scholarly journals AutoProstate: Towards Automated Reporting of Prostate MRI for Prostate Cancer Assessment Using Deep Learning

Cancers ◽  
2021 ◽  
Vol 13 (23) ◽  
pp. 6138
Author(s):  
Pritesh Mehta ◽  
Michela Antonelli ◽  
Saurabh Singh ◽  
Natalia Grondecka ◽  
Edward W. Johnston ◽  
...  
Keyword(s):  
Prostate Cancer ◽  
Deep Learning ◽  
Specific Antigen ◽  
Reporting Quality ◽  
Lesion Detection ◽  
Observer Agreement ◽  
Detection Sensitivity ◽  
Automatic Assessment ◽  
Experienced Radiologist ◽  
Clinically Significant

Multiparametric magnetic resonance imaging (mpMRI) of the prostate is used by radiologists to identify, score, and stage abnormalities that may correspond to clinically significant prostate cancer (CSPCa). Automatic assessment of prostate mpMRI using artificial intelligence algorithms may facilitate a reduction in missed cancers and unnecessary biopsies, an increase in inter-observer agreement between radiologists, and an improvement in reporting quality. In this work, we introduce AutoProstate, a deep learning-powered framework for automatic MRI-based prostate cancer assessment. AutoProstate comprises of three modules: Zone-Segmenter, CSPCa-Segmenter, and Report-Generator. Zone-Segmenter segments the prostatic zones on T2-weighted imaging, CSPCa-Segmenter detects and segments CSPCa lesions using biparametric MRI, and Report-Generator generates an automatic web-based report containing four sections: Patient Details, Prostate Size and PSA Density, Clinically Significant Lesion Candidates, and Findings Summary. In our experiment, AutoProstate was trained using the publicly available PROSTATEx dataset, and externally validated using the PICTURE dataset. Moreover, the performance of AutoProstate was compared to the performance of an experienced radiologist who prospectively read PICTURE dataset cases. In comparison to the radiologist, AutoProstate showed statistically significant improvements in prostate volume and prostate-specific antigen density estimation. Furthermore, AutoProstate matched the CSPCa lesion detection sensitivity of the radiologist, which is paramount, but produced more false positive detections.

The CADMUS trial: A paired cohort, blinded study comparing multiparametric ultrasound targeted biopsies with multiparametric MRI targeted biopsies in the detection of clinically significant prostate cancer.

2021 ◽  
Vol 39 (15_suppl) ◽  
pp. 5008-5008
Author(s):  
Alistair Grey ◽  
Rebecca Scott ◽  
Bina Shah ◽  
Peter Acher ◽  
Sidath Liyanage ◽  
...  
Keyword(s):  
Prostate Cancer ◽  
Diagnostic Performance ◽  
Point Of Care ◽  
Low Cost ◽  
Reporting Quality ◽  
Multiparametric Mri ◽  
Lesion Detection ◽  
Gleason Grade ◽  
Clinically Significant ◽  
Targeted Biopsies

5008 Background: Multiparametric MRI (mpMRI) of the prostate followed by targeted biopsy is recommended in men at risk of prostate cancer. Dissemination of this pathway may be limited by cost, variable scan and reporting quality, and contraindicated in the presence of metallic implants and claustrophobia. Multi-parametric ultrasound (mpUSS) is a point of care test with low cost that combines b-mode, colour Doppler, elastography and contrast enhancement. CADMUS compared the diagnostic performance of mpUSS to mpMRI. Methods: CADMUS recruited 370 patients from seven sites to a prospective, multicentre, paired-cohort trial (ISRCTN 38541912). Ethics committee approval was obtained. Patients underwent both mpUSS and mpMRI independently, each with a positive test defined as a Likert score of >3. Those with either a positive mpUSS or mpMRI, or both, were advised to undergo targeted biopsies. Reporting of each scan was carried out blind to the other and prior to biopsy; patients advised for biopsy were blinded to which test was positive. The order of mpUSS and mpMRI targeting was randomised. Primary outcomes were proportion of positive tests and detection of clinically significant cancer (csPCa) defined as Gleason >4+3 of any length and/or maximum cancer core length of >6mm of any grade [PROMIS definition1]. Results: 306 completed both mpUSS and mpMRI. Agreement in lesion detection between mpUSS and mpMRI was 73.2% (kappa 0.06, p = 0.14). 257 with positive results on mpUSS, mpMRI or both had targeted biopsies. Agreement on detection of csPCa was 91.1% (expected 59.8%, kappa 0.78, p < 0.01). Overall, mpUSS detected 4.3% fewer csPCa than mpMRI (95% CI = [-8.3%, -1.5%]; p = 0.042 [Bonferroni correction]). mpUSS detected 7.2% (6/83) csPCa missed by mpMRI; mpMRI detected 20.5% (17/83) csPCa that mpUSS missed. At a less stringent definition of significant cancer, Gleason grade >3+4 of any length (definition 3), agreement was 89.1% (expected 55.6%, kappa 0.75, p < 0.01) mpUSS detected 5.4% fewer definition 3 cancers than mpMRI overall. mpUSS detected 7% (7/99) definition 3 cancers that mpMRI missed; mpMRI detected 21% (21/99) definition 3 cancers that mpUSS missed. Conclusions: The CADMUS trial shows mpUSS has a diagnostic performance approaching that of mpMRI and significant cancer detection is improved by the use of both scans over mpMRI alone. Clinical trial information: 38541912. [Table: see text]

