Diagnosis and Localization of Prostate Cancer via Automated Multiparametric MRI Equipped with Artificial Intelligence

Prostate MRI scans for pre-biopsied patients are important. However, fewer radiologists are available for MRI diagnoses, which requires multi-sequential interpretations of multi-slice images. To reduce such a burden, artificial intelligence (AI)-based, computer-aided diagnosis is expected to be a critical technology. We present an AI-based method for pinpointing prostate cancer location and determining tumor morphology using multiparametric MRI. The study enrolled 15 patients who underwent radical prostatectomy between April 2008 and August 2017 at our institution. We labeled the cancer area on the peripheral zone on MR images, comparing MRI with histopathological mapping of radical prostatectomy specimens. Likelihood maps were drawn, and tumors were divided into morphologically distinct regions using the superpixel method. Likelihood maps consisted of pixels, which utilize the cancer likelihood value computed from the T2-weighted, apparent diffusion coefficient, and diffusion-weighted MRI-based texture features. Cancer location was determined based on the likelihood maps. We evaluated the diagnostic performance by the area under the receiver operating characteristic (ROC) curve according to the Chi-square test. The area under the ROC curve was 0.985. Sensitivity and specificity for our approach were 0.875 and 0.961 (p < 0.01), respectively. Our AI-based procedures were successfully applied to automated prostate cancer localization and shape estimation using multiparametric MRI.

Download Full-text

Artificial Intelligence Is a Promising Prospect for the Detection of Prostate Cancer Extracapsular Extension With Mp-mri: A Two-center Comparative Study

10.21203/rs.3.rs-298296/v1 ◽

2021 ◽

Author(s):

Ying Hou ◽

Yi-Hong Zhang ◽

Jie Bao ◽

Mei-Ling Bao ◽

Guang Yang ◽

...

Keyword(s):

Prostate Cancer ◽

Artificial Intelligence ◽

Paired Comparison ◽

Characteristic Curve ◽

Multiparametric Mri ◽

Training Data ◽

Urinary Continence ◽

Data Sets ◽

Extracapsular Extension ◽

Promising Alternative

Abstract Purpose: A balance between preserving urinary continence and achievement of negative margins is of clinical relevance while implementary difficulty. Preoperatively accurate detection of prostate cancer (PCa) extracapsular extension (ECE) is thus crucial for determining appropriate treatment options. We aimed to develop and clinically validate an artificial intelligence (AI)-assisted tool for the detection of ECE in patients with PCa using multiparametric MRI. Methods: 849 patients with localized PCa underwent multiparametric MRI before radical prostatectomy were retrospectively included from two medical centers. The AI tool was built on a ResNeXt network embedded with a spatial attention map of experts’ prior knowledges (PAGNet) from 596 training data sets. The tool was validated in 150 internal and 103 external data sets, respectively; and its clinical applicability was compared with expert-based interpretation and AI-expert interaction.Results: An index PAGNet model using a single-slice image yielded the highest areas under the receiver operating characteristic curve (AUC) of 0.857 (95% confidence interval [CI], 0.827-0.884), 0.807 (95% CI, 0.735-0.867) and 0.728 (95% CI, 0.631-0.811) in the training, internal test and external test cohorts, compared to the conventional ResNeXt networks. For experts, the inter-reader agreement was observed in only 437/849 (51.5%) patients with a Kappa value 0.343. And the performance of two experts (AUC, 0.632 to 0.741 vs 0.715 to 0.857) was lower (paired comparison, all p values < 0.05) than that of AI assessment. When expert’ interpretations were adjusted by the AI assessments, the performance of both two experts was improved.Conclusion: Our AI tool, showing improved accuracy, offers a promising alternative to human experts for imaging staging of PCa ECE using multiparametric MRI.

Download Full-text

MP66-16 CLINICAL SIGNIFICANCE OF MULTIPARAMETRIC MRI AND PSA DENSITY AS PREDICTORS OF RESIDUAL TUMOR (PT0) FOLLOWING RADICAL PROSTATECTOMY FOR T1A-T1B (INCIDENTAL) PROSTATE CANCER

The Journal of Urology ◽

10.1097/01.ju.0000556948.84535.57 ◽

2019 ◽

Vol 201 (Supplement 4) ◽

Author(s):

Doo Yong Chung* ◽

Jong Soo Lee ◽

Won Sik Jang ◽

In Rae Cho ◽

Young Sig Kim ◽

...

