Artificial Intelligence Based Algorithms for Prostate Cancer Classification and Detection on Magnetic Resonance Imaging: A Narrative Review

From molecular mechanisms of prostate cancer to translational applications: based on multi-omics fusion analysis and intelligent medicine

Prostate cancer is the most common cancer in men worldwide and has a high mortality rate. The complex and heterogeneous development of prostate cancer has become a core obstacle in the treatment of prostate cancer. Simultaneously, the issues of overtreatment in early-stage diagnosis, oligometastasis and dormant tumor recognition, as well as personalized drug utilization, are also specific concerns that require attention in the clinical management of prostate cancer. Some typical genetic mutations have been proved to be associated with prostate cancer's initiation and progression. However, single-omic studies usually are not able to explain the causal relationship between molecular alterations and clinical phenotypes. Exploration from a systems genetics perspective is also lacking in this field, that is, the impact of gene network, the environmental factors, and even lifestyle behaviors on disease progression. At the meantime, current trend emphasizes the utilization of artificial intelligence (AI) and machine learning techniques to process extensive multidimensional data, including multi-omics. These technologies unveil the potential patterns, correlations, and insights related to diseases, thereby aiding the interpretable clinical decision making and applications, namely intelligent medicine. Therefore, there is a pressing need to integrate multidimensional data for identification of molecular subtypes, prediction of cancer progression and aggressiveness, along with perosonalized treatment performing. In this review, we systematically elaborated the landscape from molecular mechanism discovery of prostate cancer to clinical translational applications. We discussed the molecular profiles and clinical manifestations of prostate cancer heterogeneity, the identification of different states of prostate cancer, as well as corresponding precision medicine practices. Taking multi-omics fusion, systems genetics, and intelligence medicine as the main perspectives, the current research results and knowledge-driven research path of prostate cancer were summarized.

Emerging Strategies in Drug Development and Clinical Care in the Era of Personalized and Precision Medicine

In the ever-changing landscape of modern medicine, we face an important moment where the interplay of disease, drugs, and patients defines a new paradigm [...].

Patient perspectives on the use of artificial intelligence in prostate cancer diagnosis on MRI

OBJECTIVES

This study investigated patients' acceptance of artificial intelligence (AI) for diagnosing prostate cancer (PCa) on MRI scans and the factors influencing their trust in AI diagnoses.

MATERIALS AND METHODS

A prospective, multicenter study was conducted between January and November 2023. Patients undergoing prostate MRI were surveyed about their opinions on hypothetical AI assessment of their MRI scans. The questionnaire included nine items: four on hypothetical scenarios of combinations between AI and the radiologist, two on trust in the diagnosis, and three on accountability for misdiagnosis. Relationships between the items and independent variables were assessed using multivariate analysis.

RESULTS

A total of 212 PCa suspicious patients undergoing prostate MRI were included. The majority preferred AI involvement in their PCa diagnosis alongside a radiologist, with 91% agreeing with AI as the primary reader and 79% as the secondary reader. If AI has a high certainty diagnosis, 15% of the respondents would accept it as the sole decision-maker. Autonomous AI outperforming radiologists would be accepted by 52%. Higher educated persons tended to accept AI when it would outperform radiologists (p < 0.05). The respondents indicated that the hospital (76%), radiologist (70%), and program developer (55%) should be held accountable for misdiagnosis.

CONCLUSIONS

Patients favor AI involvement alongside radiologists in PCa diagnosis. Trust in AI diagnosis depends on the patient's education level and the AI performance, with autonomous AI acceptance by a small majority on the condition that AI outperforms a radiologist. Respondents held the hospital, radiologist, and program developers accountable for misdiagnosis in descending order of accountability.

CLINICAL RELEVANCE STATEMENT

Patients show a high level of acceptance for AI-assisted prostate cancer diagnosis on MRI, either alongside radiologists or fully autonomous, particularly if it demonstrates superior performance to radiologists alone.

KEY POINTS

Prostate cancer suspicious patients may accept autonomous AI based on performance. Patients prefer AI involvement alongside a radiologist in diagnosing prostate cancer. Patients indicate accountability for AI should be shared among multiple stakeholders.

Artificial Intelligence Improves the Ability of Physicians to Identify Prostate Cancer Extent

PURPOSE

Defining prostate cancer contours is a complex task, undermining the efficacy of interventions such as focal therapy. A multireader multicase study compared physicians' performance using artificial intelligence (AI) vs standard-of-care methods for tumor delineation.

MATERIALS AND METHODS

Cases were interpreted by 7 urologists and 3 radiologists from 5 institutions with 2 to 23 years of experience. Each reader evaluated 50 prostatectomy cases retrospectively eligible for focal therapy. Each case included a T2-weighted MRI, contours of the prostate and region(s) of interest suspicious for cancer, and a biopsy report. First, readers defined cancer contours cognitively, manually delineating tumor boundaries to encapsulate all clinically significant disease. Then, after ≥ 4 weeks, readers contoured the same cases using AI software. Using tumor boundaries on whole-mount histopathology slides as ground truth, AI-assisted, cognitively-defined, and hemigland cancer contours were evaluated. Primary outcome measures were the accuracy and negative margin rate of cancer contours. All statistical analyses were performed using generalized estimating equations.

RESULTS

The balanced accuracy (mean of voxel-wise sensitivity and specificity) of AI-assisted cancer contours (84.7%) was superior to cognitively-defined (67.2%) and hemigland contours (75.9%; P < .0001). Cognitively-defined cancer contours systematically underestimated cancer extent, with a negative margin rate of 1.6% compared to 72.8% for AI-assisted cancer contours (P < .0001).

CONCLUSIONS

AI-assisted cancer contours reduce underestimation of prostate cancer extent, significantly improving contouring accuracy and negative margin rate achieved by physicians. This technology can potentially improve outcomes, as accurate contouring informs patient management strategy and underpins the oncologic efficacy of treatment.

Development and validation of an imageless machine-learning algorithm for the initial screening of prostate cancer

PURPOSE

Prostate specific antigen (PSA) testing is a low-cost screening method for prostate cancer (PCa). However, its accuracy is limited. While progress is being made using medical imaging for PCa screening, PSA testing can still be improved as an easily accessible first step in the screening process. We aimed to develop and validate a new model by further personalizing the analysis of PSA with demographic, medical history, lifestyle parameters, and digital rectal examination (DRE) results.

METHODS

Using data from 34,224 patients in the screening arm of the PLCO trial (22,188 for the training set and 12,036 for the validation set), we applied a gradient-boosting model whose features (Model 1) were one PSA value and the personal variables available in the PLCO trial except those that signaled an ex-ante assumption of PCa. A second algorithm (Model 2) included a DRE result. The primary outcome was the occurrence of PCa, while the aggressiveness of PCa was a secondary outcome. ROC analyses were used to compare both models to other initial screening tests.

RESULTS

The areas under the curve (AUC) for Model 2 was 0.894 overall and 0.908 for patients with a suspicious DRE, compared to 0.808 for PSA for patients with a suspicious DRE. The AUC for Model 1 was 0.814 compared to 0.821 for PSA. Model 2 predicted 58% more high-risk PCa than PSA ≥4 combined with an abnormal DRE and had a positive predictive value of 74.7% (vs. 50.6%).

CONCLUSION

Personalizing the interpretation of PSA values and DRE results with a gradient-boosting model showed promising results as a potential novel, low-cost method for the initial screening of PCa. The importance of DRE, when included in such a model, was also highlighted.

The Roadmap toward Personalized Medicine: Challenges and Opportunities

In 2019, the International Consortium for Personalised Medicine (ICPerMed) developed a vision on how the use of personalized medicine (PM) approaches will promote "next-generation" medicine in 2030 more firmly centered on the individual's personal characteristics, leading to improved health outcomes within sustainable healthcare systems through research, development, innovation, and implementation for the benefit of patients, citizens, and society. Nevertheless, there are significant hurdles that healthcare professionals, researchers, policy makers, and patients must overcome to implement PM. The ICPerMed aims to provide recommendations to increase stakeholders' awareness on actionable measures to be implemented for the realization of PM. Starting with best practice examples of PM together with consultation of experts and stakeholders, a careful analysis that underlined hurdles, opportunities, recommendations, and information, aiming at developing knowledge on the requirements for PM implementation in healthcare practices, has been provided. A pragmatic roadmap has been defined for PM integration into healthcare systems, suggesting actions to overcome existing barriers and harness the potential of PM for improved health outcomes. In fact, to facilitate the adoption of PM by diverse stakeholders, it is mandatory to have a comprehensive set of resources tailored to stakeholder needs in critical areas of PM. These include engagement strategies, collaboration frameworks, infrastructure development, education and training programs, ethical considerations, resource allocation guidelines, regulatory compliance, and data management and privacy.