scholarly journals Analysis of risk factors for determining the need for prostate biopsy in patients with negative MRI

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Linghui Liang ◽  
Feng Qi ◽  
Yifei Cheng ◽  
Lei Zhang ◽  
Dongliang Cao ◽  
...  
Keyword(s):  
Prostate Cancer ◽  
Detection Efficiency ◽  
Specific Antigen ◽  
Digital Rectal Examination ◽  
Multivariable Logistic Regression Analysis ◽  
Detection Rates ◽  
Prostate Biopsies ◽  
Clinically Significant ◽  
Definition Of ◽  
Medical University

AbstractTo analyze the clinical characteristics of patients with negative biparametric magnetic resonance imaging (bpMRI) who didn’t need prostate biopsies (PBs). A total of 1,012 male patients who underwent PBs in the First Affiliated Hospital of Nanjing Medical University from March 2018 to November 2019, of 225 had prebiopsy negative bpMRI (defined as Prostate Imaging Reporting and Data System (PI-RADS 2.1) score less than 3). The detection efficiency of clinically significant prostate cancer (CSPCa) was assessed according to age, digital rectal examination (DRE), prostate volume (PV) on bpMRI, prostate-specific antigen (PSA) and PSA density (PSAD). The definition of CSPCa for Gleason score > 6. Univariate and multivariable logistic regression analysis were used to identify predictive factors of absent CSPCa on PBs. Moreover, absent CSPCa contained clinically insignificant prostate cancer (CIPCa) and benign result. The detection rates of present prostate cancer (PCa) and CSPCa were 27.11% and 16.44%, respectively. Patients who were diagnosed as CSPCa had an older age (P < 0.001), suspicious DRE (P < 0.001), a smaller PV (P < 0.001), higher PSA value (P = 0.008) and higher PSAD (P < 0.001) compared to the CIPCa group and benign result group. PSAD < 0.15 ng/ml/cm3 (P = 0.004) and suspicious DRE (P < 0.001) were independent predictors of absent CSPCa on BPs. The negative forecast value of bpMRI for BP detection of CSPCa increased with decreasing PSAD, mainly in patients with naive PB (P < 0.001) but not in prior negative PB patients. 25.33% of the men had the combination of negative bpMRI, PSAD < 0.15 ng/ml/cm3 and PB naive, and none had CSPCa on repeat PBs. The incidence of PB was determined, CSPCa was 1.59%, 0% and 16.67% in patients with negative bpMRI and PSAD < 0.15 ng/ml/cm3, patients with negative bpMRI, PSAD < 0.15 ng/ml/cm3 and biopsy naive and patients with negative bpMRI, PSAD < 0.15 ng/ml/cm3 and prior negative PB, separately. We found that a part of patients with negative bpMRI, a younger age, no suspicious DRE and PSAD < 0.15 ng/ml/cm3 may securely avoid PBs. Conversely PB should be considered in patients regardless of negative bpMRI, especially who with a greater age, obviously suspicious DRE, significantly increased PSA value, a significantly small PV on MRI and PSAD > 0.15 ng/ml/cm3.

