate "nodule in nodule" is related to their pathology. • The PI-RADS v2.1 principle requires consideration of prostate "nodule in nodule" variants.

Deep learning algorithm performs similarly to radiologists in the assessment of prostate volume on MRI

OBJECTIVES

Prostate volume (PV) in combination with prostate specific antigen (PSA) yields PSA density which is an increasingly important biomarker. Calculating PV from MRI is a time-consuming, radiologist-dependent task. The aim of this study was to assess whether a deep learning algorithm can replace PI-RADS 2.1 based ellipsoid formula (EF) for calculating PV.

METHODS

Eight different measures of PV were retrospectively collected for each of 124 patients who underwent radical prostatectomy and preoperative MRI of the prostate (multicenter and multi-scanner MRI's 1.5 and 3 T). Agreement between volumes obtained from the deep learning algorithm (PV_DL) and ellipsoid formula by two radiologists (PV_EF1 and PV_EF2) was evaluated against the reference standard PV obtained by manual planimetry by an expert radiologist (PV_MPE). A sensitivity analysis was performed using a prostatectomy specimen as the reference standard. Inter-reader agreement was evaluated between the radiologists using the ellipsoid formula and between the expert and inexperienced radiologists performing manual planimetry.

RESULTS

PV_DL showed better agreement and precision than PV_EF1 and PV_EF2 using the reference standard PV_MPE (mean difference [95% limits of agreement] PV_DL: -0.33 [-10.80; 10.14], PV_EF1: -3.83 [-19.55; 11.89], PV_EF2: -3.05 [-18.55; 12.45]) or the PV determined based on specimen weight (PV_DL: -4.22 [-22.52; 14.07], PV_EF1: -7.89 [-30.50; 14.73], PV_EF2: -6.97 [-30.13; 16.18]). Inter-reader agreement was excellent between the two experienced radiologists using the ellipsoid formula and was good between expert and inexperienced radiologists performing manual planimetry.

CONCLUSION

Deep learning algorithm performs similarly to radiologists in the assessment of prostate volume on MRI.

KEY POINTS

• A commercially available deep learning algorithm performs similarly to radiologists in the assessment of prostate volume on MRI. • The deep-learning algorithm was previously untrained on this heterogenous multicenter day-to-day practice MRI data set.

Prostate gland volume estimation: anteroposterior diameters measured on axial versus sagittal ultrasonography and magnetic resonance images

PURPOSE

The aim of this study was to evaluate the accuracy of prostate volume estimates calculated from the ellipsoid formula using the anteroposterior (AP) diameter measured on axial and sagittal images obtained through ultrasonography (US) and magnetic resonance imaging (MRI).

METHODS

This retrospective study included 456 patients with transrectal US and MRI from two university hospitals. Two radiologists independently measured the prostate gland diameters on US and MRI: AP diameters on axial and sagittal images, transverse, and longitudinal diameters on midsagittal images. The volume estimates, volumeax and volumesag, were calculated from the ellipsoid formula by using the AP diameter on axial and sagittal images, respectively. The prostate volume extracted from MRI-based whole-gland segmentation was considered the gold standard. The intraclass correlation coefficient (ICC) was used to evaluate the inter-method agreement between volumeax and volumesag, and agreement with the gold standard. The Wilcoxon signedrank test was used to analyze the differences between the volume estimates and the gold standard.

RESULTS

The prostate gland volume estimates showed excellent inter-method agreement, and excellent agreement with the gold standard (ICCs >0.9). Compared with the gold standard, the volume estimates were significantly larger on MRI and significantly smaller on US (P<0.001). The volume difference (segmented volume-volume estimate) was greater in patients with larger prostate glands, especially on US.

CONCLUSION

Volumeax and volumesag showed excellent inter-method agreement and excellent agreement with the gold standard on both US and MRI. However, prostate volume was overestimated on MRI and underestimated on US.