Autonomous Tumor Signature Extraction Applied to Spatially Registered Bi-Parametric MRI to Predict Prostate Tumor Aggressiveness: A Pilot Study

BACKGROUND

Accurate, reliable, non-invasive assessment of patients diagnosed with prostate cancer is essential for proper disease management. Quantitative assessment of multi-parametric MRI, such as through artificial intelligence or spectral/statistical approaches, can provide a non-invasive objective determination of the prostate tumor aggressiveness without side effects or potential poor sampling from needle biopsy or overdiagnosis from prostate serum antigen measurements. To simplify and expedite prostate tumor evaluation, this study examined the efficacy of autonomously extracting tumor spectral signatures for spectral/statistical algorithms for spatially registered bi-parametric MRI.

METHODS

Spatially registered hypercubes were digitally constructed by resizing, translating, and cropping from the image sequences (Apparent Diffusion Coefficient (ADC), High B-value, T2) from 42 consecutive patients in the bi-parametric MRI PI-CAI dataset. Prostate cancer blobs exceeded a threshold applied to the registered set from normalizing the registered set into an image that maximizes High B-value, but minimizes the ADC and T2 images, appearing "green" in the color composite. Clinically significant blobs were selected based on size, average normalized green value, sliding window statistics within a blob, and position within the hypercube. The center of mass and maximized sliding window statistics within the blobs identified voxels associated with tumor signatures. We used correlation coefficients (R) and p-values, to evaluate the linear regression fits of the z-score and SCR (with processed covariance matrix) to tumor aggressiveness, as well as Area Under the Curves (AUC) for Receiver Operator Curves (ROC) from logistic probability fits to clinically significant prostate cancer.

RESULTS

The highest R (R > 0.45), AUC (>0.90), and lowest p-values (<0.01) were achieved using z-score and modified registration applied to the covariance matrix and tumor signatures selected from the "greenest" parts from the selected blob.

CONCLUSIONS

The first autonomous tumor signature applied to spatially registered bi-parametric MRI shows promise for determining prostate tumor aggressiveness.

Radiologists' perspectives on the workflow integration of an artificial intelligence-based computer-aided detection system: A qualitative study

In healthcare, artificial intelligence (AI) is expected to improve work processes, yet most research focuses on the technical features of AI rather than its real-world clinical implementation. To evaluate the implementation process of an AI-based computer-aided detection system (AI-CAD) for prostate MRI readings, we interviewed German radiologists in a pre-post design. We embedded our findings in the Model of Workflow Integration and the Technology Acceptance Model to analyze workflow effects, facilitators, and barriers. The most prominent barriers were: (i) a time delay in the work process, (ii) additional work steps to be taken, and (iii) an unstable performance of the AI-CAD. Most frequently named facilitators were (i) good self-organization, and (ii) good usability of the software. Our results underline the importance of a holistic approach to AI implementation considering the sociotechnical work system and provide valuable insights into key factors of the successful adoption of AI technologies in work systems.

Detecting MRI-Invisible Prostate Cancers Using a Weakly Supervised Deep Learning Model

Background

MRI is an important tool for accurate detection and targeted biopsy of prostate lesions. However, the imaging appearances of some prostate cancers are similar to those of the surrounding normal tissue on MRI, which are referred to as MRI-invisible prostate cancers (MIPCas). The detection of MIPCas remains challenging and requires extensive systematic biopsy for identification. In this study, we developed a weakly supervised UNet (WSUNet) to detect MIPCas.

Methods

The study included 777 patients (training set: 600; testing set: 177), all of them underwent comprehensive prostate biopsies using an MRI-ultrasound fusion system. MIPCas were identified in MRI based on the Gleason grade (≥7) from known systematic biopsy results.

Results

The WSUNet model underwent validation through systematic biopsy in the testing set with an AUC of 0.764 (95% CI: 0.728-0.798). Furthermore, WSUNet exhibited a statistically significant precision improvement of 91.3% (p < 0.01) over conventional systematic biopsy methods in the testing set. This improvement resulted in a substantial 47.6% (p < 0.01) decrease in unnecessary biopsy needles, while maintaining the same number of positively identified cores as in the original systematic biopsy.

Conclusions

In conclusion, the proposed WSUNet could effectively detect MIPCas, thereby reducing unnecessary biopsies.

Performance of a Region of Interest-based Algorithm in Diagnosing International Society of Urological Pathology Grade Group ≥2 Prostate Cancer on the MRI-FIRST Database-CAD-FIRST Study

BACKGROUND AND OBJECTIVE

Prostate multiparametric magnetic resonance imaging (MRI) shows high sensitivity for International Society of Urological Pathology grade group (GG) ≥2 cancers. Many artificial intelligence algorithms have shown promising results in diagnosing clinically significant prostate cancer on MRI. To assess a region-of-interest-based machine-learning algorithm aimed at characterising GG ≥2 prostate cancer on multiparametric MRI.

METHODS

The lesions targeted at biopsy in the MRI-FIRST dataset were retrospectively delineated and assessed using a previously developed algorithm. The Prostate Imaging-Reporting and Data System version 2 (PI-RADSv2) score assigned prospectively before biopsy and the algorithm score calculated retrospectively in the regions of interest were compared for diagnosing GG ≥2 cancer, using the areas under the curve (AUCs), and sensitivities and specificities calculated with predefined thresholds (PIRADSv2 scores ≥3 and ≥4; algorithm scores yielding 90% sensitivity in the training database). Ten predefined biopsy strategies were assessed retrospectively.

KEY FINDINGS AND LIMITATIONS

After excluding 19 patients, we analysed 232 patients imaged on 16 different scanners; 85 had GG ≥2 cancer at biopsy. At patient level, AUCs of the algorithm and PI-RADSv2 were 77% (95% confidence interval [CI]: 70-82) and 80% (CI: 74-85; p = 0.36), respectively. The algorithm's sensitivity and specificity were 86% (CI: 76-93) and 65% (CI: 54-73), respectively. PI-RADSv2 sensitivities and specificities were 95% (CI: 89-100) and 38% (CI: 26-47), and 89% (CI: 79-96) and 47% (CI: 35-57) for thresholds of ≥3 and ≥4, respectively. Using the PI-RADSv2 score to trigger a biopsy would have avoided 26-34% of biopsies while missing 5-11% of GG ≥2 cancers. Combining prostate-specific antigen density, the PI-RADSv2 and algorithm's scores would have avoided 44-47% of biopsies while missing 6-9% of GG ≥2 cancers. Limitations include the retrospective nature of the study and a lack of PI-RADS version 2.1 assessment.

CONCLUSIONS AND CLINICAL IMPLICATIONS

The algorithm provided robust results in the multicentre multiscanner MRI-FIRST database and could help select patients for biopsy.

PATIENT SUMMARY

An artificial intelligence-based algorithm aimed at diagnosing aggressive cancers on prostate magnetic resonance imaging showed results similar to expert human assessment in a prospectively acquired multicentre test database.

Reliable Prostate Cancer Risk Mapping From MRI Using Targeted and Systematic Core Needle Biopsy Histopathology

OBJECTIVE

To compute a dense prostate cancer risk map for the individual patient post-biopsy from magnetic resonance imaging (MRI) and to provide a more reliable evaluation of its fitness in prostate regions that were not identified as suspicious for cancer by a human-reader in pre- and intra-biopsy imaging analysis.

METHODS

Low-level pre-biopsy MRI biomarkers from targeted and non-targeted biopsy locations were extracted and statistically tested for representativeness against biomarkers from non-biopsied prostate regions. A probabilistic machine learning classifier was optimized to map biomarkers to their core-level pathology, followed by extrapolation of pathology scores to non-biopsied prostate regions. Goodness-of-fit was assessed at targeted and non-targeted biopsy locations for the post-biopsy individual patient.

RESULTS

Our experiments showed high predictability of imaging biomarkers in differentiating histopathology scores in thousands of non-targeted core-biopsy locations (ROC-AUCs: 0.85-0.88), but also high variability between patients (Median ROC-AUC [IQR]: 0.81-0.89 [0.29-0.40]).

CONCLUSION

The sparseness of prostate biopsy data makes the validation of a whole gland risk mapping a non-trivial task. Previous studies i) focused on targeted-biopsy locations although biopsy-specimens drawn from systematically scattered locations across the prostate constitute a more representative sample to non-biopsied regions, and ii) estimated prediction-power across predicted instances (e.g., biopsy specimens) with no patient distinction, which may lead to unreliable estimation of model fitness to the individual patient due to variation between patients in instance count, imaging characteristics, and pathologies.