Keyword(s):

Prostate Cancer ◽

Radical Prostatectomy ◽

Clinical Significance ◽

Residual Tumor ◽

Multiparametric Mri ◽

Incidental Prostate Cancer ◽

Psa Density

Download Full-text

Multiparametric MRI for Recurrent Prostate Cancer Post Radical Prostatectomy and Postradiation Therapy

BioMed Research International ◽

10.1155/2014/316272 ◽

2014 ◽

Vol 2014 ◽

pp. 1-23 ◽

Cited By ~ 28

Author(s):

Flavio Barchetti ◽

Valeria Panebianco

Keyword(s):

Prostate Cancer ◽

Radical Prostatectomy ◽

Local Recurrence ◽

Locoregional Recurrence ◽

Multiparametric Mri ◽

Cochrane Library ◽

Published Data ◽

Detection And Localization ◽

The Impact

The clinical suspicion of local recurrence of prostate cancer (PCa) after radical prostatectomy (RP) and after radiation therapy (RT) is based on the onset of biochemical failure. The aim of this paper was to review the current role of multiparametric-MRI (mp-MRI) in the detection of locoregional recurrence. A systematic literature search using the Medline and Cochrane Library databases was performed from January 1995 up to November 2013. Bibliographies of retrieved and review articles were also examined. Only those articles reporting complete data with clinical relevance for the present review were selected. This review article is divided into two major parts: the first one considers the role of mp-MRI in the detection of PCa local recurrence after RP; the second part provides an insight about the impact of mp-MRI in the depiction of locoregional recurrence after RT (interstitial or external beam). Published data indicate an emerging role for mp-MRI in the detection and localization of locally recurrent PCa both after RP and RT which represents an information of paramount importance to perform focal salvage treatments.

Download Full-text

Prediction of prostate cancer Gleason score upgrading from biopsy to radical prostatectomy using pre-biopsy multiparametric MRI PIRADS scoring system

Scientific Reports ◽

10.1038/s41598-020-64693-y ◽

2020 ◽

Vol 10 (1) ◽

Cited By ~ 1

Author(s):

Saeed Alqahtani ◽

Cheng Wei ◽

Yilong Zhang ◽

Magdalena Szewczyk-Bieda ◽

Jennifer Wilson ◽

...

Keyword(s):

Prostate Cancer ◽

Radical Prostatectomy ◽

Gleason Score ◽

Scoring System ◽

Multiparametric Mri

Download Full-text

The impact of multiparametric MRI on utilization of nerve sparing and surgical margins status in patients with clinically localized prostate cancer treated with robot-assisted radical prostatectomy

European Urology Open Science ◽

10.1016/s2666-1683(20)33789-7 ◽

2020 ◽

Vol 19 ◽

pp. e1773

Author(s):

G. Giannarini ◽

R. Girometti ◽

S. Pizzolitto ◽

F. Zattoni ◽

C. Valotto ◽

...

Keyword(s):

Prostate Cancer ◽

Radical Prostatectomy ◽

Multiparametric Mri ◽

Surgical Margins ◽

Localized Prostate Cancer ◽

Nerve Sparing ◽

Robot Assisted ◽

The Impact

Download Full-text

11C-choline PET scanning is more accurate than biopsy in assessment of localized prostate cancer planned for radical prostatectomy.

Journal of Clinical Oncology ◽

10.1200/jco.2012.30.5_suppl.182 ◽

2012 ◽

Vol 30 (5_suppl) ◽

pp. 182-182

Author(s):

Ian D. Davis ◽

Sze Ting Lee ◽

Lekshmy Shanker ◽

David Clouston ◽

Damien M Bolton ◽

...

Keyword(s):

Prostate Cancer ◽

Radical Prostatectomy ◽

Treatment Plan ◽

Curative Intent ◽

Trus Biopsy ◽

Department Of Health ◽

Imaging Results ◽

Mri Scans ◽

Treat Prostate Cancer ◽

Ct And Mri

182 Background: A decision to treat prostate cancer (PC) with radical prostatectomy (RP) with curative intent requires confidence that the PC is confined to the prostate. PC outcomes will improve with better selection of surgical candidates. Current imaging modalities include CT and MRI but have limited accuracy. We assessed 18F-FDG (FDG) and 11C-choline (CHOL) PET in men planned for RP to determine the accuracy of PET, effects of PET on decision making by surgeons, and correlation with PSA. Methods: Written informed consent was obtained from eligible participants (pts) planned for RP. All men underwent TRUS-guided prostatic biopsies, CT and MRI scans, PSA and standard tests of organ function. The urologist then documented the treatment plan based on these results. Pts then underwent FDG and CHOL PET and the urologist then determined whether this information altered the treatment plan. After surgery the RP specimen was reconstructed, examined histologically and correlated with TRUS and imaging results on a sextant-based analysis (apex/mid/base on both sides). Results: 30 pts entered and completed the trial. Outcomes are shown in the table. Neither PET modality significantly affected decisions about surgery. Preoperative PSA did not correlate with degree of involvement. FDG PET was unhelpful. Conclusions: CHOL PET was the most sensitive and most accurate modality with highest congruity with pathology and had excellent positive predictive value, but was least specific. CHOL PET was superior to both TRUS biopsy and MRI. Supported by grant 487916 through Cancer Australia, Prostate Cancer Foundation Australia, Australian Government Department of Health and Aging. [Table: see text]

Download Full-text

Is systematic 10 core prostate needle biopsy enough for treatment determination of localized prostate cancer?