Evaluation of transabdominal and transperineal ultrasound-derived prostate specific antigen (PSA) density and clinical utility compared to MRI prostate volumes: A feasibility study

Purpose

To investigate the accuracy of surface-based ultrasound-derived PSA-density (US-PSAD) versus gold-standard MRI-PSAD as a risk-stratification tool.

Methods

Single-centre prospective study of patients undergoing MRI for suspected prostate cancer (PCa). Four combinations of US-volumes were calculated using transperineal (TP) and transabdominal (TA) views, with triplanar measurements to calculate volume and US-PSAD. Intra-class correlation coefficient (ICC) was used to compare US and MRI volumes. Categorical comparison of MRI-PSAD and US-PSAD was performed at PSAD cut-offs <0.15, 0.15–0.20, and >0.20 ng/mL² to assess agreement with MRI-PSAD risk-stratification decisions.

Results

64 men were investigated, mean age 69 years and PSA 7.0 ng/mL. 36/64 had biopsy-confirmed prostate cancer (18 Gleason 3+3, 18 Gleason ≥3+4). Mean MRI-derived gland volume was 60 mL, compared to 56 mL for TA-US, and 65 mL TP-US. ICC demonstrated good agreement for all US volumes with MRI, with highest agreement for transabdominal US, followed by combined TA/TP volumes. Risk-stratification decisions to biopsy showed concordant agreement between triplanar MRI-PSAD and ultrasound-PSAD in 86–91% and 92–95% at PSAD thresholds of >0.15 ng/mL² and >0.12 ng/mL², respectively. Decision to biopsy at threshold >0.12 ng/mL², demonstrated sensitivity ranges of 81–100%, specificity 85–100%, PPV 86–100% and NPV 83–100%. Transabdominal US provided optimal sensitivity of 100% for this clinical decision, with specificity 85%, and transperineal US provided optimal specificity of 100%, with sensitivity 87%.

Conclusion

Transperineal-US and combined TA-TP US-derived PSA density values compare well with standard MRI-derived values and could be used to provide accurate PSAD at presentation and inform the need for further investigations.

Prostate volume prediction on MRI: tools, accuracy and variability

OBJECTIVE

A reliable estimation of prostate volume (PV) is essential to prostate cancer management. The objective of our multi-rater study was to compare intra- and inter-rater variability of PV from manual planimetry and ellipsoid formulas.

METHODS

Forty treatment-naive patients who underwent prostate MRI were selected from a local database. PV and corresponding PSA density (PSAd) were estimated on 3D T2-weighted MRI (3 T) by 7 independent radiologists using the traditional ellipsoid formula (TEF), the newer biproximate ellipsoid formula (BPEF), and the manual planimetry method (MPM) used as ground truth. Intra- and inter-rater variability was calculated using the mixed model-based intraclass correlation coefficient (ICC).

RESULTS

Mean volumes were 67.00 (± 36.61), 66.07 (± 35.03), and 64.77 (± 38.27) cm³ with the TEF, BPEF, and MPM methods, respectively. Both TEF and BPEF overestimated PV relative to MPM, with the former presenting significant differences (+ 1.91 cm³, IQ = [- 0.33 cm³, 5.07 cm³], p val = 0.03). Both intra- (ICC > 0.90) and inter-rater (ICC > 0.90) reproducibility were excellent. MPM had the highest inter-rater reproducibility (ICC = 0.999). Inter-rater PV variation led to discrepancies in classification according to the clinical criterion of PSAd > 0.15 ng/mL for 2 patients (5%), 7 patients (17.5%), and 9 patients (22.5%) when using MPM, TEF, and BPEF, respectively.

CONCLUSION

PV measurements using ellipsoid formulas and MPM are highly reproducible. MPM is a robust method for PV assessment and PSAd calculation, with the lowest variability. TEF showed a high degree of concordance with MPM but a slight overestimation of PV. Precise anatomic landmarks as defined with the BPEF led to a more accurate PV estimation, but also to a higher variability.