SIGNIFICANCE

This study proposes a personalized whole-gland prostate cancer risk mapping post-biopsy to allow clinicians to better stage and personalize focal therapy treatment plans.

A multivariate curve resolution analysis of multicenter proton spectroscopic imaging of the prostate for cancer localization and assessment of aggressiveness

In this study, we investigated the potential of the multivariate curve resolution alternating least squares (MCR-ALS) algorithm for analyzing three-dimensional (3D) 1 H-MRSI data of the prostate in prostate cancer (PCa) patients. MCR-ALS generates relative intensities of components representing spectral profiles derived from a large training set of patients, providing an interpretable model. Our objectives were to classify magnetic resonance (MR) spectra, differentiating tumor lesions from benign tissue, and to assess PCa aggressiveness. We included multicenter 3D 1 H-MRSI data from 106 PCa patients across eight centers. The patient cohort was divided into a training set (N = 63) and an independent test set (N = 43). Singular value decomposition determined that MR spectra were optimally represented by five components. The profiles of these components were extracted from the training set by MCR-ALS and assigned to specific tissue types. Using these components, MCR-ALS was applied to the test set for a quantitative analysis to discriminate tumor lesions from benign tissue and to assess tumor aggressiveness. Relative intensity maps of the components were reconstructed and compared with histopathology reports. The quantitative analysis demonstrated a significant separation between tumor and benign voxels (t-test, p < 0.001). This result was achieved including voxels with low-quality MR spectra. A receiver operating characteristic analysis of the relative intensity of the tumor component revealed that low- and high-risk tumor lesions could be distinguished with an area under the curve of 0.88. Maps of this component properly identified the extent of tumor lesions. Our study demonstrated that MCR-ALS analysis of 1 H-MRSI of the prostate can reliably identify tumor lesions and assess their aggressiveness. It handled multicenter data with minimal preprocessing and without using prior knowledge or quality control. These findings indicate that MCR-ALS can serve as an automated tool to assess the presence, extent, and aggressiveness of tumor lesions in the prostate, enhancing diagnostic capabilities and treatment planning of PCa patients.

Prospective effects of an artificial intelligence-based computer-aided detection system for prostate imaging on routine workflow and radiologists' outcomes

OBJECTIVES

Artificial intelligence (AI) is expected to alleviate the negative consequences of rising case numbers for radiologists. Currently, systematic evaluations of the impact of AI solutions in real-world radiological practice are missing. Our study addresses this gap by investigating the impact of the clinical implementation of an AI-based computer-aided detection system (CAD) for prostate MRI reading on clinicians' workflow, workflow throughput times, workload, and stress.

MATERIALS AND METHODS

CAD was newly implemented into radiology workflow and accompanied by a prospective pre-post study design. We assessed prostate MRI case readings using standardized work observations and questionnaires. The observation period was three months each in a single department. Workflow throughput times, PI-RADS score, CAD usage and radiologists' self-reported workload and stress were recorded. Linear mixed models were employed for effect identification.

RESULTS

In data analyses, 91 observed case readings (pre: 50, post: 41) were included. Variation of routine workflow was observed following CAD implementation. A non-significant increase in overall workflow throughput time was associated with CAD implementation (mean 16.99 ± 6.21 vs 18.77 ± 9.69 min, p = .51), along with an increase in diagnostic reading time for high suspicion cases (mean 15.73 ± 4.99 vs 23.07 ± 8.75 min, p = .02). Changes in radiologists' self-reported workload or stress were not found.

CONCLUSION

Implementation of an AI-based detection aid was associated with lower standardization and no effects over time on radiologists' workload or stress. Expectations of AI decreasing the workload of radiologists were not confirmed by our real-world study.

PRE-REGISTRATION

German register for clinical trials https://drks.de/; DRKS00027391.

Evaluation of a Deep Learning-based Algorithm for Post-Radiotherapy Prostate Cancer Local Recurrence Detection Using Biparametric MRI

OBJECTIVE

To evaluate a biparametric MRI (bpMRI)-based artificial intelligence (AI) model for the detection of local prostate cancer (PCa) recurrence in patients with radiotherapy history.

MATERIALS AND METHODS

This study included post-radiotherapy patients undergoing multiparametric MRI and subsequent MRI/US fusion-guided and/or systematic biopsy. Histopathology results were used as ground truth. The recurrent cancer detection sensitivity of a bpMRI-based AI model, which was developed on a large dataset to primarily identify lesions in treatment-naïve patients, was compared to a prospective radiologist assessment using the Wald test. Subanalysis was conducted on patients stratified by the treatment modality (external beam radiation treatment [EBRT] and brachytherapy) and the prostate volume quartiles.

RESULTS

Of the 62 patients included (median age = 70 years; median PSA = 3.51 ng/ml; median prostate volume = 27.55 ml), 56 recurrent PCa foci were identified within 46 patients. The AI model detected 40 lesions in 35 patients. The AI model performance was lower than the prospective radiology interpretation (Rad) on a patient-(AI: 76.1% vs. Rad: 91.3%, p = 0.02) and lesion-level (AI: 71.4% vs. Rad: 87.5%, p = 0.01). The mean number of false positives per patient was 0.35 (range: 0-2). The AI model performance was higher in EBRT group both on patient-level (EBRT: 81.5% [22/27] vs. brachytherapy: 68.4% [13/19]) and lesion-level (EBRT: 79.4% [27/34] vs. brachytherapy: 59.1% [13/22]). In patients with gland volumes >34 ml (n = 25), detection sensitivities were 100% (11/11) and 94.1% (16/17) on patient- and lesion-level, respectively.

CONCLUSION

The reported bpMRI-based AI model detected the majority of locally recurrent prostate cancer after radiotherapy. Further testing including external validation of this model is warranted prior to clinical implementation.

Artificial intelligence and allied subsets in early detection and preclusion of gynecological cancers

Gynecological cancers including breast, cervical, ovarian, uterine, and vaginal, pose the greatest threat to world health, with early identification being crucial to patient outcomes and survival rates. The application of machine learning (ML) and artificial intelligence (AI) approaches to the study of gynecological cancer has shown potential to revolutionize cancer detection and diagnosis. The current review outlines the significant advancements, obstacles, and prospects brought about by AI and ML technologies in the timely identification and accurate diagnosis of different types of gynecological cancers. The AI-powered technologies can use genomic data to discover genetic alterations and biomarkers linked to a particular form of gynecologic cancer, assisting in the creation of targeted treatments. Furthermore, it has been shown that the potential benefits of AI and ML technologies in gynecologic tumors can greatly increase the accuracy and efficacy of cancer diagnosis, reduce diagnostic delays, and possibly eliminate the need for needless invasive operations. In conclusion, the review focused on the integrative part of AI and ML based tools and techniques in the early detection and exclusion of various cancer types; together with a collaborative coordination between research clinicians, data scientists, and regulatory authorities, which is suggested to realize the full potential of AI and ML in gynecologic cancer care.

Relationship between Eccentricity and Volume Determined by Spectral Algorithms Applied to Spatially Registered Bi-Parametric MRI and Prostate Tumor Aggressiveness: A Pilot Study

(1) Background: Non-invasive prostate cancer assessments using multi-parametric MRI are essential to the reliable detection of lesions and proper management of patients. While current guidelines call for the administration of Gadolinium-containing intravenous contrast injections, eliminating such injections would simplify scanning and reduce patient risk and costs. However, augmented image analysis is necessary to extract important diagnostic information from MRIs. Purpose: This study aims to extend previous work on the signal to clutter ratio and test whether prostate tumor eccentricity and volume are indicators of tumor aggressiveness using bi-parametric (BP)-MRI. (2) Methods: This study retrospectively processed 42 consecutive prostate cancer patients from the PI-CAI data collection. BP-MRIs (apparent diffusion coefficient, high b-value, and T2 images) were resized, translated, cropped, and stitched to form spatially registered BP-MRIs. The International Society of Urological Pathology (ISUP) grade was used to judge cases of prostate cancer as either clinically significant prostate cancer (CsPCa) (ISUP ≥ 2) or clinically insignificant prostate cancer (CiPCa) (ISUP < 2). The Adaptive Cosine Estimator (ACE) algorithm was applied to the BP-MRIs, followed by thresholding, and then eccentricity and volume computations, from the labeled and blobbed detection maps. Then, univariate and multivariate linear regression fittings of eccentricity and volume were applied to the ISUP grade. The fits were quantitatively evaluated by computing correlation coefficients (R) and p-values. Area under the curve (AUC) and receiver operator characteristic (ROC) curve scores were used to assess the logistic fitting to CsPCa/CiPCa. (3) Results: Modest correlation coefficients (R) (>0.35) and AUC scores (0.70) for the linear and/or logistic fits from the processed prostate tumor eccentricity and volume computations for the spatially registered BP-MRIs exceeded fits using the parameters of prostate serum antigen, prostate volume, and patient age (R~0.17). (4) Conclusions: This is the first study that applied spectral approaches to BP-MRIs to generate tumor eccentricity and volume metrics to assess tumor aggressiveness. This study found significant values of R and AUC (albeit below those from multi-parametric MRI) to fit and relate the metrics to the ISUP grade and CsPCA/CiPCA, respectively.