The lesion detection efficacy of deep learning on automatic breast ultrasound and factors affecting its efficacy: a pilot study

2021 ◽  
Author(s):  
Xiao Luo PhD ◽  
Min Xu ◽  
Guoxue Tang ◽  
Yi Wang PhD ◽  
Na Wang ◽  
...  
Keyword(s):  
Deep Learning ◽  
Breast Tissue ◽  
Test Group ◽  
Lesion Size ◽  
Breast Ultrasound ◽  
Lesion Detection ◽  
Detection Sensitivity ◽  
Control Group ◽  
Factors Affecting ◽  
Handheld Ultrasound

Objectives: The aim of this study was to investigate the detection efficacy of deep learning (DL) for automatic breast ultrasound (ABUS) and factors affecting its efficacy. Methods: Women who underwent ABUS and handheld ultrasound from May 2016 to June 2017 (N = 397) were enrolled and divided into training (n = 163 patients with breast cancer and 33 with benign lesions), test (n = 57) and control (n = 144) groups. A convolutional neural network was optimised to detect lesions in ABUS. The sensitivity and false positives (FPs) were evaluated and compared for different breast tissue compositions, lesion sizes, morphologies and echo patterns. Results: In the training set, with 688 lesion regions (LRs), the network achieved sensitivities of 93.8%, 97.2 and 100%, based on volume, lesion and patient, respectively, with 1.9 FPs per volume. In the test group with 247 LRs, the sensitivities were 92.7%, 94.5 and 96.5%, respectively, with 2.4 FPs per volume. The control group, with 900 volumes, showed 0.24 FPs per volume. The sensitivity was 98% for lesions > 1 cm3, but 87% for those ≤1 cm3 (p < 0.05). Similar sensitivities and FPs were observed for different breast tissue compositions (homogeneous, 97.5%, 2.1; heterogeneous, 93.6%, 2.1), lesion morphologies (mass, 96.3%, 2.1; non-mass, 95.8%, 2.0) and echo patterns (homogeneous, 96.1%, 2.1; heterogeneous 96.8%, 2.1). Conclusions: DL had high detection sensitivity with a low FP but was affected by lesion size. Advances in knowledge: DL is technically feasible for the automatic detection of lesions in ABUS.

scholarly journals Who’s too old to screen? Prostate cancer in elderly men

2013 ◽  
Vol 3 (3) ◽  
pp. 205 ◽  
Author(s):  
Sandeep Mistry ◽  
Wesley Mayer ◽  
Rose Khavari ◽  
Gustavo Ayala ◽  
Brian Miles
Keyword(s):  
Prostate Cancer ◽  
Specific Antigen ◽  
Threshold Value ◽  
The Elderly ◽  
Cancer De La Prostate ◽  
Gleason Grade ◽  
Elderly Men ◽  
Patients Âgés ◽  
Positive Core ◽  
Clinically Significant