KEY POINTS

• Manual planimetry used for prostate volume estimation is robust and reproducible, with the lowest variability between readers. • Ellipsoid formulas are accurate and reproducible but with higher variability between readers. • The traditional ellipsoid formula tends to overestimate prostate volume.

Which measurement method should be used for prostate volume for PI-RADS? A comparison of ellipsoid and segmentation methods

PURPOSE

Prostate volume and PSA density (PSAd) are important in the risk stratification of suspected prostate cancer (Pca). PI-RADS v2.1 allows for determining volume via segmentation or ellipsoid calculation. The purpose of our study was to compare ellipsoid and segmentation volume calculation methods and evaluate if PSAd diagnostic performance is altered.

METHODS

We retrospectively assessed 397 patients (mean age/standard deviation: 63.7/7.4 years) who underwent MRI and prostate biopsy or prostatectomy, with Pca classified by Gleason ≥3 + 4 and ≥4 + 4 disease. Prostate total volumes were determined with ellipsoid calculations (TVe) and with semi-automated segmentation (TVs), along with inter-rater reliability with intraclass correlation coefficient (ICC). PSAd was calculated for TVe and TVs and ROC curves were created to compare performance for Gleason ≥3 + 4 and ≥4 + 4 disease.

RESULTS

TVe was significantly higher than TVs (p < 0.0001), with mean TVe = 55.4 mL and TVs = 51.0 mL. ROC area under the curve for PSAd derived with TVe (0.63, 95%CI:0.59-0.68) and TVs (0.64, 95%CI:0.59-0.68) showed no significant difference for Gleason ≥3 + 4 disease (p = 0.45), but PSAd derived with TVs (0.63, 95%CI: 0.58-0.68) significantly outperformed TVe (0.61, 95%CI: 0.57-0.67) for Gleason ≥4 + 4 disease (p = 0.02). Both methods demonstrated excellent inter-rater reliability with TVe with ICC of 0.93(95%CI: 0.92-0.94) and TVs with ICC of 0.98(95%CI: 0.98-0.99).

CONCLUSION

Traditional ellipsoid measurements tend to overestimate total prostate volume compared to segmentation, but both methods demonstrate similar diagnostic performance of derived PSA density for PI-RADS clinically significant disease. For higher grade disease, PSAd derived from segmentation volumes demonstrates statistically significant superior performance. Both methods are viable, but segmentation volume is potentially better.

[PI-RADS 2.1 and structured reporting of magnetic resonance imaging of the prostate]

Klinisches/methodisches Problem

Die Identifikation klinisch signifikanter Prostatakarzinome bei gleichzeitigem Vermeiden einer Überdiagnostik niedrigmaligner Tumoren stellt eine Herausforderung in der klinischen Routine dar.

Radiologische Standardverfahren

Die gemäß PI-RADS-Richtlinien (Prostate Imaging Reporting and Data System Guidelines) akquirierte und interpretierte multiparametrische Magnetresonanztomographie (MRT) der Prostata ist als klinischer Standard bei Urologen und Radiologen akzeptiert.

Methodische Innovationen

Die PI-RADS-Richtlinien sind neu auf Version 2.1 aktualisiert worden und beinhalten neben präzisierten technischen Anforderungen einzelne Änderungen in der Läsionsbewertung.

Leistungsfähigkeit

Die PI-RADS-Richtlinien haben entscheidende Bedeutung in der Standardisierung der multiparametrischen MRT der Prostata erlangt und bieten Vorlagen zur strukturierten Befundung, was die Kommunikation mit dem Zuweiser erleichtert.

Bewertung

Die nun auf Version 2.1 aktualisierten Richtlinien stellen eine Verfeinerung der verbreiteten Version 2.0 dar. Dabei wurden viele Aspekte der Befundung präzisiert, einige vorbekannte Limitationen bleiben jedoch bestehen und erfordern die weitere Verbesserung der Richtlinien in kommenden Versionen.