Journal of Clinical Oncology ◽

10.1200/jco.2012.30.5_suppl.163 ◽

2012 ◽

Vol 30 (5_suppl) ◽

pp. 163-163

Author(s):

Ryo Kishimoto ◽

Ryuta Tanimoto ◽

Kensuke Bekku ◽

Yasuyuki Kobayashi ◽

Shin Ebara ◽

...

Keyword(s):

Prostate Cancer ◽

Multivariate Analysis ◽

Radical Prostatectomy ◽

Gleason Score ◽

Needle Biopsy ◽

Chi Square ◽

Prostate Needle Biopsy ◽

Significant Difference ◽

Chi Square Test

163 Background: To evaluate whether the systematic 10 cores prostate needle biopsy is enough for determination of NCCN risk classification (NRC), we analyzed migration of Gleason score (GS), cancer location, and NRC between pre and postoperative periods in a cohort of patients who underwent radical prostatectomy. Methods: A total of 197 patients were included in this study. These patients were divided into three groups along the number of biopsy cores: less than 10 (L), 10, and more than 10 (M). We compared between three groups about Gleason score, cancer location and NCCN risk classification change (CC) between prostate biopsy and radical prostatectomy specimen. Statistical analysis were performed with chi-square test, and multiple logistic regression with p<0.05, and Bonferroni correction with p<0.017 considered significant difference. Results: The rate of CC in L, 10, M was 55.1%, 43.0%, 26.5%, respectively. On chi-square test rates of CC were significantly different between three groups (P=0.035), but rates of Gleason score and cancer location were not. On univariate analysis, PSA (Odds rate (OR) 0.872 p<0.001), preoperative NRC (low vs. intermediate, and poor, OR 0.157 and 0.241, p<0.001), prostate volume (normal vs. mild or moderate, OR 1.989 p=0.025), the number of biopsy cores (L vs. M, OR 0.293 p=0.011), GS (6 vs. 8, OR 2.374 p=0.021) were correlated with CC. On multivariate analysis, the most important independent predictive factors for CC were preoperative NRC (low vs. intermediate, p<0.001, OR 0.198, 95% CI 0.09-0.45) and PSA (p=0.007, OR 0.903, 95%CI 0.83-0.98), but the number of biopsy cores was not associated CC significantly. Conclusions: Although multivariate analysis showed no significant difference, the more biopsy cores reduced the risk of CC. Systematic 10 core biopsy might be insufficient for accurate diagnosis and treatment decision of prostate cancer.

Download Full-text

Location of local recurrence after MRI-guided partial prostate brachytherapy targeting only the peripheral zone: Implications for focal therapy.

Journal of Clinical Oncology ◽

10.1200/jco.2013.31.6_suppl.149 ◽

2013 ◽

Vol 31 (6_suppl) ◽

pp. 149-149 ◽

Cited By ~ 1

Author(s):

Konstantin Kovtun ◽

Tobias Penzkofer ◽

Neha Agrawal ◽

Tina Kapur ◽

Andriy Fedorov ◽

...

Keyword(s):

Prostate Cancer ◽

Local Recurrence ◽

Central Zone ◽

Peripheral Zone ◽

Focal Therapy ◽

Prostate Brachytherapy ◽

Local Recurrences ◽

Endorectal Coil ◽

Mri Scans ◽

Mri Guided

149 Background: Prostate cancer local recurrences usually occur at the same site as the dominant primary tumor in patients treated with radiation therapy to the whole gland. We characterized location of local recurrences in patients who were treated with MRI Guided Partial Brachytherapy in which only the peripheral zone was targeted. Methods: We retrospectively reviewed ten patients with initial cT1c, Gleason score 3+4 or less prostate cancer who developed biopsy proven local recurrences and had available imaging after MRI Guided Partial Brachytherapy targeting the peripheral zone from 1998 to 2006. All 10 patients had 1.5T endorectal coil MRI at diagnosis, performed primarily for staging and not for tumor localization, while at recurrence 8 had 3T endorectal coil MRI and 2 had 1.5T endorectal coil MRI. Scans consisted of at least T1 and T2 sequences. Two radiologists (C.T. and T.P.) blinded to clinical data reviewed diagnosis MRI scans together and quantified likelihood of tumor on a 1 to 5 scale in each section of an eight part prostate in both pre-treatment and recurrence scans. Local recurrence was judged to be in the same location as the baseline tumor if at least 50% of the tumor location overlapped. Results: Only 3 of 10 patients had local recurrences at the same location as the baseline tumor with a mean overlap of 64%. 7 of 10 patients had local recurrences at a different location with a mean overlap of 5%. 5 of 10 patients had recurrences in the central zone of the prostate which did not definitively show tumor on review of the initial 1.5T staging scan. Conclusions: After MRI-guided brachytherapy targeting only the peripheral zone in men initially staged with 1.5T MRI, 50% of the local recurrences occurred at the non-targeted central zone, raising the possibility that focal therapy directed only at the dominant tumor will result in increased out-of-field recurrences. Whether the superior ability of modern 3T multiparametric MRI to detect and precisely localize occult prostate cancer foci will reduce this risk is the subject of current study.

Download Full-text