Characterization of high-grade prostate cancer at multiparametric MRI using a radiomic-based computer-aided diagnosis system as standalone and second reader

PURPOSE

The purpose of this study was to develop and test across various scanners a zone-specific region-of-interest (ROI)-based computer-aided diagnosis system (CAD) aimed at characterizing, on MRI, International Society of Urological Pathology (ISUP) grade≥2 prostate cancers.

MATERIALS AND METHODS

ROI-based quantitative models were selected in multi-vendor training (265 pre-prostatectomy MRIs) and pre-test (112 pre-biopsy MRIs) datasets. The best peripheral and transition zone models were combined and retrospectively assessed in internal (158 pre-biopsy MRIs) and external (104 pre-biopsy MRIs) test datasets. Two radiologists (R1/R2) retrospectively delineated the lesions targeted at biopsy in test datasets. The CAD area under the receiver operating characteristic curve (AUC) for characterizing ISUP≥2 cancers was compared to that of the Prostate Imaging-Reporting and Data System version2 (PI-RADSv2) score prospectively assigned to targeted lesions.

RESULTS

The best models used the 25th apparent diffusion coefficient (ADC) percentile in transition zone and the 2nd ADC percentile and normalized wash-in rate in peripheral zone. The PI-RADSv2 AUCs were 82% (95% confidence interval [CI]: 74-87) and 86% (95% CI: 81-91) in the internal and external test datasets respectively. They were not different from the CAD AUCs obtained with R1 and R2 delineations, in the internal (82% [95% CI: 76-89], P = 0.95 and 85% [95% CI: 78-91], P = 0.55) and external (82% [95% CI: 74-91], P = 0.41 and 86% [95% CI:78-95], P = 0.98) test datasets. The CAD yielded sensitivities of 86-89% and 90-91%, and specificities of 64-65% and 69-75% in the internal and external test datasets respectively.

CONCLUSION

The CAD performance for characterizing ISUP grade≥2 prostate cancers on MRI is not different from that of PI-RADSv2 score across two test datasets.

Label-set impact on deep learning-based prostate segmentation on MRI

BACKGROUND

Prostate segmentation is an essential step in computer-aided detection and diagnosis systems for prostate cancer. Deep learning (DL)-based methods provide good performance for prostate gland and zones segmentation, but little is known about the impact of manual segmentation (that is, label) selection on their performance. In this work, we investigated these effects by obtaining two different expert label-sets for the PROSTATEx I challenge training dataset (n = 198) and using them, in addition to an in-house dataset (n = 233), to assess the effect on segmentation performance. The automatic segmentation method we used was nnU-Net.

RESULTS

The selection of training/testing label-set had a significant (p < 0.001) impact on model performance. Furthermore, it was found that model performance was significantly (p < 0.001) higher when the model was trained and tested with the same label-set. Moreover, the results showed that agreement between automatic segmentations was significantly (p < 0.0001) higher than agreement between manual segmentations and that the models were able to outperform the human label-sets used to train them.

CONCLUSIONS

We investigated the impact of label-set selection on the performance of a DL-based prostate segmentation model. We found that the use of different sets of manual prostate gland and zone segmentations has a measurable impact on model performance. Nevertheless, DL-based segmentation appeared to have a greater inter-reader agreement than manual segmentation. More thought should be given to the label-set, with a focus on multicenter manual segmentation and agreement on common procedures.

CRITICAL RELEVANCE STATEMENT

Label-set selection significantly impacts the performance of a deep learning-based prostate segmentation model. Models using different label-set showed higher agreement than manual segmentations.

KEY POINTS

• Label-set selection has a significant impact on the performance of automatic segmentation models. • Deep learning-based models demonstrated true learning rather than simply mimicking the label-set. • Automatic segmentation appears to have a greater inter-reader agreement than manual segmentation.

Prediction and Mapping of Intraprostatic Tumor Extent with Artificial Intelligence

Background

Magnetic resonance imaging (MRI) underestimation of prostate cancer extent complicates the definition of focal treatment margins.

Objective

To validate focal treatment margins produced by an artificial intelligence (AI) model.

Design setting and participants

Testing was conducted retrospectively in an independent dataset of 50 consecutive patients who had radical prostatectomy for intermediate-risk cancer. An AI deep learning model incorporated multimodal imaging and biopsy data to produce three-dimensional cancer estimation maps and margins. AI margins were compared with conventional MRI regions of interest (ROIs), 10-mm margins around ROIs, and hemigland margins. The AI model also furnished predictions of negative surgical margin probability, which were assessed for accuracy.

Outcome measurements and statistical analysis

Comparing AI with conventional margins, sensitivity was evaluated using Wilcoxon signed-rank tests and negative margin rates using chi-square tests. Predicted versus observed negative margin probability was assessed using linear regression. Clinically significant prostate cancer (International Society of Urological Pathology grade ≥2) delineated on whole-mount histopathology served as ground truth.

Results and limitations

The mean sensitivity for cancer-bearing voxels was higher for AI margins (97%) than for conventional ROIs (37%, p < 0.001), 10-mm ROI margins (93%, p = 0.24), and hemigland margins (94%, p < 0.001). For index lesions, AI margins were more often negative (90%) than conventional ROIs (0%, p < 0.001), 10-mm ROI margins (82%, p = 0.24), and hemigland margins (66%, p = 0.004). Predicted and observed negative margin probabilities were strongly correlated (R2 = 0.98, median error = 4%). Limitations include a validation dataset derived from a single institution's prostatectomy population.

Conclusions

The AI model was accurate and effective in an independent test set. This approach could improve and standardize treatment margin definition, potentially reducing cancer recurrence rates. Furthermore, an accurate assessment of negative margin probability could facilitate informed decision-making for patients and physicians.

Patient summary

Artificial intelligence was used to predict the extent of tumors in surgically removed prostate specimens. It predicted tumor margins more accurately than conventional methods.

Novel Multiparametric Magnetic Resonance Imaging-Based Deep Learning and Clinical Parameter Integration for the Prediction of Long-Term Biochemical Recurrence-Free Survival in Prostate Cancer after Radical Prostatectomy

Radical prostatectomy (RP) is the main treatment of prostate cancer (PCa). Biochemical recurrence (BCR) following RP remains the first sign of aggressive disease; hence, better assessment of potential long-term post-RP BCR-free survival is crucial. Our study aimed to evaluate a combined clinical-deep learning (DL) model using multiparametric magnetic resonance imaging (mpMRI) for predicting long-term post-RP BCR-free survival in PCa. A total of 437 patients with PCa who underwent mpMRI followed by RP between 2008 and 2009 were enrolled; radiomics features were extracted from T2-weighted imaging, apparent diffusion coefficient maps, and contrast-enhanced sequences by manually delineating the index tumors. Deep features from the same set of imaging were extracted using a deep neural network based on pretrained EfficentNet-B0. Here, we present a clinical model (six clinical variables), radiomics model, DL model (DLM-Deep feature), combined clinical-radiomics model (CRM-Multi), and combined clinical-DL model (CDLM-Deep feature) that were built using Cox models regularized with the least absolute shrinkage and selection operator. We compared their prognostic performances using stratified fivefold cross-validation. In a median follow-up of 61 months, 110/437 patients experienced BCR. CDLM-Deep feature achieved the best performance (hazard ratio [HR] = 7.72), followed by DLM-Deep feature (HR = 4.37) or RM-Multi (HR = 2.67). CRM-Multi performed moderately. Our results confirm the superior performance of our mpMRI-derived DL algorithm over conventional radiomics.

