Dataset Information

Improving workflow in prostate MRI: AI-based decision-making on biparametric or multiparametric MRI.

ABSTRACT:

Objectives

To develop and validate an artificial intelligence algorithm to decide on the necessity of dynamic contrast-enhanced sequences (DCE) in prostate MRI.

Methods

This study was approved by the institutional review board and requirement for study-specific informed consent was waived. A convolutional neural network (CNN) was developed on 300 prostate MRI examinations. Consensus of two expert readers on the necessity of DCE acted as reference standard. The CNN was validated in a separate cohort of 100 prostate MRI examinations from the same vendor and 31 examinations from a different vendor. Sensitivity/specificity were calculated using ROC curve analysis and results were compared to decisions made by a radiology technician.

Results

The CNN reached a sensitivity of 94.4% and specificity of 68.8% (AUC: 0.88) for the necessity of DCE, correctly assigning 44%/34% of patients to a biparametric/multiparametric protocol. In 2% of all patients, the CNN incorrectly decided on omitting DCE. With a technician reaching a sensitivity of 63.9% and specificity of 89.1%, the use of the CNN would allow for an increase in sensitivity of 30.5%. The CNN achieved an AUC of 0.73 in a set of examinations from a different vendor.

Conclusions

The CNN would have correctly assigned 78% of patients to a biparametric or multiparametric protocol, with only 2% of all patients requiring re-examination to add DCE sequences. Integrating this CNN in clinical routine could render the requirement for on-table monitoring obsolete by performing contrast-enhanced MRI only when needed.

SUBMITTER: Hotker AM

PROVIDER: S-EPMC8353049 | biostudies-literature | 2021 Aug

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Improving workflow in prostate MRI: AI-based decision-making on biparametric or multiparametric MRI.

Hötker Andreas M AM Da Mutten Raffaele R Tiessen Anja A Konukoglu Ender E Donati Olivio F OF

Insights into imaging 20210809 1

<h4>Objectives</h4>To develop and validate an artificial intelligence algorithm to decide on the necessity of dynamic contrast-enhanced sequences (DCE) in prostate MRI.<h4>Methods</h4>This study was approved by the institutional review board and requirement for study-specific informed consent was waived. A convolutional neural network (CNN) was developed on 300 prostate MRI examinations. Consensus of two expert readers on the necessity of DCE acted as reference standard. The CNN was validated in ...[more]

PMID: 34370164

Similar Datasets

Project description:ObjectivesTo evaluate the feasibility of automatic longitudinal analysis of consecutive biparametric MRI (bpMRI) scans to detect clinically significant (cs) prostate cancer (PCa).MethodsThis retrospective study included a multi-center dataset of 1513 patients who underwent bpMRI (T2 + DWI) between 2014 and 2020, of whom 73 patients underwent at least two consecutive bpMRI scans and repeat biopsies. A deep learning PCa detection model was developed to produce a heatmap of all PIRADS ≥ 2 lesions across prior and current studies. The heatmaps for each patient's prior and current examination were used to extract differential volumetric and likelihood features reflecting explainable changes between examinations. A machine learning classifier was trained to predict from these features csPCa (ISUP > 1) at the current examination according to biopsy. A classifier trained on the current study only was developed for comparison. An extended classifier was developed to incorporate clinical parameters (PSA, PSA density, and age). The cross-validated diagnostic accuracies were compared using ROC analysis. The diagnostic performance of the best model was compared to the radiologist scores.ResultsThe model including prior and current study (AUC 0.81, CI: 0.69, 0.91) resulted in a higher (p = 0.04) diagnostic accuracy than the current only model (AUC 0.73, CI: 0.61, 0.84). Adding clinical variables further improved diagnostic performance (AUC 0.86, CI: 0.77, 0.93). The diagnostic performance of the surveillance AI model was significantly better (p = 0.02) than of radiologists (AUC 0.69, CI: 0.54, 0.81).ConclusionsOur proposed AI-assisted surveillance of prostate MRI can pick up explainable, diagnostically relevant changes with promising diagnostic accuracy.Key points• Sequential prostate MRI scans can be automatically evaluated using a hybrid deep learning and machine learning approach. • The diagnostic accuracy of our csPCa detection AI model improved by including clinical parameters.