Introduction: Prostate cancer is the most common nonskin malignancyaffecting men and is the second leading cause of cancerrelateddeath in North America. The incidence of prostate cancerincreases dramatically with age. However, many healthauthorities advocate the cessation of routine prostate cancer testingin men older than 75 because of the belief that most patientswill have a clinically insignificant cancer and will not benefit fromtherapy. The true prevalence of clinically significant prostate cancerin elderly men is not known.Methods: We analyzed 1446 needle biopsies of the prostate inmen aged 75 or older. All pathological reviews were conductedby the pathology department at the Methodist Hospital in Houston,Tex. Data were collected from pathology reports, hospital andclinic databases, and medical records when available. Dataobtained included age at biopsy, serum prostate-specific antigen(PSA) levels, number of positive core biopsies and Gleasongrade. Statistical analysis was performed using Stata. Clinicallysignificant cancer was defined by the pathological presence ofGleason grade 6 adenocarcinoma in more than 1 biopsy coreor the presence of any Gleason 4 or 5 component in the biopsy.Results: The median age of the patients included in the studywas 78.8 and 95% of the patients were between the ages of 75and 85. The mean serum PSA level for patients biopsied was10.4 μg/L. Of all biopsies reviewed, 53% were positive for prostatecancer and 78% of these would be defined as clinically significantcancer. Regression analysis revealed age to be a significant(p < 0.05) factor for increased Gleason grade in positive biopsies.Logistic regression revealed age as a significant factor (p <0.05) for clinically significant prostate cancer even when controllingfor PSA. A serum PSA threshold value of 6.5 μg/L would havemissed 38% of significant cancers and a threshold of 4.0 μg/Lwould have missed 8% of significant cancers.Conclusion: Our findings suggest that the prevalence of clinicallysignificant prostate cancer in the elderly population may be higherthan previously thought. As the population continues to livelonger and healthier lives, it will become more common to confrontprostate cancer morbidity in the eldery population. Usinghigher serum PSA thresholds to eliminate unnecessary biopsies inolder men does not appear to help identify patients at greaterrisk of having clinically significant prostate cancer. Patients withprostate cancer having aggressive clinical features may benefitfrom treatment of their prostate cancer well into their eighth andninth decades of life. Testing and diagnostic recommendationsshould reflect the potential benefit of identifying patients withaggressive prostate cancer even after age 75.Introduction : Le cancer de la prostate est le type de cancer noncutané le plus fréquent chez les hommes et la seconde causede décès lié au cancer en importance en Amérique du Nord.L’incidence du cancer de la prostate augmente grandement avecl’âge. Néanmoins, de nombreuses autorités en matière de santéavancent l’idée de mettre fin au dépistage systématique du cancerde la prostate chez les hommes de plus de 75 ans en raisonde la croyance selon laquelle la plupart des patients présenterontun cancer non significatif sur le plan clinique et ne bénéficierontpas d’un traitement. La véritable prévalence des cas decancer de la prostate cliniquement significatif chez les hommesâgés n’est pas établie.Méthodes : Nous avons analysé 1446 échantillons de biopsie àl’aiguille prélevés au niveau de la prostate chez des patients de75 ans ou plus. Toutes les analyses de pathologie ont été effectuéespar le service de pathologie du Methodist Hospital deHouston, au Texas. Les données ont été tirées des rapports depathologie, des bases de données des hôpitaux et des cliniques,et des dossiers médicaux lorsque possible. Les données obtenuesincluaient l’âge au moment de la biopsie, les valeurs d’antigèneprostatique spécifique (APS), le nombre de microbiopsies positiveset le score de Gleason. Les analyses statistiques ont été effectuéesà l’aide du système Stata. Le cancer cliniquement significatifest défini comme la présence d’un adénocarcinome avec un scorede Gleason de 6 dans plus d’une zone de biopsie ou un scorede Gleason de 4 ou 5 dans toute partie de l’échantillon.Résultats : L’âge moyen des patients inclus était de 78,8 ans et95 % des patients avaient entre 75 et 85 ans. La valeur moyennede l’APS chez les patients ayant subi une biopsie était de 10,4 μg/L.De tous les échantillons examinés, 53 % confirmaient la présenced’un cancer de la prostate, et le cancer était défini comme étantcliniquement significatif dans 78 % de ces cas. Une analyse derégression a révélé que l’âge était un facteur significatif (p <0,05) lié à un score de Gleason plus élevé dans les biopsies positives.Une analyse de régression logistique a révélé que l’âge étaitaussi un facteur significatif (p < 0,05) lié à un cancer de la prostatecliniquement significatif même en tenant compte du taux d’APS.Une valeur seuil d’APS de 6,5 μg/L serait passée à côté de 38 %des cas de cancer significatif, alors qu’une valeur seuil d’APSde 4,0 μg/L serait passée à côté de 8 % des cancers significatifs.Conclusion : Nos observations portent à croire que la prévalencedu cancer de la prostate significatif sur le plan clinique chezles patients âgés pourrait être plus élevée qu’on le croit. Avecl’augmentation de l’espérance de vie, l’incidence de la morbiditéliée au cancer de la prostate augmentera. Le recours àdes valeurs seuils d’APS plus élevées pour éliminer les cas debiopsies non nécessaires chez les hommes âgés ne semble pasaider à cerner les patients présentant un risque plus élevé de cancerde la prostate cliniquement significatif. Les patients atteintsde cancer de la prostate cliniquement agressif peuvent bénéficierd’un traitement contre le cancer même lorsqu’ils dépassentlargement les 80 ou les 90 ans. Les recommandations concernantle dépistage et le diagnostic devraient refléter les avantages potentielsliés au dépistage d’un cancer de la prostate agressif, mêmeaprès 75 ans.

scholarly journals Analysis of Urinary Prostate-Specific Antigen Glycoforms in Samples of Prostate Cancer and Benign Prostate Hyperplasia

2016 ◽  
Vol 2016 ◽  
pp. 1-12 ◽  
Author(s):  
Chun-Jen Hsiao ◽  
Tzong-Shin Tzai ◽  
Chein-Hung Chen ◽  
Wen-Horng Yang ◽  
Chung-Hsuan Chen
Keyword(s):  
Prostate Cancer ◽  
Prostate Specific Antigen ◽  
Benign Prostate Hyperplasia ◽  
Clinical Sample ◽  
Specific Antigen ◽  
Ion Accumulation ◽  
Detection Sensitivity ◽  
Prostate Hyperplasia ◽  
Liquid Chromatography Tandem Mass ◽  
Benign Prostate

Glycans of prostate-specific antigen (PSA) in prostate cancer were found to be different from that in benign disease. It is difficult to analyze heterogeneous PSA glycoforms in each individual specimen because of low protein abundance and the limitation of detection sensitivity. We developed a method for prostate cancer diagnosis based on PSA glycoforms. Specific glycoforms were screened in each clinical sample based on liquid chromatography-tandem mass spectrometry with ion accumulation. To look for potential biomarkers, normalized abundance of each glycoform in benign prostate hyperplasia (BPH) and in prostate cancer was evaluated. The PSA glycoform, Hex5HexNAc4NeuAc1dHex1, and monosialylated, sialylated, and unfucosylated glycoforms differed significantly between the prostate cancer and BPH samples. The detection sensitivity (87.5%) and specificity (60%) for prostate cancer identification are higher than those of the serum PSA marker. As low as 100 amol PSA could be detected with the ion accumulation method which has not been reported before. The improved detection specificity can help reduce unnecessary examinations.

Multiparametric ultrasound and micro-ultrasound in prostate cancer: a comprehensive review

2021 ◽  
Author(s):  
Adriano Basso Dias ◽  
Ciara O’Brien ◽  
Jean-Michel Correas ◽  
Sangeet Ghai
Keyword(s):  
Prostate Cancer ◽  
Specific Antigen ◽  
Transrectal Ultrasound ◽  
Digital Rectal Examination ◽  
High Sensitivity ◽  
Cost Effective ◽  
Cognitive Fusion ◽  
Clinically Significant ◽  
Multiparametric Ultrasound ◽  
Targeted Biopsies

Prostate cancer (PCa) is the most common non-cutaneous cancer diagnosed in males. Traditional tools for screening and diagnosis, such as prostate-specific antigen, digital rectal examination and conventional transrectal ultrasound (TRUS), present low accuracy for PCa detection. Multiparametric MRI has become a game changer in the PCa diagnosis pathway and MRI-targeted biopsies are currently recommended for males at risk of clinically significant PCa, even in biopsy-naïve patients. Recent advances in ultrasound have also emerged with the goal to provide a readily accessible and cost-effective tool for detection of PCa. These newer techniques include elastography and contrast-enhanced ultrasound, as well as improved B-mode and Doppler techniques. These modalities can be combined to define a novel ultrasound approach, multiparametric ultrasound. High frequency Micro-ultrasound has emerged as a promising imaging technology for PCa diagnosis. Initial results have shown high sensitivity of Micro-ultrasound in detecting PCa in addition to its potential in improving the accuracy of targeted biopsies, based on targeting under real-time visualization, rather than relying on cognitive/fusion software MRI-transrectal ultrasound-guided biopsy.

scholarly journals Accuracy of Tumour-Associated Circulating Endothelial Cells as a Screening Biomarker for Clinically Significant Prostate Cancer

Cancers ◽  
2019 ◽  
Vol 11 (8) ◽  
pp. 1064 ◽  
Author(s):  
Sebastian Chakrit Bhakdi ◽  
Prapat Suriyaphol ◽  
Ponpan Thaicharoen ◽  
Sebastian Tobias Karl Grote ◽  
Chulaluk Komoltri ◽  
...  
Keyword(s):  
Prostate Cancer ◽  
Endothelial Cells ◽  
Predictive Value ◽  
Specific Antigen ◽  
Transrectal Ultrasound ◽  
Circulating Endothelial Cells ◽  
Index Test ◽  
Psa Testing ◽  
Diagnostic Potential ◽  
Clinically Significant

Even though more than 350,000 men die from prostate cancer every year, broad-based screening for the disease remains a controversial topic. Guidelines demand that the only commonly accepted screening tool, prostate-specific antigen (PSA) testing, must be followed by prostate biopsy if results are elevated. Due to the procedure’s low positive predictive value (PPV), however, over 80% of biopsies are performed on healthy men or men with clinically insignificant cancer—prompting calls for new ways of vetting equivocal PSA readings prior to the procedure. Responding to the challenge, the present study investigated the diagnostic potential of tumour-associated circulating endothelial cells (tCECs), which have previously been described as a novel, blood-based biomarker for clinically significant cancers. Specifically, the objective was to determine the diagnostic accuracy of a tCEC-based blood test to detect clinically significant prostate cancer (defined as Gleason score ≥ 3 + 4) in high-risk patients. Performed in a blinded, prospective, single-centre set-up, it compared a novel tCEC index test with transrectal ultrasound-guided biopsy biopsy as a reference on a total of 170 patients and found that a tCEC add-on test will almost double the PPV of a standalone PSA test (32% vs. 17%; p = 0.0012), while retaining a negative predictive value above 90%.

Sign in / Sign up

Export Citation Format

Share Document