Dataset Information

Automated detection of aggressive and indolent prostate cancer on magnetic resonance imaging.

ABSTRACT:

Purpose

While multi-parametric magnetic resonance imaging (MRI) shows great promise in assisting with prostate cancer diagnosis and localization, subtle differences in appearance between cancer and normal tissue lead to many false positive and false negative interpretations by radiologists. We sought to automatically detect aggressive cancer (Gleason pattern ≥ 4) and indolent cancer (Gleason pattern 3) on a per-pixel basis on MRI to facilitate the targeting of aggressive cancer during biopsy.

Methods

We created the Stanford Prostate Cancer Network (SPCNet), a convolutional neural network model, trained to distinguish between aggressive cancer, indolent cancer, and normal tissue on MRI. Ground truth cancer labels were obtained by registering MRI with whole-mount digital histopathology images from patients who underwent radical prostatectomy. Before registration, these histopathology images were automatically annotated to show Gleason patterns on a per-pixel basis. The model was trained on data from 78 patients who underwent radical prostatectomy and 24 patients without prostate cancer. The model was evaluated on a pixel and lesion level in 322 patients, including six patients with normal MRI and no cancer, 23 patients who underwent radical prostatectomy, and 293 patients who underwent biopsy. Moreover, we assessed the ability of our model to detect clinically significant cancer (lesions with an aggressive component) and compared it to the performance of radiologists.

Results

Our model detected clinically significant lesions with an area under the receiver operator characteristics curve of 0.75 for radical prostatectomy patients and 0.80 for biopsy patients. Moreover, the model detected up to 18% of lesions missed by radiologists, and overall had a sensitivity and specificity that approached that of radiologists in detecting clinically significant cancer.

Conclusions

Our SPCNet model accurately detected aggressive prostate cancer. Its performance approached that of radiologists, and it helped identify lesions otherwise missed by radiologists. Our model has the potential to assist physicians in specifically targeting the aggressive component of prostate cancers during biopsy or focal treatment.

SUBMITTER: Seetharaman A

PROVIDER: S-EPMC8360053 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:BackgroundDistinguishing men with aggressive from indolent prostate cancer is critical to decisions in the management of clinically localized prostate cancer. Molecular signatures of aggressive disease could help men overcome this major clinical challenge by reducing unnecessary treatment and allowing more appropriate treatment of aggressive disease.MethodsWe performed a mass spectrometry-based proteomic analysis of normal and malignant prostate tissues from 22 men who underwent surgery for prostate cancer. Prostate cancer samples included Grade Groups (3-5), with 8 patients experiencing recurrence and 14 without evidence of recurrence with a mean of 6.8 years of follow-up. To better understand the biological pathways underlying prostate cancer aggressiveness, we performed a systems biology analysis and gene enrichment analysis. Proteins that distinguished recurrent from nonrecurrent cancer were chosen for validation by immunohistochemical analysis on tissue microarrays containing samples from a larger cohort of patients with recurrent and nonrecurrent prostate cancer.ResultsIn all, 24,037 unique peptides (false discovery rate < 1%) corresponding to 3,313 distinct proteins were identified with absolute abundance ranges spanning seven orders of magnitude. Of these proteins, 115 showed significantly (p < 0.01) different levels in tissues from recurrent versus nonrecurrent cancers. Analysis of all differentially expressed proteins in recurrent and nonrecurrent cases identified several protein networks, most prominently one in which approximately 24% of the proteins in the network were regulated by the YY1 transcription factor (adjusted p < 0.001). Strong immunohistochemical staining levels of three differentially expressed proteins, POSTN, CALR, and CTSD, on a tissue microarray validated their association with shorter patient survival.ConclusionsThe protein signatures identified could improve understanding of the molecular drivers of aggressive prostate cancer and be used as candidate prognostic biomarkers.

Project description:PurposeGiven the limitations of prostate specific antigen and standard biopsies for detecting prostate cancer, we evaluated the cancer detection rate and external validity of a magnetic resonance imaging/transrectal ultrasound fusion guided prostate biopsy system used at the National Institutes of Health.Materials and methodsWe performed a phase III trial of a magnetic resonance imaging/transrectal ultrasound fusion guided prostate biopsy system with participants enrolled between 2012 and 2013. A total of 153 men consented to the study and underwent 3 Tesla multiparametric magnetic resonance imaging with an endorectal coil for clinical suspicion of prostate cancer. Lesions were classified as low or moderate/high risk for prostate cancer. Magnetic resonance imaging/transrectal ultrasound fusion guided biopsy and standard 12-core prostate biopsy were performed and 105 men were eligible for analysis.ResultsMean patient age was 65.8 years and mean prostate specific antigen was 9.5 ng/ml. The overall cancer detection rate was 62.9% (66 of 105 patients). The cancer detection rate in those with moderate/high risk on imaging was 72.3% (47 of 65) vs 47.5% (19 of 40) in those classified as low risk for prostate cancer (p<0.05). Mean tumor core length was 4.6 and 3.7 mm for fusion biopsy and standard 12-core biopsy, respectively (p<0.05). Magnetic resonance imaging/transrectal ultrasound fusion guided biopsy detected prostate cancer that was missed by standard 12-core biopsy in 14.3% of cases (15 of 105), of which 86.7% (13 of 15) were clinically significant. This biopsy upgraded 23.5% of cancers (4 of 17) deemed clinically insignificant on 12-core biopsy to clinically significant prostate cancer necessitating treatment.ConclusionsMagnetic resonance imaging/transrectal ultrasound fusion guided biopsy can improve prostate cancer detection. The results of this trial support the external validity of this platform and may be the next step in the evolution of prostate cancer management.

Project description:Rationale and objectivesProstate MRI improves detection of clinically significant prostate cancer; however, its diagnostic performance has wide variation. Artificial intelligence (AI) has the potential to assist radiologists in the detection and classification of prostatic lesions. Herein, we aimed to develop and test a cascaded deep learning detection and classification system trained on biparametric prostate MRI using PI-RADS for assisting radiologists during prostate MRI read out.Materials and methodsT2-weighted, diffusion-weighted (ADC maps, high b value DWI) MRI scans obtained at 3 Tesla from two institutions (n = 1043 in-house and n = 347 Prostate-X, respectively) acquired between 2015 to 2019 were used for model training, validation, testing. All scans were retrospectively reevaluated by one radiologist. Suspicious lesions were contoured and assigned a PI-RADS category. A 3D U-Net-based deep neural network was used to train an algorithm for automated detection and segmentation of prostate MRI lesions. Two 3D residual neural network were used for a 4-class classification task to predict PI-RADS categories 2 to 5 and BPH. Training and validation used 89% (n = 1290 scans) of the data using 5 fold cross-validation, the remaining 11% (n = 150 scans) were used for independent testing. Algorithm performance at lesion level was assessed using sensitivities, positive predictive values (PPV), false discovery rates (FDR), classification accuracy, Dice similarity coefficient (DSC). Additional analysis was conducted to compare AI algorithm's lesion detection performance with targeted biopsy results.ResultsMedian age was 66 years (IQR = 60-71), PSA 6.7 ng/ml (IQR = 4.7-9.9) from in-house cohort. In the independent test set, algorithm correctly detected 111 of 198 lesions leading to 56.1% (49.3%-62.6%) sensitivity. PPV was 62.7% (95% CI 54.7%-70.7%) with FDR of 37.3% (95% CI 29.3%-45.3%). Of 79 true positive lesions, 82.3% were tumor positive at targeted biopsy, whereas of 57 false negative lesions, 50.9% were benign at targeted biopsy. Median DSC for lesion segmentation was 0.359. Overall PI-RADS classification accuracy was 30.8% (95% CI 24.6%-37.8%).ConclusionOur cascaded U-Net, residual network architecture can detect, classify cancer suspicious lesions at prostate MRI with good detection, reasonable classification performance metrics.