Application of Spectral Algorithm Applied to Spatially Registered Bi-Parametric MRI to Predict Prostate Tumor Aggressiveness: A Pilot Study

Background: Current prostate cancer evaluation can be inaccurate and burdensome. Quantitative evaluation of Magnetic Resonance Imaging (MRI) sequences non-invasively helps prostate tumor assessment. However, including Dynamic Contrast Enhancement (DCE) in the examined MRI sequence set can add complications, inducing possible side effects from the IV placement or injected contrast material and prolonging scanning time. More accurate quantitative MRI without DCE and artificial intelligence approaches are needed. Purpose: Predict the risk of developing Clinically Significant (Insignificant) prostate cancer CsPCa (CiPCa) and correlate with the International Society of Urologic Pathology (ISUP) grade using processed Signal to Clutter Ratio (SCR) derived from spatially registered bi-parametric MRI (SRBP-MRI) and thereby enhance non-invasive management of prostate cancer. Methods: This pilot study retrospectively analyzed 42 consecutive prostate cancer patients from the PI-CAI data collection. BP-MRI (Apparent Diffusion Coefficient, High B-value, T2) were resized, translated, cropped, and stitched to form spatially registered SRBP-MRI. Efficacy of noise reduction was tested by regularizing, eliminating principal components (PC), and minimizing elliptical volume from the covariance matrix to optimize the SCR. MRI guided biopsy (MRBx), Systematic Biopsy (SysBx), combination (MRBx + SysBx), or radical prostatectomy determined the ISUP grade for each patient. ISUP grade ≥ 2 (<2) was judged as CsPCa (CiPCa). Linear and logistic regression were fitted to ISUP grade and CsPCa/CiPCa SCR. Correlation Coefficients (R) and Area Under the Curves (AUC) for Receiver Operator Curves (ROC) evaluated the performance. Results: High correlation coefficients (R) (>0.55) and high AUC (=1.0) for linear and/or logistic fit from processed SCR and z-score for SRBP-MRI greatly exceed fits using prostate serum antigen, prostate volume, and patient age (R ~ 0.17). Patients assessed with combined MRBx + SysBx and from individual MRI scanners achieved higher R (DR = 0.207+/-0.118) than all patients used in the fits. Conclusions: In the first study, to date, spectral approaches for assessing tumor aggressiveness on SRBP-MRI have been applied and tested and achieved high values of R and exceptional AUC to fit the ISUP grade and CsPCA/CiPCA, respectively.

A multicenter study of artificial intelligence-aided software for detecting visible clinically significant prostate cancer on mpMRI

BACKGROUND

AI-based software may improve the performance of radiologists when detecting clinically significant prostate cancer (csPCa). This study aims to compare the performance of radiologists in detecting MRI-visible csPCa on MRI with and without AI-based software.

MATERIALS AND METHODS

In total, 480 multiparametric MRI (mpMRI) images were retrospectively collected from eleven different MR devices, with 349 csPCa lesions in 180 (37.5%) cases. The csPCa areas were annotated based on pathology. Sixteen radiologists from four hospitals participated in reading. Each radiologist was randomly assigned to 30 cases and diagnosed twice. Half cases were interpreted without AI, and the other half were interpreted with AI. After four weeks, the cases were read again in switched mode. The mean diagnostic performance was compared using sensitivity and specificity on lesion level and patient level. The median reading time and diagnostic confidence were assessed.

RESULTS

On lesion level, AI-aided improved the sensitivity from 40.1% to 59.0% (18.9% increased; 95% confidence interval (CI) [11.5, 26.1]; p < .001). On patient level, AI-aided improved the specificity from 57.7 to 71.7% (14.0% increase, 95% CI [6.4, 21.4]; p < .001) while preserving the sensitivity (88.3% vs. 93.9%, p = 0.06). AI-aided reduced the median reading time of one case by 56.3% from 423 to 185 s (238-s decrease, 95% CI [219, 260]; p < .001), and the median diagnostic confidence score was increased by 10.3% from 3.9 to 4.3 (0.4-score increase, 95% CI [0.3, 0.5]; p < .001).

CONCLUSIONS

AI software improves the performance of radiologists by reducing false positive detection of prostate cancer patients and also improving reading times and diagnostic confidence.

CLINICAL RELEVANCE STATEMENT

This study involves the process of data collection, randomization and crossover reading procedure.

Value of handcrafted and deep radiomic features towards training robust machine learning classifiers for prediction of prostate cancer disease aggressiveness

There is a growing piece of evidence that artificial intelligence may be helpful in the entire prostate cancer disease continuum. However, building machine learning algorithms robust to inter- and intra-radiologist segmentation variability is still a challenge. With this goal in mind, several model training approaches were compared: removing unstable features according to the intraclass correlation coefficient (ICC); training independently with features extracted from each radiologist's mask; training with the feature average between both radiologists; extracting radiomic features from the intersection or union of masks; and creating a heterogeneous dataset by randomly selecting one of the radiologists' masks for each patient. The classifier trained with this last resampled dataset presented with the lowest generalization error, suggesting that training with heterogeneous data leads to the development of the most robust classifiers. On the contrary, removing features with low ICC resulted in the highest generalization error. The selected radiomics dataset, with the randomly chosen radiologists, was concatenated with deep features extracted from neural networks trained to segment the whole prostate. This new hybrid dataset was then used to train a classifier. The results revealed that, even though the hybrid classifier was less overfitted than the one trained with deep features, it still was unable to outperform the radiomics model.

Biomarkers of Aggressive Prostate Cancer at Diagnosis

In the United States, prostate cancer (CaP) remains the second leading cause of cancer deaths in men. CaP is predominantly indolent at diagnosis, with a small fraction (25-30%) representing an aggressive subtype (Gleason score 7-10) that is prone to metastatic progression. This fact, coupled with the criticism surrounding the role of prostate specific antigen in prostate cancer screening, demonstrates the current need for a biomarker(s) that can identify clinically significant CaP and avoid unnecessary biopsy procedures and psychological implications of being diagnosed with low-risk prostate cancer. Although several diagnostic biomarkers are available to clinicians, very few comparative trials have been performed to assess the clinical effectiveness of these biomarkers. It is of note, however, that a majority of these clinical trials have been over-represented by men of Caucasian origin, despite the fact that African American men have a 1.7 times higher incidence and 2.1 times higher rate of mortality from prostate cancer. Biomarkers for CaP diagnosis based on the tissue of origin include urine-based gene expression assays (PCA3, Select MDx, ExoDx Prostate IntelliScore, Mi-Prostate Score, PCA3-PCGEM1 gene panel), blood-based protein biomarkers (4K, PHI), and tissue-based DNA biomarker (Confirm MDx). Another potential direction that has emerged to aid in the CaP diagnosis include multi-parametric magnetic resonance imaging (mpMRI) and bi-parametric magnetic resonance imaging (bpMRI), which in conjunction with clinically validated biomarkers may provide a better approach to predict clinically significant CaP at diagnosis. In this review, we discuss some of the adjunctive biomarker tests along with newer imaging modalities that are currently available to help clinicians decide which patients are at risk of having high-grade CaP on prostate biopsy with the emphasis on clinical utility of the tests across African American (AA) and Caucasian (CA) men.

Artificial Intelligence for Clinical Diagnosis and Treatment of Prostate Cancer

As medical science and technology progress towards the era of "big data", a multi-dimensional dataset pertaining to medical diagnosis and treatment is becoming accessible for mathematical modelling. However, these datasets are frequently inconsistent, noisy, and often characterized by a significant degree of redundancy. Thus, extensive data processing is widely advised to clean the dataset before feeding it into the mathematical model. In this context, Artificial intelligence (AI) techniques, including machine learning (ML) and deep learning (DL) algorithms based on artificial neural networks (ANNs) and their types, are being used to produce a precise and cross-sectional illustration of clinical data. For prostate cancer patients, datasets derived from the prostate-specific antigen (PSA), MRI-guided biopsies, genetic biomarkers, and the Gleason grading are primarily used for diagnosis, risk stratification, and patient monitoring. However, recording diagnoses and further stratifying risks based on such diagnostic data frequently involves much subjectivity. Thus, implementing an AI algorithm on a PC's diagnostic data can reduce the subjectivity of the process and assist in decision making. In addition, AI is used to cut down the processing time and help with early detection, which provides a superior outcome in critical cases of prostate cancer. Furthermore, this also facilitates offering the service at a lower cost by reducing the amount of human labor. Herein, the prime objective of this review is to provide a deep analysis encompassing the existing AI algorithms that are being deployed in the field of prostate cancer (PC) for diagnosis and treatment. Based on the available literature, AI-powered technology has the potential for extensive growth and penetration in PC diagnosis and treatment to ease and expedite the existing medical process.

Comparative Performance of Deep Learning and Radiologists for the Diagnosis and Localization of Clinically Significant Prostate Cancer at MRI: A Systematic Review

Background: Deep learning (DL)-based models have demonstrated an ability to automatically diagnose clinically significant prostate cancer (PCa) on MRI scans and are regularly reported to approach expert performance. The aim of this work was to systematically review the literature comparing deep learning (DL) systems to radiologists in order to evaluate the comparative performance of current state-of-the-art deep learning models and radiologists. Methods: This systematic review was conducted in accordance with the 2020 Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) checklist. Studies investigating DL models for diagnosing clinically significant (cs) PCa on MRI were included. The quality and risk of bias of each study were assessed using the checklist for AI in medical imaging (CLAIM) and QUADAS-2, respectively. Patient level and lesion-based diagnostic performance were separately evaluated by comparing the sensitivity achieved by DL and radiologists at an identical specificity and the false positives per patient, respectively. Results: The final selection consisted of eight studies with a combined 7337 patients. The median study quality with CLAIM was 74.1% (IQR: 70.6–77.6). DL achieved an identical patient-level performance to the radiologists for PI-RADS ≥ 3 (both 97.7%, SD = 2.1%). DL had a lower sensitivity for PI-RADS ≥ 4 (84.2% vs. 88.8%, p = 0.43). The sensitivity of DL for lesion localization was also between 2% and 12.5% lower than that of the radiologists. Conclusions: DL models for the diagnosis of csPCa on MRI appear to approach the performance of experts but currently have a lower sensitivity compared to experienced radiologists. There is a need for studies with larger datasets and for validation on external data.

Serum PSA-based early detection of prostate cancer in Europe and globally: past, present and future

In the pre-PSA-detection era, a large proportion of men were diagnosed with metastatic prostate cancer and died of the disease; after the introduction of the serum PSA test, randomized controlled prostate cancer screening trials in the USA and Europe were conducted to assess the effect of PSA screening on prostate cancer mortality. Contradictory outcomes of the trials and the accompanying overdiagnosis resulted in recommendations against prostate cancer screening by organizations such as the United States Preventive Services Task Force. These recommendations were followed by a decline in PSA testing and a rise in late-stage diagnosis and prostate cancer mortality. Re-evaluation of the randomized trials, which accounted for contamination, showed that PSA-based screening does indeed reduce prostate cancer mortality; however, the debate about whether to screen or not to screen continues because of the considerable overdiagnosis that occurs using PSA-based screening. Meanwhile, awareness among the population of prostate cancer as a potentially lethal disease stimulates opportunistic screening practices that further increase overdiagnosis without the benefit of mortality reduction. However, in the past decade, new screening tools have been developed that make the classic PSA-only-based screening an outdated strategy. With improved use of PSA, in combination with age, prostate volume and with the application of prostate cancer risk calculators, a risk-adapted strategy enables improved stratification of men with prostate cancer and avoidance of unnecessary diagnostic procedures. This combination used with advanced detection techniques (such as MRI and targeted biopsy), can reduce overdiagnosis. Moreover, new biomarkers are becoming available and will enable further improvements in risk stratification.

Machine learning prediction of prostate cancer from transrectal ultrasound video clips

Objective

To build a machine learning (ML) prediction model for prostate cancer (PCa) from transrectal ultrasound video clips of the whole prostate gland, diagnostic performance was compared with magnetic resonance imaging (MRI).

Methods

We systematically collated data from 501 patients—276 with prostate cancer and 225 with benign lesions. From a final selection of 231 patients (118 with prostate cancer and 113 with benign lesions), we randomly chose 170 for the purpose of training and validating a machine learning model, while using the remaining 61 to test a derived model. We extracted 851 features from ultrasound video clips. After dimensionality reduction with the least absolute shrinkage and selection operator (LASSO) regression, 14 features were finally selected and the support vector machine (SVM) and random forest (RF) algorithms were used to establish radiomics models based on those features. In addition, we creatively proposed a machine learning models aided diagnosis algorithm (MLAD) composed of SVM, RF, and radiologists’ diagnosis based on MRI to evaluate the performance of ML models in computer-aided diagnosis (CAD). We evaluated the area under the curve (AUC) as well as the sensitivity, specificity, and precision of the ML models and radiologists’ diagnosis based on MRI by employing receiver operator characteristic curve (ROC) analysis.

Results

The AUC, sensitivity, specificity, and precision of the SVM in the diagnosis of PCa in the validation set and the test set were 0.78, 63%, 80%; 0.75, 65%, and 67%, respectively. Additionally, the SVM model was found to be superior to senior radiologists’ (SR, more than 10 years of experience) diagnosis based on MRI (AUC, 0.78 vs. 0.75 in the validation set and 0.75 vs. 0.72 in the test set), and the difference was statistically significant (p< 0.05).

Conclusion

The prediction model constructed by the ML algorithm has good diagnostic efficiency for prostate cancer. The SVM model’s diagnostic efficiency is superior to that of MRI, as it has a more focused application value. Overall, these prediction models can aid radiologists in making better diagnoses.

Quantib Prostate Compared to an Expert Radiologist for the Diagnosis of Prostate Cancer on mpMRI: A Single-Center Preliminary Study

Background: To evaluate the clinical utility of an Artificial Intelligence (AI) radiology solution, Quantib Prostate, for prostate cancer (PCa) lesions detection on multiparametric Magnetic Resonance Images (mpMRI). Methods: Prostate mpMRI exams of 108 patients were retrospectively studied. The diagnostic performance of an expert radiologist (>8 years of experience) and of an inexperienced radiologist aided by Quantib software were compared. Three groups of patients were assessed: patients with positive mpMRI, positive target biopsy, and/or at least one positive random biopsy (group A, 73 patients); patients with positive mpMRI and a negative biopsy (group B, 14 patients), and patients with negative mpMRI who did not undergo biopsy (group-C, 21 patients). Results: In group A, the AI-assisted radiologist found new lesions with positive biopsy correlation, increasing the diagnostic PCa performance when compared with the expert radiologist, reaching an SE of 92.3% and a PPV of 90.1% (vs. 71.7% and 84.4%). In group A, the expert radiologist found 96 lesions on 73 mpMRI exams (17.7% PIRADS3, 56.3% PIRADS4, and 26% PIRADS5). The AI-assisted radiologist found 121 lesions (0.8% PIRADS3, 53.7% PIRADS4, and 45.5% PIRADS5). At biopsy, 33.9% of the lesions were ISUP1, 31.4% were ISUP2, 22% were ISUP3, 10.2% were ISUP4, and 2.5% were ISUP5. In group B, where biopsies were negative, the AI-assisted radiologist excluded three lesions but confirmed all the others. In group-C, the AI-assisted radiologist found 37 new lesions, most of them PIRADS 3, with 32.4% localized in the peripherical zone and 67.6% in the transition zone. Conclusions: Quantib software is a very sensitive tool to use specifically in high-risk patients (high PIRADS and high Gleason score).

Artificial intelligence for prostate MRI: open datasets, available applications, and grand challenges

Artificial intelligence (AI) for prostate magnetic resonance imaging (MRI) is starting to play a clinical role for prostate cancer (PCa) patients. AI-assisted reading is feasible, allowing workflow reduction. A total of 3,369 multi-vendor prostate MRI cases are available in open datasets, acquired from 2003 to 2021 in Europe or USA at 3 T (n = 3,018; 89.6%) or 1.5 T (n = 296; 8.8%), 346 cases scanned with endorectal coil (10.3%), 3,023 (89.7%) with phased-array surface coils; 412 collected for anatomical segmentation tasks, 3,096 for PCa detection/classification; for 2,240 cases lesions delineation is available and 56 cases have matching histopathologic images; for 2,620 cases the PSA level is provided; the total size of all open datasets amounts to approximately 253 GB. Of note, quality of annotations provided per dataset highly differ and attention must be paid when using these datasets (e.g., data overlap). Seven grand challenges and commercial applications from eleven vendors are here considered. Few small studies provided prospective validation. More work is needed, in particular validation on large-scale multi-institutional, well-curated public datasets to test general applicability. Moreover, AI needs to be explored for clinical stages other than detection/characterization (e.g., follow-up, prognosis, interventions, and focal treatment).

Tasks for artificial intelligence in prostate MRI

The advent of precision medicine, increasing clinical needs, and imaging availability among many other factors in the prostate cancer diagnostic pathway has engendered the utilization of artificial intelligence (AI). AI carries a vast number of potential applications in every step of the prostate cancer diagnostic pathway from classifying/improving prostate multiparametric magnetic resonance image quality, prostate segmentation, anatomically segmenting cancer suspicious foci, detecting and differentiating clinically insignificant cancers from clinically significant cancers on a voxel-level, and classifying entire lesions into Prostate Imaging Reporting and Data System categories/Gleason scores. Multiple studies in all these areas have shown many promising results approximating accuracies of radiologists. Despite this flourishing research, more prospective multicenter studies are needed to uncover the full impact and utility of AI on improving radiologist performance and clinical management of prostate cancer. In this narrative review, we aim to introduce emerging medical imaging AI paper quality metrics such as the Checklist for Artificial Intelligence in Medical Imaging (CLAIM) and Field-Weighted Citation Impact (FWCI), dive into some of the top AI models for segmentation, detection, and classification.

The Mount Sinai Prebiopsy Risk Calculator for Predicting any Prostate Cancer and Clinically Significant Prostate Cancer: Development of a Risk Predictive Tool and Validation with Advanced Neural Networking, Prostate Magnetic Resonance Imaging Outcome Database, and European Randomized Study of Screening for Prostate Cancer Risk Calculator

Background

The European Association of Urology guidelines recommend the use of imaging, biomarkers, and risk calculators in men at risk of prostate cancer. Risk predictive calculators that combine multiparametric magnetic resonance imaging with prebiopsy variables aid as an individualized decision-making tool for patients at risk of prostate cancer, and advanced neural networking increases reliability of these tools.

Objective

To develop a comprehensive risk predictive online web-based tool using magnetic resonance imaging (MRI) and clinical data, to predict the risk of any prostate cancer (PCa) and clinically significant PCa (csPCa) applicable to biopsy-naïve men, men with a prior negative biopsy, men with prior positive low-grade cancer, and men with negative MRI.

Design setting and participants

Institutional review board-approved prospective data of 1902 men undergoing biopsy from October 2013 to September 2021 at Mount Sinai were collected.

Outcome measurements and statistical analysis

Univariable and multivariable analyses were used to evaluate clinical variables such as age, race, digital rectal examination, family history, prostate-specific antigen (PSA), biopsy status, Prostate Imaging Reporting and Data System score, and prostate volume, which emerged as predictors for any PCa and csPCa. Binary logistic regression was performed to study the probability. Validation was performed with advanced neural networking (ANN), multi-institutional European cohort (Prostate MRI Outcome Database [PROMOD]), and European Randomized Study of Screening for Prostate Cancer Risk Calculator (ERSPC RC) 3/4.

Results and limitations

Overall, 2363 biopsies had complete clinical information, with 57.98% any cancer and 31.40% csPCa. The prediction model was significantly associated with both any PCa and csPCa having an area under the curve (AUC) of 81.9% including clinical data. The AUC for external validation was calculated in PROMOD, ERSPC RC, and ANN for any PCa (0.82 vs 0.70 vs 0.90) and csPCa (0.82 vs 0.78 vs 0.92), respectively. This study is limited by its retrospective design and overestimation of csPCa in the PROMOD cohort.

Conclusions

The Mount Sinai Prebiopsy Risk Calculator combines PSA, imaging and clinical data to predict the risk of any PCa and csPCa for all patient settings. With accurate validation results in a large European cohort, ERSPC RC, and ANN, it exhibits its efficiency and applicability in a more generalized population. This calculator is available online in the form of a free web-based tool that can aid clinicians in better patients counseling and treatment decision-making.

Patient summary

We developed the Mount Sinai Prebiopsy Risk Calculator (MSP-RC) to assess the likelihood of any prostate cancer and clinically significant disease based on a combination of clinical and imaging characteristics. MSP-RC is applicable to all patient settings and accessible online.

Artificial Intelligence in Oncology: Current Capabilities, Future Opportunities, and Ethical Considerations

The promise of highly personalized oncology care using artificial intelligence (AI) technologies has been forecasted since the emergence of the field. Cumulative advances across the science are bringing this promise to realization, including refinement of machine learning- and deep learning algorithms; expansion in the depth and variety of databases, including multiomics; and the decreased cost of massively parallelized computational power. Examples of successful clinical applications of AI can be found throughout the cancer continuum and in multidisciplinary practice, with computer vision-assisted image analysis in particular having several U.S. Food and Drug Administration-approved uses. Techniques with emerging clinical utility include whole blood multicancer detection from deep sequencing, virtual biopsies, natural language processing to infer health trajectories from medical notes, and advanced clinical decision support systems that combine genomics and clinomics. Substantial issues have delayed broad adoption, with data transparency and interpretability suffering from AI's "black box" mechanism, and intrinsic bias against underrepresented persons limiting the reproducibility of AI models and perpetuating health care disparities. Midfuture projections of AI maturation involve increasing a model's complexity by using multimodal data elements to better approximate an organic system. Far-future positing includes living databases that accumulate all aspects of a person's health into discrete data elements; this will fuel highly convoluted modeling that can tailor treatment selection, dose determination, surveillance modality and schedule, and more. The field of AI has had a historical dichotomy between its proponents and detractors. The successful development of recent applications, and continued investment in prospective validation that defines their impact on multilevel outcomes, has established a momentum of accelerated progress.

Comparative performance of fully-automated and semi-automated artificial intelligence methods for the detection of clinically significant prostate cancer on MRI: a systematic review

OBJECTIVES

We systematically reviewed the current literature evaluating the ability of fully-automated deep learning (DL) and semi-automated traditional machine learning (TML) MRI-based artificial intelligence (AI) methods to differentiate clinically significant prostate cancer (csPCa) from indolent PCa (iPCa) and benign conditions.

METHODS

We performed a computerised bibliographic search of studies indexed in MEDLINE/PubMed, arXiv, medRxiv, and bioRxiv between 1 January 2016 and 31 July 2021. Two reviewers performed the title/abstract and full-text screening. The remaining papers were screened by four reviewers using the Checklist for Artificial Intelligence in Medical Imaging (CLAIM) for DL studies and Radiomics Quality Score (RQS) for TML studies. Papers that fulfilled the pre-defined screening requirements underwent full CLAIM/RQS evaluation alongside the risk of bias assessment using QUADAS-2, both conducted by the same four reviewers. Standard measures of discrimination were extracted for the developed predictive models.

RESULTS

17/28 papers (five DL and twelve TML) passed the quality screening and were subject to a full CLAIM/RQS/QUADAS-2 assessment, which revealed a substantial study heterogeneity that precluded us from performing quantitative analysis as part of this review. The mean RQS of TML papers was 11/36, and a total of five papers had a high risk of bias. AUCs of DL and TML papers with low risk of bias ranged between 0.80-0.89 and 0.75-0.88, respectively.

CONCLUSION

We observed comparable performance of the two classes of AI methods and identified a number of common methodological limitations and biases that future studies will need to address to ensure the generalisability of the developed models.

Quality in MR reporting of the prostate – improving acquisition, the role of AI and future perspectives

The high quality of MRI reporting of the prostate is the most critical component of the service provided by a radiologist. Prostate MRI structured reporting with PI-RADS v. 2.1 has been proven to improve consistency, quality, guideline-based care in the management of prostate cancer. There is room for improved accuracy of prostate mpMRI reporting, particularly as PI-RADS core criteria are subjective for radiologists. The application of artificial intelligence may support radiologists in interpreting MRI scans. This review addresses the quality of prostate multiparametric MRI (mpMRI) structured reporting (include improvements in acquisition using artificial intelligence) in terms of size of prostate gland, imaging quality, lesion location, lesion size, TNM staging, sector map, and discusses the future prospects of quality in MR reporting.

Deep learning-based artificial intelligence applications in prostate MRI: brief summary

Prostate cancer (PCa) is the most common cancer type in males in the Western World. MRI has an established role in diagnosis of PCa through guiding biopsies. Due to multistep complex nature of the MRI-guided PCa diagnosis pathway, diagnostic performance has a big variation. Developing artificial intelligence (AI) models using machine learning, particularly deep learning, has an expanding role in radiology. Specifically, for prostate MRI, several AI approaches have been defined in the literature for prostate segmentation, lesion detection and classification with the aim of improving diagnostic performance and interobserver agreement. In this review article, we summarize the use of radiology applications of AI in prostate MRI.

Machine Learning in Prostate MRI for Prostate Cancer: Current Status and Future Opportunities

Advances in our understanding of the role of magnetic resonance imaging (MRI) for the detection of prostate cancer have enabled its integration into clinical routines in the past two decades. The Prostate Imaging Reporting and Data System (PI-RADS) is an established imaging-based scoring system that scores the probability of clinically significant prostate cancer on MRI to guide management. Image fusion technology allows one to combine the superior soft tissue contrast resolution of MRI, with real-time anatomical depiction using ultrasound or computed tomography. This allows the accurate mapping of prostate cancer for targeted biopsy and treatment. Machine learning provides vast opportunities for automated organ and lesion depiction that could increase the reproducibility of PI-RADS categorisation, and improve co-registration across imaging modalities to enhance diagnostic and treatment methods that can then be individualised based on clinical risk of malignancy. In this article, we provide a comprehensive and contemporary review of advancements, and share insights into new opportunities in this field.

Machine and Deep Learning Prediction Of Prostate Cancer Aggressiveness Using Multiparametric MRI

Prostate cancer (PCa) is the most frequent male malignancy and the assessment of PCa aggressiveness, for which a biopsy is required, is fundamental for patient management. Currently, multiparametric (mp) MRI is strongly recommended before biopsy. Quantitative assessment of mpMRI might provide the radiologist with an objective and noninvasive tool for supporting the decision-making in clinical practice and decreasing intra- and inter-reader variability. In this view, high dimensional radiomics features and Machine Learning (ML) techniques, along with Deep Learning (DL) methods working on raw images directly, could assist the radiologist in the clinical workflow. The aim of this study was to develop and validate ML/DL frameworks on mpMRI data to characterize PCas according to their aggressiveness. We optimized several ML/DL frameworks on T2w, ADC and T2w+ADC data, using a patient-based nested validation scheme. The dataset was composed of 112 patients (132 peripheral lesions with Prostate Imaging Reporting and Data System (PI-RADS) score ≥ 3) acquired following both PI-RADS 2.0 and 2.1 guidelines. Firstly, ML/DL frameworks trained and validated on PI-RADS 2.0 data were tested on both PI-RADS 2.0 and 2.1 data. Then, we trained, validated and tested ML/DL frameworks on a multi PI-RADS dataset. We reported the performances in terms of Area Under the Receiver Operating curve (AUROC), specificity and sensitivity. The ML/DL frameworks trained on T2w data achieved the overall best performance. Notably, ML and DL frameworks trained and validated on PI-RADS 2.0 data obtained median AUROC values equal to 0.750 and 0.875, respectively, on unseen PI-RADS 2.0 test set. Similarly, ML/DL frameworks trained and validated on multi PI-RADS T2w data showed median AUROC values equal to 0.795 and 0.750, respectively, on unseen multi PI-RADS test set. Conversely, all the ML/DL frameworks trained and validated on PI-RADS 2.0 data, achieved AUROC values no better than the chance level when tested on PI-RADS 2.1 data. Both ML/DL techniques applied on mpMRI seem to be a valid aid in predicting PCa aggressiveness. In particular, ML/DL frameworks fed with T2w images data (objective, fast and non-invasive) show good performances and might support decision-making in patient diagnostic and therapeutic management, reducing intra- and inter-reader variability.

Artificial Intelligence (AI) for Automated Cancer Detection on Prostate MRI: Opportunities and Ongoing Challenges, From the AJR Special Series on AI Applications

Prostate MRI utilization has increased greatly in the past decade, serving primarily to direct targeted prostate biopsies. However, prostate MRI interpretation remains prone to consideration interreader variation. Artificial intelligence (AI) has the potential to standardize the detection of lesions on MRI that are suspicious for prostate cancer (PCa). This review explores the current status of AI for the automated detection of PCa by MRI. Recent literature providing promising results for AI models for PCa detection by MRI is highlighted. Numerous limitations of the existing literature are also described, including biases in model validation, heterogeneity in reporting of performance metrics, and lack of sufficient evidence of clinical translation. Challenges related to AI ethics and data governance are also discussed. An outlook is provided for AI in lesion detection by prostate MRI in the coming years, emphasizing current research needs. Future investigations, incorporating large-scale diverse multi-institutional training and testing datasets, are anticipated to enable the development of more robust AI models for PCa detection by MRI, though prospective clinical trials will ultimately be required to establish benefit of AI in patient management.

Deep learning-assisted prostate cancer detection on bi-parametric MRI: minimum training data size requirements and effect of prior knowledge

Objectives

To assess Prostate Imaging Reporting and Data System (PI-RADS)–trained deep learning (DL) algorithm performance and to investigate the effect of data size and prior knowledge on the detection of clinically significant prostate cancer (csPCa) in biopsy-naïve men with a suspicion of PCa.

Methods

Multi-institution data included 2734 consecutive biopsy-naïve men with elevated PSA levels (≥ 3 ng/mL) that underwent multi-parametric MRI (mpMRI). mpMRI exams were prospectively reported using PI-RADS v2 by expert radiologists. A DL framework was designed and trained on center 1 data (n = 1952) to predict PI-RADS ≥ 4 (n = 1092) lesions from bi-parametric MRI (bpMRI). Experiments included varying the number of cases and the use of automatic zonal segmentation as a DL prior. Independent center 2 cases (n = 296) that included pathology outcome (systematic and MRI targeted biopsy) were used to compute performance for radiologists and DL. The performance of detecting PI-RADS 4–5 and Gleason > 6 lesions was assessed on 782 unseen cases (486 center 1, 296 center 2) using free-response ROC (FROC) and ROC analysis.

Results

The DL sensitivity for detecting PI-RADS ≥ 4 lesions was 87% (193/223, 95% CI: 82–91) at an average of 1 false positive (FP) per patient, and an AUC of 0.88 (95% CI: 0.84–0.91). The DL sensitivity for the detection of Gleason > 6 lesions was 85% (79/93, 95% CI: 77–83) @ 1 FP compared to 91% (85/93, 95% CI: 84–96) @ 0.3 FP for a consensus panel of expert radiologists. Data size and prior zonal knowledge significantly affected performance (4%, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$p<0.05$$\end{document}p<0.05).

Conclusion

PI-RADS-trained DL can accurately detect and localize Gleason > 6 lesions. DL could reach expert performance using substantially more than 2000 training cases, and DL zonal segmentation.

Key Points

• AI for prostate MRI analysis depends strongly on data size and prior zonal knowledge.

• AI needs substantially more than 2000 training cases to achieve expert performance.

Supplementary Information

The online version contains supplementary material available at 10.1007/s00330-021-08320-y.

Multiparametric prostate MRI quality assessment using a semi-automated PI-QUAL software program

The technical requirements for the acquisition of multiparametric magnetic resonance imaging (mpMRI) of the prostate have been clearly outlined in the Prostate Imaging Reporting and Data System (PI-RADS) guidelines, but there is still huge variability in image quality among centres across the world. It has been difficult to quantify what constitutes a good-quality image, and a first attempt to address this matter has been the publication of the Prostate Imaging Quality (PI-QUAL) score and its dedicated scoring sheet. This score includes the assessment of technical parameters that can be obtained from the DICOM files along with a visual evaluation of certain features on prostate MRI (e.g., anatomical structures). We retrospectively analysed the image quality of 10 scans from different vendors and magnets using a semiautomated dedicated PI-QUAL software program and compared the time needed for assessing image quality using two methods (semiautomated assessment versus manual filling of the scoring sheet). This semiautomated software is able to assess the technical parameters automatically, but the visual assessment is still performed by the radiologist. There was a significant reduction in the reporting time of prostate mpMRI quality according to PI-QUAL using the dedicated software program compared to manual filling (5'54″ versus 7'59″; p = 0.005). A semiautomated PI-QUAL software program allows the radiologist to assess the technical details related to the image quality of prostate mpMRI in a quick and reliable manner, allowing clinicians to have more confidence that the quality of mpMRI of the prostate is sufficient to determine patient care.

Prostate-specific Antigen Testing as Part of a Risk-Adapted Early Detection Strategy for Prostate Cancer: European Association of Urology Position and Recommendations for 2021

BACKGROUND

Recommendations against prostate-specific antigen (PSA) testing in 2012 have increased advanced-stage diagnosis and prostate cancer-specific mortality rates.

OBJECTIVE

To present the position of the European Association of Urology (EAU) in 2021 and provide recommendations for the use of PSA testing as part of a risk-adapted strategy for the early detection of prostate cancer.

EVIDENCE ACQUISITION

The authors combined their review of relevant literature, including the EAU prostate cancer guidelines 2021 update, with their own knowledge to provide an expert opinion, representing the EAU's position in 2021.

EVIDENCE SYNTHESIS

The EAU has developed a risk-adapted early prostate cancer detection strategy for well-informed men based on PSA testing, risk calculators, and multiparametric magnetic resonance imaging, which can differentiate significant from insignificant prostate cancer. This approach largely avoids the overdiagnosis/overtreatment of men unlikely to experience disease-related symptoms during their lifetime and facilitates an early diagnosis of men with significant cancer to receive active treatment. It also reduces advanced-stage diagnosis, thereby potentially reducing prostate cancer-specific mortality and improving quality of life. Education is required among urologists, general practitioners, radiologists, policy makers, and healthy men, including endorsement by the European Commission to adapt the European Council's screening recommendations in its 2022 plan and requests to individual countries for its incorporation into national cancer plans.

CONCLUSIONS

This risk-adapted approach for the early detection of prostate cancer will reverse current unfavourable trends and ultimately save lives.

PATIENT SUMMARY

The European Association of Urology has developed a patient information leaflet and algorithm for the early diagnosis of prostate cancer. It can identify men who do not need magnetic resonance imaging or a biopsy and those who would not show any symptoms versus those with more aggressive disease who require further tests/treatment. We need to raise awareness of this algorithm to ensure that all well-informed men at risk of significant prostate cancer are offered a prostate-specific antigen test.