Project description:PurposeThe purpose of this study is to explore the value of combining bpMRI and clinical indicators in the diagnosis of clinically significant prostate cancer (csPCa), and developing a prediction model and Nomogram to guide clinical decision-making.MethodsWe retrospectively analyzed 530 patients who underwent prostate biopsy due to elevated serum prostate specific antigen (PSA) levels and/or suspicious digital rectal examination (DRE). Enrolled patients were randomly assigned to the training group (n = 371, 70%) and validation group (n = 159, 30%). All patients underwent prostate bpMRI examination, and T2-weighted imaging (T2WI) and diffusion-weighted imaging (DWI) sequences were collected before biopsy and were scored, which were respectively named T2WI score and DWI score according to Prostate Imaging Reporting and Data System version 2 (PI-RADS v.2) scoring protocol, and then PI-RADS scoring was performed. We defined a new bpMRI-based parameter named Total score (Total score = T2WI score + DWI score). PI-RADS score and Total score were separately included in the multivariate analysis of the training group to determine independent predictors for csPCa and establish prediction models. Then, prediction models and clinical indicators were compared by analyzing the area under the curve (AUC) and decision curves. A Nomogram for predicting csPCa was established using data from the training group.ResultsIn the training group, 160 (43.1%) patients had prostate cancer (PCa), including 128 (34.5%) with csPCa. Multivariate regression analysis showed that the PI-RADS score, Total score, f/tPSA, and PSA density (PSAD) were independent predictors of csPCa. The prediction model that was defined by Total score, f/tPSA, and PSAD had the highest discriminatory power of csPCa (AUC = 0.931), and the diagnostic sensitivity and specificity were 85.1% and 87.5%, respectively. Decision curve analysis (DCA) showed that the prediction model achieved an optimal overall net benefit in both the training group and the validation group. In addition, the Nomogram predicted csPCa revealed good estimation when compared with clinical indicators.ConclusionThe prediction model and Nomogram based on bpMRI and clinical indicators exhibit a satisfactory predictive value and improved risk stratification for csPCa, which could be used for clinical biopsy decision-making.

Project description:There is a growing need to localize prostate cancers on magnetic resonance imaging (MRI) to facilitate the use of image guided biopsy, focal therapy, and active surveillance follow up. Our goal was to develop a decision support system (DSS) for detecting and localizing peripheral zone prostate cancers by using machine learning approach to calculate a cancer probability map from multiparametric MR images (MP-MRI).This IRB approved Health Insurance Portability and Accountability Act compliant retrospective study consisted of 31 patients (mean age and serum prostate specific antigen of 60.4 and 6.62 ng∕ml, respectively) who had MP-MRI at 3 T followed by radical prostatectomy. Seven patients were excluded due to technical issues with their MP-MRI (e.g., motion artifact, failure to perform all sequences). Cancer and normal regions were identified in the peripheral zone by correlating them to whole mount histology slides of the excised prostatectomy specimens. To facilitate the correlation, tissue blocks matching the MR slices were obtained using a MR-based patient-specific mold. Segmented regions on the MP-MRI were correlated to histopathology and used as training sets for the learning system that generated the cancer probability maps. Leave-one-patient-out cross-validation on the cancer and normal regions was performed to determine the learning system's efficacy, an evolutionary strategies approach (also known as a genetic algorithm) was used to find the optimal values for a set of parameters, and finally a cancer probability map was generated.For the 24 patients that were used in the study, 225 cancer and 264 noncancerous regions were identified from the region maps. The efficacy of DSS was first determined without optimizing support vector machines (SVM) parameters, where a region having a cancer probability greater than or equal to 50% was considered as a correct classification. The nonoptimized system had an f-measure of 85% and the Kappa coefficient of 71% (Rater's agreement, where raters are DSS and ground truth histology). The efficacy of the DSS after optimizing SVM parameters using a genetic algorithm had an f-measure of 89% and a Kappa coefficient of 80%. Thus, after optimization of the DSS there was a 4% increase in the f-measure and a 9% increase in the Kappa coefficient.This DSS provides a cancer probability map for peripheral zone prostate tumors based on endorectal MP-MRI. These cancer probability maps can potentially aid radiologists in accurately localizing peripheral zone prostate cancers for planning targeted biopsies, focal therapy, and follow up for active surveillance.

Project description:Background. Multiple studies have shown that digitally mediated decision aids help prepare patients for medical decision making with their providers. However, few studies have investigated whether decision-support preferences differ between non-English-speaking and English-speaking Latino men with limited literacy. Objective. To identify and compare health information seeking patterns, preferences for information presentation, and interest in digital decision aids in a sample of Southern Californian underserved Latino men with newly diagnosed prostate cancer at a county hospital. Methods. We conducted semistructured, in-depth telephone interviews with 12 Spanish-speaking and 8 English-speaking Latino men using a purposive sampling technique. Following transcription of taped interviews, Spanish interviews were translated. Using a coding protocol developed by the team, two bilingual members jointly analyzed the transcripts for emerging themes. Coder agreement exceeded 80%. Differences were resolved through discussion. Results. Thematic differences between groups with different preferred languages emerged. Most respondents engaged in online health information seeking using cellphones, perceived a paternalistic patient-provider relationship, and expressed willingness to use hypothetical digital decision aids if recommended by their provider. English speakers reported higher digital technology proficiency for health-related searches. They also more frequently indicated family involvement in digital search related to their condition and preferred self-guided, web-based decision aids. In comparison, Spanish speakers reported lower digital technology proficiency and preferred family-involved, coach-guided, paper and visual decision aids. English speakers reported substantially higher levels of formal education. Conclusion. Preferences regarding the use of digital technology to inform prostate cancer treatment decision making among underserved Latino men varied depending on preferred primary language. Effective preparation of underserved Latino men for shared decision making requires consideration of alternative approaches depending on level of education attainment and preferred primary language.

Dataset Information

Improving workflow in prostate MRI: AI-based decision-making on biparametric or multiparametric MRI.

Objectives

Methods

Results

Conclusions

Publications

Improving workflow in prostate MRI: AI-based decision-making on biparametric or multiparametric MRI.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets