Machine learning classifiers can predict Gleason pattern 4 prostate cancer with greater accuracy than experienced radiologists

Should systematic prostatic biopsies be discontinued?

INTRODUCTION

The use of systematic biopsies in addition to targeted biopsies is based on multiple studies showing that 15-20% of "clinically significant" cancers are missed on targeted biopsies. Concern about these 'missed' cancers drives many interventions. This includes systematic biopsies in men with negative imaging and in men having targeted biopsies, and drives a preference for total gland treatment in men who may be candidates for partial gland ablation. This article summarizes recent genomic and clinical data indicating that, despite "clinically significant" histology, MRI invisible lesions are genomically and clinically favorable. These studies have demonstrated that the genetic aberrations associated with cancer visibility are the same aberrations that drive cancer invasiveness and metastasis. Thus invisible cancers, even if undiagnosed at baseline, are in most cases indolent and pose little threat to the patient. The implications are that patients should be monitored with imaging rather than systematic biopsy, and subject to repeat targeted biopsy for evidence of MR progression. Patients prefer this strategy. It has many advantages in terms of reduced burden of care, cost, psychological benefits, and less diagnosis of insignificant cancer.

CONCLUSION

It is now appropriate to abandon systematic biopsies in most patients.

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.

Radiomics from multisite MRI and clinical data to predict clinically significant prostate cancer

BACKGROUND

Magnetic resonance imaging (MRI) is useful in the diagnosis of clinically significant prostate cancer (csPCa). MRI-derived radiomics may support the diagnosis of csPCa.

PURPOSE

To investigate whether adding radiomics from biparametric MRI to predictive models based on clinical and MRI parameters improves the prediction of csPCa in a multisite-multivendor setting.

MATERIAL AND METHODS

Clinical information (PSA, PSA density, prostate volume, and age), MRI reviews (PI-RADS 2.1), and radiomics (histogram and texture features) were retrieved from prospectively included patients examined at different radiology departments and with different MRI systems, followed by MRI-ultrasound fusion guided biopsies of lesions PI-RADS 3-5. Predictive logistic regression models of csPCa (Gleason score ≥7) for the peripheral (PZ) and transition zone (TZ), including clinical data and PI-RADS only, and combined with radiomics, were built and compared using receiver operating characteristic (ROC) curves.

RESULTS

In total, 456 lesions in 350 patients were analyzed. In PZ and TZ, PI-RADS 4-5 and PSA density, and age in PZ, were independent predictors of csPCa in models without radiomics. In models including radiomics, PI-RADS 4-5, PSA density, age, and ADC energy were independent predictors in PZ, and PI-RADS 5, PSA density and ADC mean in TZ. Comparison of areas under the ROC curve (AUC) for the models without radiomics (PZ: AUC = 0.82, TZ: AUC = 0.80) versus with radiomics (PZ: AUC = 0.82, TZ: AUC = 0.82) showed no significant differences (PZ: P = 0.366; TZ: P = 0.171).

CONCLUSION

PSA density and PI-RADS are potent predictors of csPCa. Radiomics do not add significant information to our multisite-multivendor dataset.

Combiner and HyperCombiner networks: Rules to combine multimodality MR images for prostate cancer localisation

One of the distinct characteristics of radiologists reading multiparametric prostate MR scans, using reporting systems like PI-RADS v2.1, is to score individual types of MR modalities, including T2-weighted, diffusion-weighted, and dynamic contrast-enhanced, and then combine these image-modality-specific scores using standardised decision rules to predict the likelihood of clinically significant cancer. This work aims to demonstrate that it is feasible for low-dimensional parametric models to model such decision rules in the proposed Combiner networks, without compromising the accuracy of predicting radiologic labels. First, we demonstrate that either a linear mixture model or a nonlinear stacking model is sufficient to model PI-RADS decision rules for localising prostate cancer. Second, parameters of these combining models are proposed as hyperparameters, weighing independent representations of individual image modalities in the Combiner network training, as opposed to end-to-end modality ensemble. A HyperCombiner network is developed to train a single image segmentation network that can be conditioned on these hyperparameters during inference for much-improved efficiency. Experimental results based on 751 cases from 651 patients compare the proposed rule-modelling approaches with other commonly-adopted end-to-end networks, in this downstream application of automating radiologist labelling on multiparametric MR. By acquiring and interpreting the modality combining rules, specifically the linear-weights or odds ratios associated with individual image modalities, three clinical applications are quantitatively presented and contextualised in the prostate cancer segmentation application, including modality availability assessment, importance quantification and rule discovery.

Sliding Window-Based Machine Learning for Environmental Inspection Resource Allocation

Environmental regulation is pivotal in mitigating environmental risks and promoting sustainable development, yet regulators frequently encounter resource constraints when inspecting enterprises. To address this limitation, we employed four sliding window-based machine learning techniques to enhance effective environmental inspections. Utilizing feature-engineered time-series data of characteristic and compliance records from 16,777 chemical enterprises in Jiangsu between 2010 and 2021, the four models were used to establish the predictive models that link enterprise characteristics to the likelihood of inspection failure. The results indicated that the models were comparable with the widely utilized deep learning sequence model, the Long Short-Term Memory model, achieving areas under the ROC (receiver operating characteristic) curves exceeding 0.83. Past violation significantly influences future violations, with a more recent violation history exerting stronger impacts. Besides, using predicted failure rates, we proposed seven resource allocation scenarios, considering different regulation intensities to target high-risk enterprises. Notably, the provincial-level risk-based method yielded a significant inspection failure rate increase (detecting violation), surpassing 8-fold the baseline. Overall, our study offers regulators an optimized inspection resource allocation method, thereby enhancing the regulatory effectiveness. Moreover, this study demonstrates the potential of window-based machine learning techniques in environmental regulation and highlights the importance of data-driven decision-making for promoting sustainable development.

What benefit can be obtained from magnetic resonance imaging diagnosis with artificial intelligence in prostate cancer compared with clinical assessments?

The present study aimed to explore the potential of artificial intelligence (AI) methodology based on magnetic resonance (MR) images to aid in the management of prostate cancer (PCa). To this end, we reviewed and summarized the studies comparing the diagnostic and predictive performance for PCa between AI and common clinical assessment methods based on MR images and/or clinical characteristics, thereby investigating whether AI methods are generally superior to common clinical assessment methods for the diagnosis and prediction fields of PCa. First, we found that, in the included studies of the present study, AI methods were generally equal to or better than the clinical assessment methods for the risk assessment of PCa, such as risk stratification of prostate lesions and the prediction of therapeutic outcomes or PCa progression. In particular, for the diagnosis of clinically significant PCa, the AI methods achieved a higher summary receiver operator characteristic curve (SROC-AUC) than that of the clinical assessment methods (0.87 vs. 0.82). For the prediction of adverse pathology, the AI methods also achieved a higher SROC-AUC than that of the clinical assessment methods (0.86 vs. 0.75). Second, as revealed by the radiomics quality score (RQS), the studies included in the present study presented a relatively high total average RQS of 15.2 (11.0-20.0). Further, the scores of the individual RQS elements implied that the AI models in these studies were constructed with relatively perfect and standard radiomics processes, but the exact generalizability and clinical practicality of the AI models should be further validated using higher levels of evidence, such as prospective studies and open-testing datasets.

Novel Histopathological Biomarkers in Prostate Cancer: Implications and Perspectives

Prostate cancer (PCa) is the second most frequently diagnosed cancer in men. Despite the significant progress in cancer diagnosis and treatment over the last few years, the approach to disease detection and therapy still does not include histopathological biomarkers. The dissemination of PCa is strictly related to the creation of a premetastatic niche, which can be detected by altered levels of specific biomarkers. To date, the risk factors for biochemical recurrence include lymph node status, prostate-specific antigen (PSA), PSA density (PSAD), body mass index (BMI), pathological Gleason score, seminal vesicle invasion, extraprostatic extension, and intraductal carcinoma. In the future, biomarkers might represent another prognostic factor, as discussed in many studies. In this review, we focus on histopathological biomarkers (particularly CD169 macrophages, neuropilin-1, cofilin-1, interleukin-17, signal transducer and activator of transcription protein 3 (STAT3), LIM domain kinase 1 (LIMK1), CD15, AMACR, prostate-specific membrane antigen (PSMA), Appl1, Sortilin, Syndecan-1, and p63) and their potential application in decision making regarding the prognosis and treatment of PCa patients. We refer to studies that found a correlation between the levels of biomarkers and tumor characteristics as well as clinical outcomes. We also hypothesize about the potential use of histopathological markers as a target for novel immunotherapeutic drugs or targeted radionuclide therapy, which may be used as adjuvant therapy in the future.

Exploring the Use of Artificial Intelligence in the Management of Prostate Cancer

PURPOSE OF REVIEW

This review aims to explore the current state of research on the use of artificial intelligence (AI) in the management of prostate cancer. We examine the various applications of AI in prostate cancer, including image analysis, prediction of treatment outcomes, and patient stratification. Additionally, the review will evaluate the current limitations and challenges faced in the implementation of AI in prostate cancer management.

RECENT FINDINGS

Recent literature has focused particularly on the use of AI in radiomics, pathomics, the evaluation of surgical skills, and patient outcomes. AI has the potential to revolutionize the future of prostate cancer management by improving diagnostic accuracy, treatment planning, and patient outcomes. Studies have shown improved accuracy and efficiency of AI models in the detection and treatment of prostate cancer, but further research is needed to understand its full potential as well as limitations.

The Low Rate of Adherence to Checklist for Artificial Intelligence in Medical Imaging Criteria Among Published Prostate MRI Artificial Intelligence Algorithms

OBJECTIVE

To determine the rigor, generalizability, and reproducibility of published classification and detection artificial intelligence (AI) models for prostate cancer (PCa) on MRI using the Checklist for Artificial Intelligence in Medical Imaging (CLAIM) guidelines, a 42-item checklist that is considered a measure of best practice for presenting and reviewing medical imaging AI research.

MATERIALS AND METHODS

This review searched English literature for studies proposing PCa AI detection and classification models on MRI. Each study was evaluated with the CLAIM checklist. The additional outcomes for which data were sought included measures of AI model performance (eg, area under the curve [AUC], sensitivity, specificity, free-response operating characteristic curves), training and validation and testing group sample size, AI approach, detection versus classification AI, public data set utilization, MRI sequences used, and definition of gold standard for ground truth. The percentage of CLAIM checklist fulfillment was used to stratify studies into quartiles. Wilcoxon's rank-sum test was used for pair-wise comparisons.

RESULTS

In all, 75 studies were identified, and 53 studies qualified for analysis. The original CLAIM items that most studies did not fulfill includes item 12 (77% no): de-identification methods; item 13 (68% no): handling missing data; item 15 (47% no): rationale for choosing ground truth reference standard; item 18 (55% no): measurements of inter- and intrareader variability; item 31 (60% no): inclusion of validated interpretability maps; item 37 (92% no): inclusion of failure analysis to elucidate AI model weaknesses. An AUC score versus percentage CLAIM fulfillment quartile revealed a significant difference of the mean AUC scores between quartile 1 versus quartile 2 (0.78 versus 0.86, P = .034) and quartile 1 versus quartile 4 (0.78 versus 0.89, P = .003) scores. Based on additional information and outcome metrics gathered in this study, additional measures of best practice are defined. These new items include disclosure of public dataset usage, ground truth definition in comparison to other referenced works in the defined task, and sample size power calculation.

CONCLUSION

A large proportion of AI studies do not fulfill key items in CLAIM guidelines within their methods and results sections. The percentage of CLAIM checklist fulfillment is weakly associated with improved AI model performance. Additions or supplementations to CLAIM are recommended to improve publishing standards and aid reviewers in determining study rigor.

Artificial intelligence in prostate cancer: Definitions, current research, and future directions

Multiple novel modalities tasking artificial intelligence based computational pathology applications and integrating other variables, such as risk factors, tumor microenvironment, genomic testing data, laboratory findings, clinical history, and radiology findings, will improve diagnostic consistency and generate a synergistic diagnostic workflow. In this article, we present the concise and contemporary review on the utilization of artificial intelligence in prostate cancer and identify areas for possible future applications.

Prostate MRI: Is Endorectal Coil Necessary?—A Review

To assess the necessity of endorectal coil use in 3 Tesla (T) prostate magnetic resonance imaging (MRI), a literature review comparing the image quality and diagnostic performance with an endorectal coil (ERC) and a without endorectal coil (NERC), with a phased array coil or a wearable perineal coil (WPC), was performed. A PubMed search of 3T prostate MRI using an endorectal coil for studies published until 31 July 2021 was performed. A total of 14 studies comparing 3T prostate MRI with and without endorectal coil use were identified. The quality scores and diagnostic performances were recorded for each study. In total, five studies compared image quality; five studies compared quality and performance; and four studies compared performance of detection, size of detected lesions, accuracy of cancer localization, and aggressiveness/staging. The use of an endorectal coil improved image quality with a higher overall signal to noise ratio, posterior and peripheral zone signal to noise ratio, high b-value attenuation diffusion coefficient (ADC) signal to noise ratio, and contrast to noise ratio. Endorectal coil use improved subjective image quality for anatomic detail on T2 weighted images (T2WI) and diffusion weighted images (DWI). Endorectal coil use had less motion artifact on DWI than non-endorectal coil use, but produced a higher occurrence of other artifacts on DWI. Endorectal coils had higher sensitivity, specificity, and positive predictive value (PPV) in the detection of overall and index lesions, as well as smaller and less aggressive lesions, missing fewer and smaller lesions than non-endorectal coils. Endorectal coils had higher sensitivity than non-endorectal coils in localizing and staging lesions. Endorectal coils improved quantitative and qualitative image quality and diagnostic performance in the detection of smaller and less aggressive cancers in 3T prostate MRI.

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.

Current Value of Biparametric Prostate MRI with Machine-Learning or Deep-Learning in the Detection, Grading, and Characterization of Prostate Cancer: A Systematic Review

Prostate cancer detection with magnetic resonance imaging is based on a standardized MRI-protocol according to the PI-RADS guidelines including morphologic imaging, diffusion weighted imaging, and perfusion. To facilitate data acquisition and analysis the contrast-enhanced perfusion is often omitted resulting in a biparametric prostate MRI protocol. The intention of this review is to analyze the current value of biparametric prostate MRI in combination with methods of machine-learning and deep learning in the detection, grading, and characterization of prostate cancer; if available a direct comparison with human radiologist performance was performed. PubMed was systematically queried and 29 appropriate studies were identified and retrieved. The data show that detection of clinically significant prostate cancer and differentiation of prostate cancer from non-cancerous tissue using machine-learning and deep learning is feasible with promising results. Some techniques of machine-learning and deep-learning currently seem to be equally good as human radiologists in terms of classification of single lesion according to the PIRADS score.

External validation of the computerized analysis of TRUS of the prostate with the ANNA/C-TRUS system: a potential role of artificial intelligence for improving prostate cancer detection

PURPOSE

Prostate cancer (PCa) imaging has been revolutionized by the introduction of multi-parametric Magnetic Resonance Imaging (mpMRI). Transrectal ultrasound (TRUS) has always been considered a low-performance modality. To overcome this, a computerized artificial neural network analysis (ANNA/C-TRUS) of the TRUS based on an artificial intelligence (AI) analysis has been proposed. Our aim was to evaluate the diagnostic performance of the ANNA/C-TRUS system and its ability to improve conventional TRUS in PCa diagnosis.

METHODS

We retrospectively analyzed data from 64 patients with PCa and scheduled for radical prostatectomy who underwent TRUS followed by ANNA/C-TRUS analysis before the procedure. The results of ANNA/C-TRUS analysis with whole mount sections from final pathology.

RESULTS

On a per-sectors analysis, sensitivity, specificity, negative predictive value (NPV), positive predictive value (PPV) and accuracy were 62%, 81%, 80%, 64% and 78% respectively. The values for the detection of clinically significant prostate cancer were 69%, 77%, 88%, 50% and 75%. The diagnostic values for high grade tumours were 70%, 74%, 91%, 41% and 74%, respectively. Cancer volume (≤ 0.5 or greater) did not influence the diagnostic performance of the ANNA/C-TRUS system.

CONCLUSIONS

ANNA/C-TRUS represents a promising diagnostic tool and application of AI for PCa diagnosis. It improves the ability of conventional TRUS to diagnose prostate cancer, preserving its simplicity and availability. Since it is an AI system, it does not hold the inter-observer variability nor a learning curve. Multicenter biopsy-based studies with the inclusion of an adequate number of patients are needed to confirm these results.

Development and Validation of a Radiomics Nomogram for Predicting Clinically Significant Prostate Cancer in PI-RADS 3 Lesions

Purpose

To develop and validate a radiomics nomogram for the prediction of clinically significant prostate cancer (CsPCa) in Prostate Imaging-Reporting and Data System (PI-RADS) category 3 lesions.

Methods

We retrospectively enrolled 306 patients within PI-RADS 3 lesion from January 2015 to July 2020 in institution 1; the enrolled patients were randomly divided into the training group (n = 199) and test group (n = 107). Radiomics features were extracted from T2-weighted imaging (T2WI), apparent diffusion coefficient (ADC) imaging, and dynamic contrast-enhanced (DCE) imaging. Synthetic minority oversampling technique (SMOTE) was used to address the class imbalance. The ANOVA and least absolute shrinkage and selection operator (LASSO) regression model were used for feature selection and radiomics signature building. Then, a radiomics score (Rad-score) was acquired. Combined with serum prostate-specific antigen density (PSAD) level, a multivariate logistic regression analysis was used to construct a radiomics nomogram. Receiver operating characteristic (ROC) curve analysis was used to evaluate radiomics signature and nomogram. The radiomics nomogram calibration and clinical usefulness were estimated through calibration curve and decision curve analysis (DCA). External validation was assessed, and the independent validation cohort contained 65 patients within PI-RADS 3 lesion from January 2020 to July 2021 in institution 2.

Results

A total of 75 (24.5%) and 16 (24.6%) patients had CsPCa in institution 1 and 2, respectively. The radiomics signature with SMOTE augmentation method had a higher area under the ROC curve (AUC) [0.840 (95% CI, 0.776–0.904)] than that without SMOTE method [0.730 (95% CI, 0.624–0.836), p = 0.08] in the test group and significantly increased in the external validation group [0.834 (95% CI, 0.709–0.959) vs. 0.718 (95% CI, 0.562–0.874), p = 0.017]. The radiomics nomogram showed good discrimination and calibration, with an AUC of 0.939 (95% CI, 0.913–0.965), 0.884 (95% CI, 0.831–0.937), and 0.907 (95% CI, 0.814–1) in the training, test, and external validation groups, respectively. The DCA demonstrated the clinical usefulness of radiomics nomogram.

Conclusion

The radiomics nomogram that incorporates the MRI-based radiomics signature and PSAD can be conveniently used to individually predict CsPCa in patients within PI-RADS 3 lesion.

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.

A review of artificial intelligence in prostate cancer detection on imaging

A multitude of studies have explored the role of artificial intelligence (AI) in providing diagnostic support to radiologists, pathologists, and urologists in prostate cancer detection, risk-stratification, and management. This review provides a comprehensive overview of relevant literature regarding the use of AI models in (1) detecting prostate cancer on radiology images (magnetic resonance and ultrasound imaging), (2) detecting prostate cancer on histopathology images of prostate biopsy tissue, and (3) assisting in supporting tasks for prostate cancer detection (prostate gland segmentation, MRI-histopathology registration, MRI-ultrasound registration). We discuss both the potential of these AI models to assist in the clinical workflow of prostate cancer diagnosis, as well as the current limitations including variability in training data sets, algorithms, and evaluation criteria. We also discuss ongoing challenges and what is needed to bridge the gap between academic research on AI for prostate cancer and commercial solutions that improve routine clinical care.

A Combined Radiomics and Machine Learning Approach to Distinguish Clinically Significant Prostate Lesions on a Publicly Available MRI Dataset

Although prostate cancer is one of the most common causes of mortality and morbidity in advancing-age males, early diagnosis improves prognosis and modifies the therapy of choice. The aim of this study was the evaluation of a combined radiomics and machine learning approach on a publicly available dataset in order to distinguish a clinically significant from a clinically non-significant prostate lesion. A total of 299 prostate lesions were included in the analysis. A univariate statistical analysis was performed to prove the goodness of the 60 extracted radiomic features in distinguishing prostate lesions. Then, a 10-fold cross-validation was used to train and test some models and the evaluation metrics were calculated; finally, a hold-out was performed and a wrapper feature selection was applied. The employed algorithms were Naïve bayes, K nearest neighbour and some tree-based ones. The tree-based algorithms achieved the highest evaluation metrics, with accuracies over 80%, and area-under-the-curve receiver-operating characteristics below 0.80. Combined machine learning algorithms and radiomics based on clinical, routine, multiparametric, magnetic-resonance imaging were demonstrated to be a useful tool in prostate cancer stratification.

Artificial Intelligence-assisted Prostate Cancer Diagnosis: Radiologic-Pathologic Correlation

The classic prostate cancer (PCa) diagnostic pathway that is based on prostate-specific antigen (PSA) levels and the findings of digital rectal examination followed by systematic biopsy has shown multiple limitations. The use of multiparametric MRI (mpMRI) is now widely accepted in men with clinical suspicion for PCa. In addition, clinical information, PSA density, risk calculators, and genomic and other "omics" biomarkers are being used to improve risk stratification. On the basis of mpMRI and MRI-targeted biopsies (MRI-TBx), new diagnostic pathways have been established, aiming to improve the limitations of the classic diagnostic approach. However, these pathways still show limitations associated with mpMRI and MRI-TBx. Definitive PCa diagnosis is made on the basis of histopathologic Gleason grading, which has demonstrated an excellent correlation with clinical outcomes. However, Gleason grading is done subjectively by pathologists and involves poor reproducibility, and PCa may have a heterogeneous distribution of histologic patterns. Thus, important discrepancies persist between biopsy tumor grading and final whole-organ pathologic assessment after radical prostatectomy. PCa offers a unique opportunity to establish a real radiologic-pathologic correlation, as whole-mount radical prostatectomy specimens permit a complete spatial relationship with mpMRI. Artificial intelligence is increasingly being applied to radiologic and pathologic images to improve clinical accuracy and efficiency in PCa diagnosis. This review delineates current PCa diagnostic pathways, with a focus on the role of mpMRI, MRI-TBx, and pathologic analysis. An overview of the expected improvements in PCa diagnosis derived from the use of artificial intelligence, integrated radiologic-pathologic systems, and decision support tools for multidisciplinary teams is provided. An invited commentary by Purysko is available online. Online supplemental material is available for this article. ^©RSNA, 2021.

Revisiting DCE-MRI: Classification of Prostate Tissue Using Descriptive Signal Enhancement Features Derived From DCE-MRI Acquisition With High Spatiotemporal Resolution

METHODS

A retrospective study (from January 2016 to July 2019) including 75 subjects (mean, 65 years; 46-80 years) with 2.5-second temporal resolution DCE-MRI and PIRADS 4 or 5 lesions was performed. Fifty-four subjects had biopsy-proven prostate cancer (Gleason 6, 15; Gleason 7, 20; Gleason 8, 13; Gleason 9, 6), whereas 21 subjects had negative MRI/ultrasound fusion-guided biopsies. Voxel-wise analysis of contrast signal enhancement was performed for all time points using custom-developed software, including automatic arterial input function detection. Seven descriptive parameter maps were calculated: normalized maximum signal intensity, time to start, time to maximum, time-to-maximum slope, and maximum slope with normalization on maximum signal and the arterial input function (SMN1, SMN2). The parameters were compared with ADC using multiparametric machine-learning models to determine classification accuracy. A Wilcoxon test was used for the hypothesis test and the Spearman coefficient for correlation.

RESULTS

There were significant differences (P < 0.05) for all 7 DCE-derived parameters between the normal peripheral zone versus PIRADS 4 or 5 lesions and the biopsy-positive versus biopsy-negative lesions. Multiparametric analysis showed better performance when combining ADC + DCE as input (accuracy/sensitivity/specificity, 97%/93%/100%) relative to ADC alone (accuracy/sensitivity/specificity, 94%/95%/95%) and to DCE alone (accuracy/sensitivity/specificity, 78%/79%/77%) in differentiating the normal peripheral zone from PIRADS lesions, biopsy-positive versus biopsy-negative lesions (accuracy/sensitivity/specificity, 68%/33%/81%), and Gleason 6 versus ≥7 prostate cancer (accuracy/sensitivity/specificity, 69%/60%/72%).

CONCLUSIONS

Descriptive perfusion characteristics derived from high-resolution DCE-MRI using model-free computations show significant differences between normal and cancerous tissue but do not reach the accuracy achieved with solely ADC-based classification. Combining ADC with DCE-based input features improved classification accuracy for PIRADS lesions, discrimination of biopsy-positive versus biopsy-negative lesions, and differentiation between Gleason 6 versus Gleason ≥7 lesions.

Particle Swarm Optimized Gaussian Process Classifier for Treatment Discontinuation Prediction in Multicohort Metastatic Castration-Resistant Prostate Cancer Patients

Prostate cancer is the second leading cancer in men, according to the WHO world cancer report. Its prevention and treatment demand proper attention. Despite numerous attempts for disease prevention, prostate tumours can still become metastatic by blood circulation to other organs. Several treatments have been adopted. However, findings show that the docetaxel treatment induces adverse reactions in patients. Particle Swarm Optimized Gaussian Process Classifier (PSO-GPC) is proposed to determine when to discontinue treatment. Based on three cohorts of prostate cancer patients, we propose and compare several classifiers for the best performance in determining treatment discontinuation. Given the data skewness and class imbalance, the models are evaluated based on both the area under receiver operating characteristics curve (AUC) and area under precision recall curve (AUPRC). With the AUCs ranging between 0.6717 - 0.8499, and AUPRCs ranging between 0.1392 - 0.5423, PSO-GPC performs better than the state-of-the-art. We have carried out statistical analysis for ranking methods and analyzed independent cohort data with PSO-GPC, demonstrating its unbiased performance. A proper determination of treatment discontinuation in metastatic castration-resistant prostate cancer patients will reduce the mortality rate in cancer patients.

Detecting Spurious Correlations With Sanity Tests for Artificial Intelligence Guided Radiology Systems

Artificial intelligence (AI) has been successful at solving numerous problems in machine perception. In radiology, AI systems are rapidly evolving and show progress in guiding treatment decisions, diagnosing, localizing disease on medical images, and improving radiologists' efficiency. A critical component to deploying AI in radiology is to gain confidence in a developed system's efficacy and safety. The current gold standard approach is to conduct an analytical validation of performance on a generalization dataset from one or more institutions, followed by a clinical validation study of the system's efficacy during deployment. Clinical validation studies are time-consuming, and best practices dictate limited re-use of analytical validation data, so it is ideal to know ahead of time if a system is likely to fail analytical or clinical validation. In this paper, we describe a series of sanity tests to identify when a system performs well on development data for the wrong reasons. We illustrate the sanity tests' value by designing a deep learning system to classify pancreatic cancer seen in computed tomography scans.

Artificial Intelligence Compared to Radiologists for the Initial Diagnosis of Prostate Cancer on Magnetic Resonance Imaging: A Systematic Review and Recommendations for Future Studies

Computer-aided diagnosis (CAD) of prostate cancer on multiparametric magnetic resonance imaging (mpMRI), using artificial intelligence (AI), may reduce missed cancers and unnecessary biopsies, increase inter-observer agreement between radiologists, and alleviate pressures caused by rising case incidence and a shortage of specialist radiologists to read prostate mpMRI. However, well-designed evaluation studies are required to prove efficacy above current clinical practice. A systematic search of the MEDLINE, EMBASE, and arXiv electronic databases was conducted for studies that compared CAD for prostate cancer detection or classification on MRI against radiologist interpretation and a histopathological reference standard, in treatment-naïve men with a clinical suspicion of prostate cancer. Twenty-seven studies were included in the final analysis. Due to substantial heterogeneities in the included studies, a narrative synthesis is presented. Several studies reported superior diagnostic accuracy for CAD over radiologist interpretation on small, internal patient datasets, though this was not observed in the few studies that performed evaluation using external patient data. Our review found insufficient evidence to suggest the clinical deployment of artificial intelligence algorithms at present. Further work is needed to develop and enforce methodological standards, promote access to large diverse datasets, and conduct prospective evaluations before clinical adoption can be considered.

Applications of artificial intelligence in prostate cancer imaging

PURPOSE OF REVIEW

The purpose of this review was to identify the most recent lines of research focusing on the application of artificial intelligence (AI) in the diagnosis and staging of prostate cancer (PCa) with imaging.

RECENT FINDINGS

The majority of studies focused on the improvement in the interpretation of bi-parametric and multiparametric magnetic resonance imaging, and in the planning of image guided biopsy. These initial studies showed that AI methods based on convolutional neural networks could achieve a diagnostic performance close to that of radiologists. In addition, these methods could improve segmentation and reduce inter-reader variability. Methods based on both clinical and imaging findings could help in the identification of high-grade PCa and more aggressive disease, thus guiding treatment decisions. Though these initial results are promising, only few studies addressed the repeatability and reproducibility of the investigated AI tools. Further, large-scale validation studies are missing and no diagnostic phase III or higher studies proving improved outcomes regarding clinical decision making have been conducted.

SUMMARY

AI techniques have the potential to significantly improve and simplify diagnosis, risk stratification and staging of PCa. Larger studies with a focus on quality standards are needed to allow a widespread introduction of AI in clinical practice.

Computer-aided diagnosis of prostate cancer using multiparametric MRI and clinical features: A patient-level classification framework

Computer-aided diagnosis (CAD) of prostate cancer (PCa) using multiparametric magnetic resonance imaging (mpMRI) is actively being investigated as a means to provide clinical decision support to radiologists. Typically, these systems are trained using lesion annotations. However, lesion annotations are expensive to obtain and inadequate for characterizing certain tumor types e.g. diffuse tumors and MRI invisible tumors. In this work, we introduce a novel patient-level classification framework, denoted PCF, that is trained using patient-level labels only. In PCF, features are extracted from three-dimensional mpMRI and derived parameter maps using convolutional neural networks and subsequently, combined with clinical features by a multi-classifier support vector machine scheme. The output of PCF is a probability value that indicates whether a patient is harboring clinically significant PCa (Gleason score ≥3+4) or not. PCF achieved mean area under the receiver operating characteristic curves of 0.79 and 0.86 on the PICTURE and PROSTATEx datasets respectively, using five-fold cross-validation. Clinical evaluation over a temporally separated PICTURE dataset cohort demonstrated comparable sensitivity and specificity to an experienced radiologist. We envision PCF finding most utility as a second reader during routine diagnosis or as a triage tool to identify low-risk patients who do not require a clinical read.

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

Due to the upfront role of magnetic resonance imaging (MRI) for prostate cancer (PCa) diagnosis, a multitude of artificial intelligence (AI) applications have been suggested to aid in the diagnosis and detection of PCa. In this review, we provide an overview of the current field, including studies between 2018 and February 2021, describing AI algorithms for (1) lesion classification and (2) lesion detection for PCa. Our evaluation of 59 included studies showed that most research has been conducted for the task of PCa lesion classification (66%) followed by PCa lesion detection (34%). Studies showed large heterogeneity in cohort sizes, ranging between 18 to 499 patients (median = 162) combined with different approaches for performance validation. Furthermore, 85% of the studies reported on the stand-alone diagnostic accuracy, whereas 15% demonstrated the impact of AI on diagnostic thinking efficacy, indicating limited proof for the clinical utility of PCa AI applications. In order to introduce AI within the clinical workflow of PCa assessment, robustness and generalizability of AI applications need to be further validated utilizing external validation and clinical workflow experiments.

Rethinking PICO in the Machine Learning Era: ML-PICO

BACKGROUND

Machine learning (ML) has captured the attention of many clinicians who may not have formal training in this area but are otherwise increasingly exposed to ML literature that may be relevant to their clinical specialties. ML papers that follow an outcomes-based research format can be assessed using clinical research appraisal frameworks such as PICO (Population, Intervention, Comparison, Outcome). However, the PICO frameworks strain when applied to ML papers that create new ML models, which are akin to diagnostic tests. There is a need for a new framework to help assess such papers.

OBJECTIVE

We propose a new framework to help clinicians systematically read and evaluate medical ML papers whose aim is to create a new ML model: ML-PICO (Machine Learning, Population, Identification, Crosscheck, Outcomes). We describe how the ML-PICO framework can be applied toward appraising literature describing ML models for health care.

CONCLUSION

The relevance of ML to practitioners of clinical medicine is steadily increasing with a growing body of literature. Therefore, it is increasingly important for clinicians to be familiar with how to assess and best utilize these tools. In this paper we have described a practical framework on how to read ML papers that create a new ML model (or diagnostic test): ML-PICO. We hope that this can be used by clinicians to better evaluate the quality and utility of ML papers.

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

PURPOSE

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

METHODS

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

RESULTS

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

CONCLUSIONS

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

Role of Machine Learning and Artificial Intelligence in Interventional Oncology

PURPOSE OF REVIEW

The purpose of this review is to highlight the current role of machine learning and artificial intelligence and in the field of interventional oncology.

RECENT FINDINGS

With advancements in technology, there is a significant amount of research regarding the application of artificial intelligence and machine learning in medicine. Interventional oncology is a field that can benefit greatly from this research through enhanced image analysis and intraprocedural guidance. These software developments can increase detection of cancers through routine screening and improve diagnostic accuracy in classifying tumors. They may also aid in selecting the most effective treatment for the patient by predicting outcomes based on a combination of both clinical and radiologic factors. Furthermore, machine learning and artificial intelligence can advance intraprocedural guidance for the interventional oncologist through more accurate needle tracking and image fusion technology. This minimizes damage to nearby healthy tissue and maximizes treatment of the tumor. While there are several exciting developments, this review also discusses limitations before incorporating machine learning and artificial intelligence in the field of interventional oncology. These include data capture and processing, lack of transparency among developers, validating models, integrating workflow, and ethical challenged. In summary, machine learning and artificial intelligence have the potential to positively impact interventional oncologists and how they provide cancer care treatments.

Prediction of Pathological Upgrading at Radical Prostatectomy in Prostate Cancer Eligible for Active Surveillance: A Texture Features and Machine Learning-Based Analysis of Apparent Diffusion Coefficient Maps

Objective

To evaluate a combination of texture features and machine learning-based analysis of apparent diffusion coefficient (ADC) maps for the prediction of Grade Group (GG) upgrading in Gleason score (GS) ≤6 prostate cancer (PCa) (GG1) and GS 3 + 4 PCa (GG2).

Materials and methods

Fifty-nine patients who were biopsy-proven to have GG1 or GG2 and underwent MRI examination with the same MRI scanner prior to transrectal ultrasound (TRUS)-guided systemic biopsy were included. All these patients received radical prostatectomy to confirm the final GG. Patients were divided into training cohort and test cohort. 94 texture features were extracted from ADC maps for each patient. The independent sample t-test or Mann−Whitney U test was used to identify the texture features with statistically significant differences between GG upgrading group and GG non-upgrading group. Texture features of GG1 and GG2 were compared based on the final pathology of radical prostatectomy. We used the least absolute shrinkage and selection operator (LASSO) algorithm to filter features. Four supervised machine learning methods were employed. The prediction performance of each model was evaluated by area under the receiver operating characteristic curve (AUC). The statistical comparison between AUCs was performed.

Results

Six texture features were selected for the machine learning models building. These texture features were significantly different between GG upgrading group and GG non-upgrading group (P < 0.05). The six features had no significant difference between GG1 and GG2 based on the final pathology of radical prostatectomy. All machine learning methods had satisfactory predictive efficacy. The diagnostic performance of nearest neighbor algorithm (NNA) and support vector machine (SVM) was better than random forests (RF) in the training cohort. The AUC, sensitivity, and specificity of NNA were 0.872 (95% CI: 0.750−0.994), 0.967, and 0.778, respectively. The AUC, sensitivity, and specificity of SVM were 0.861 (95%CI: 0.732−0.991), 1.000, and 0.722, respectively. There had no significant difference between AUCs in the test cohort.

Conclusion

A combination of texture features and machine learning-based analysis of ADC maps could predict PCa GG upgrading from biopsy to radical prostatectomy non-invasively with satisfactory predictive efficacy.

A Systematic Review of Artificial Intelligence in Prostate Cancer

The diagnosis and management of prostate cancer involves the interpretation of data from multiple modalities to aid in decision making. Tools like PSA levels, MRI guided biopsies, genomic biomarkers, and Gleason grading are used to diagnose, risk stratify, and then monitor patients during respective follow-ups. Nevertheless, diagnosis tracking and subsequent risk stratification often lend itself to significant subjectivity. Artificial intelligence (AI) can allow clinicians to recognize difficult relationships and manage enormous data sets, which is a task that is both extraordinarily difficult and time consuming for humans. By using AI algorithms and reducing the level of subjectivity, it is possible to use fewer resources while improving the overall efficiency and accuracy in prostate cancer diagnosis and management. Thus, this systematic review focuses on analyzing advancements in AI-based artificial neural networks (ANN) and their current role in prostate cancer diagnosis and management.

Semi-automated PIRADS scoring via mpMRI analysis

Purpose: Prostate cancer (PCa) is the most common solid organ cancer and second leading cause of death in men. Multiparametric magnetic resonance imaging (mpMRI) enables detection of the most aggressive, clinically significant PCa (csPCa) tumors that require further treatment. A suspicious region of interest (ROI) detected on mpMRI is now assigned a Prostate Imaging-Reporting and Data System (PIRADS) score to standardize interpretation of mpMRI for PCa detection. However, there is significant inter-reader variability among radiologists in PIRADS score assignment and a minimal input semi-automated artificial intelligence (AI) system is proposed to harmonize PIRADS scores with mpMRI data. Approach: The proposed deep learning model (the seed point model) uses a simulated single-click seed point as input to annotate the lesion on mpMRI. This approach is in contrast to typical medical AI-based approaches that require annotation of the complete lesion. The mpMRI data from 617 patients used in this study were prospectively collected at a major tertiary U.S. medical center. The model was trained and validated to classify whether an mpMRI image had a lesion with a PIRADS score greater than or equal to PIRADS 4. Results: The model yielded an average receiver-operator characteristic (ROC) area under the curve (ROC-AUC) of 0.704 over a 10-fold cross-validation, which is significantly higher than the previously published benchmark. Conclusions: The proposed model could aid in PIRADS scoring of mpMRI, providing second reads to promote quality as well as offering expertise in environments that lack a radiologist with training in prostate mpMRI interpretation. The model could help identify tumors with a higher PIRADS for better clinical management and treatment of PCa patients at an early stage.

Improving prostate cancer classification in H&E tissue micro arrays using Ki67 and P63 histopathology

Histopathology of Hematoxylin and Eosin (H&E)-stained tissue obtained from biopsy is commonly used in prostate cancer (PCa) diagnosis. Automatic PCa classification of digitized H&E slides has been developed before, but no attempts have been made to classify PCa using additional tissue stains registered to H&E. In this paper, we demonstrate that using H&E, Ki67 and p63-stained (3-stain) tissue improves PCa classification relative to H&E alone. We also show that we can infer PCa-relevant Ki67 and p63 information from the H&E slides alone, and use it to achieve H&E-based PCa classification that is comparable to the 3-stain classification. Reported improvements apply to classifying benign vs. malignant tissue, and low grade (Gleason group 2) vs. high grade (Gleason groups 3,4,5) cancer. Specifically, we conducted four classification tasks using 333 tissue samples extracted from 231 radical prostatectomy patients: regression tree-based classification using either (i) 3-stain features, with a benign vs malignant area under the curve (AUC = 92.9%), or (ii) real H&E features and H&E features learned from Ki67 and p63 stains (AUC = 92.4%), as well as deep learning classification using either (iii) real 3-stain tissue patches (AUC = 94.3%) and (iv) real H&E patches and generated Ki67 and p63 patches (AUC = 93.0%) using a deep convolutional generative adversarial network. Classification performance was assessed with Monte Carlo cross validation and quantified in terms of the Area Under the Curve, Brier score, sensitivity, and specificity. Our results are interpretable and indicate that the standard H&E classification could be improved by mimicking other stain types.

Application of hierarchical clustering to multi-parametric MR in prostate: Differentiation of tumor and normal tissue with high accuracy

PURPOSE

Hierarchical clustering (HC), an unsupervised machine learning (ML) technique, was applied to multi-parametric MR (mp-MR) for prostate cancer (PCa). The aim of this study is to demonstrate HC can diagnose PCa in a straightforward interpretable way, in contrast to deep learning (DL) techniques.

METHODS

HC was constructed using mp-MR including intravoxel incoherent motion, diffusion kurtosis imaging, and dynamic contrast-enhanced MRI from 40 tumor and normal tissues in peripheral zone (PZ) and 23 tumor and normal tissues in transition zone (TZ). HC model was optimized by assessing the combinations of several dissimilarity and linkage methods. Goodness of HC model was validated by internal methods.

RESULTS

Accuracy for differentiating tumor and normal tissue by optimal HC model was 96.3% in PZ and 97.8% in TZ, comparable to current clinical standards. Relationship between input (DWI and permeability parameters) and output (tumor and normal tissue cluster) was shown by heat maps, consistent with literature.

CONCLUSION

HC can accurately differentiate PCa and normal tissue, comparable to state-of-the-art diffusion based parameters. Contrary to DL techniques, HC is an operator-independent ML technique producing results that can be interpreted such that the results can be knowledgeably judged.

Radiologist-like artificial intelligence for grade group prediction of radical prostatectomy for reducing upgrading and downgrading from biopsy

Rationale: To reduce upgrading and downgrading between needle biopsy (NB) and radical prostatectomy (RP) by predicting patient-level Gleason grade groups (GGs) of RP to avoid over- and under-treatment. Methods: In this study, we retrospectively enrolled 575 patients from two medical institutions. All patients received prebiopsy magnetic resonance (MR) examinations, and pathological evaluations of NB and RP were available. A total of 12,708 slices of original male pelvic MR images (T2-weighted sequences with fat suppression, T2WI-FS) containing 5405 slices of prostate tissue, and 2,753 tumor annotations (only T2WI-FS were annotated using RP pathological sections as ground truth) were analyzed for the prediction of patient-level RP GGs. We present a prostate cancer (PCa) framework, PCa-GGNet, that mimics radiologist behavior based on deep reinforcement learning (DRL). We developed and validated it using a multi-center format. Results: Accuracy (ACC) of our model outweighed NB results (0.815 [95% confidence interval (CI): 0.773-0.857] vs. 0.437 [95% CI: 0.335-0.539]). The PCa-GGNet scored higher (kappa value: 0.761) than NB (kappa value: 0.289). Our model significantly reduced the upgrading rate by 27.9% (P < 0.001) and downgrading rate by 6.4% (P = 0.029). Conclusions: DRL using MRI can be applied to the prediction of patient-level RP GGs to reduce upgrading and downgrading from biopsy, potentially improving the clinical benefits of prostate cancer oncologic controls.

Prediction of pituitary adenoma surgical consistency: radiomic data mining and machine learning on T2-weighted MRI

Purpose

Pituitary macroadenoma consistency can influence the ease of lesion removal during surgery, especially when using a transsphenoidal approach. Unfortunately, it is not assessable on standard qualitative MRI. Radiomic texture analysis could help in extracting mineable quantitative tissue characteristics. We aimed to assess the accuracy of texture analysis combined with machine learning in the preoperative evaluation of pituitary macroadenoma consistency in patients undergoing endoscopic endonasal surgery.

Methods

Data of 89 patients (68 soft and 21 fibrous macroadenomas) who underwent MRI and transsphenoidal surgery at our institution were retrospectively reviewed. After manual segmentation, radiomic texture features were extracted from original and filtered MR images. Feature stability analysis and a multistep feature selection were performed. After oversampling to balance the classes, 80% of the data was used for hyperparameter tuning via stratified 5-fold cross-validation, while a 20% hold-out set was employed for its final testing, using an Extra Trees ensemble meta-algorithm. The reference standard was based on surgical findings.

Results

A total of 1118 texture features were extracted, of which 741 were stable. After removal of low variance (n = 4) and highly intercorrelated (n = 625) parameters, recursive feature elimination identified a subset of 14 features. After hyperparameter tuning, the Extra Trees classifier obtained an accuracy of 93%, sensitivity of 100%, and specificity of 87%. The area under the receiver operating characteristic and precision-recall curves was 0.99.

Conclusion

Preoperative T2-weighted MRI texture analysis and machine learning could predict pituitary macroadenoma consistency.

Electronic supplementary material

The online version of this article (10.1007/s00234-020-02502-z) contains supplementary material, which is available to authorized users.

Evaluation of Multiple Prognostic Factors of Hepatocellular Carcinoma with Intra-Voxel Incoherent Motions Imaging by Extracting the Histogram Metrics

Purpose

To predict multiple prognostic factors of HCC including histopathologic grade, the expression of Ki67 as well as capsule formation with intravoxel incoherent motions imaging by extracting the histogram metrics.

Patients and Methods

A total of 52 patients with HCC were recruited with the MR examinations undertaken at a 3T scanner. Histogram metrics were extracted from IVIM-derived parametric maps. Independent student t-test was performed to explore the differences in metrics across different subtypes of prognostic factors. Spearman correlation test was utilized to evaluate the correlations between the IVIM metrics and prognostic factors. ROC analysis was applied to evaluate the diagnostic performance.

Results

According to the independent student t-test, there were 18, 4, and 8 IVIM-derived histogram metrics showing the capability for differentiating the subtypes of histopathologic grade, Ki67, and capsule formation, respectively, with P-values of less than 0.05. Besides, there existed a lot of significant correlations between IVIM metrics and prognostic factors. Finally, by integrating different histogram metrics showing significant differences between various subgroups together via establishing logistic regression based diagnostic models, greatest diagnostic power was obtained for grading HCC (AUC=0.917), diagnosing patients with highly expressed Ki67 (AUC=0.861) and diagnosing patients with capsule formation (AUC=0.839).

Conclusion

Multiple prognostic factors including histopathologic grade, Ki67 expression status, and capsule formation can be accurately predicted with assistance of histogram metrics sourced from a single IVIM scan.

Multiparametric MRI for Prostate Cancer Characterization: Combined Use of Radiomics Model with PI-RADS and Clinical Parameters

Radiomics is an emerging field of image analysis with potential applications in patient risk stratification. This study developed and evaluated machine learning models using quantitative radiomic features extracted from multiparametric magnetic resonance imaging (mpMRI) to detect and classify prostate cancer (PCa). In total, 191 patients that underwent prostatic mpMRI and combined targeted and systematic fusion biopsy were retrospectively included. Segmentations of the whole prostate glands and index lesions were performed manually in apparent diffusion coefficient (ADC) maps and T2-weighted MRI. Radiomic features were extracted from regions corresponding to the whole prostate gland and index lesion. The best performing combination of feature setup and classifier was selected to compare its predictive ability of the radiologist’s evaluation (PI-RADS), mean ADC, prostate specific antigen density (PSAD) and digital rectal examination (DRE) using receiver operating characteristic (ROC) analysis. Models were evaluated using repeated 5-fold cross-validation and a separate independent test cohort. In the test cohort, an ensemble model combining a radiomics model, with models for PI-RADS, PSAD and DRE achieved high predictive AUCs for the differentiation of (i) malignant from benign prostatic lesions (AUC = 0.889) and of (ii) clinically significant (csPCa) from clinically insignificant PCa (cisPCa) (AUC = 0.844). Our combined model was numerically superior to PI-RADS for cancer detection (AUC = 0.779; p = 0.054) as well as for clinical significance prediction (AUC = 0.688; p = 0.209) and showed a significantly better performance compared to mADC for csPCa prediction (AUC = 0.571; p = 0.022). In our study, radiomics accurately characterizes prostatic index lesions and shows performance comparable to radiologists for PCa characterization. Quantitative image data represent a potential biomarker, which, when combined with PI-RADS, PSAD and DRE, predicts csPCa more accurately than mADC. Prognostic machine learning models could assist in csPCa detection and patient selection for MRI-guided biopsy.

Machine learning for the identification of clinically significant prostate cancer on MRI: a meta-analysis

OBJECTIVES

The aim of this study was to systematically review the literature and perform a meta-analysis of machine learning (ML) diagnostic accuracy studies focused on clinically significant prostate cancer (csPCa) identification on MRI.

METHODS

Multiple medical databases were systematically searched for studies on ML applications in csPCa identification up to July 31, 2019. Two reviewers screened all papers independently for eligibility. The area under the receiver operating characteristic curves (AUC) was pooled to quantify predictive accuracy. A random-effects model estimated overall effect size while statistical heterogeneity was assessed with the I² value. A funnel plot was used to investigate publication bias. Subgroup analyses were performed based on reference standard (biopsy or radical prostatectomy) and ML type (deep and non-deep).

RESULTS

After the final revision, 12 studies were included in the analysis. Statistical heterogeneity was high both in overall and in subgroup analyses. The overall pooled AUC for ML in csPCa identification was 0.86, with 0.81-0.91 95% confidence intervals (95%CI). The biopsy subgroup (n = 9) had a pooled AUC of 0.85 (95%CI = 0.79-0.91) while the radical prostatectomy one (n = 3) of 0.88 (95%CI = 0.76-0.99). Deep learning ML (n = 4) had a 0.78 AUC (95%CI = 0.69-0.86) while the remaining 8 had AUC = 0.90 (95%CI = 0.85-0.94).

CONCLUSIONS

ML pipelines using prostate MRI to identify csPCa showed good accuracy and should be further investigated, possibly with better standardisation in design and reporting of results.

KEY POINTS

• Overall pooled AUC was 0.86 with 0.81-0.91 95% confidence intervals. • In the reference standard subgroup analysis, algorithm accuracy was similar with pooled AUCs of 0.85 (0.79-0.91 95% confidence intervals) and 0.88 (0.76-0.99 95% confidence intervals) for studies employing biopsies and radical prostatectomy, respectively. • Deep learning pipelines performed worse (AUC = 0.78, 0.69-0.86 95% confidence intervals) than other approaches (AUC = 0.90, 0.85-0.94 95% confidence intervals).

Automated Classification of Significant Prostate Cancer on MRI: A Systematic Review on the Performance of Machine Learning Applications

Significant prostate carcinoma (sPCa) classification based on MRI using radiomics or deep learning approaches has gained much interest, due to the potential application in assisting in clinical decision-making.

OBJECTIVE

To systematically review the literature (i) to determine which algorithms are most frequently used for sPCa classification, (ii) to investigate whether there exists a relation between the performance and the method or the MRI sequences used, (iii) to assess what study design factors affect the performance on sPCa classification, and (iv) to research whether performance had been evaluated in a clinical setting Methods: The databases Embase and Ovid MEDLINE were searched for studies describing machine learning or deep learning classification methods discriminating between significant and nonsignificant PCa on multiparametric MRI that performed a valid validation procedure. Quality was assessed by the modified radiomics quality score. We computed the median area under the receiver operating curve (AUC) from overall methods and the interquartile range.

RESULTS

From 2846 potentially relevant publications, 27 were included. The most frequent algorithms used in the literature for PCa classification are logistic regression (22%) and convolutional neural networks (CNNs) (22%). The median AUC was 0.79 (interquartile range: 0.77-0.87). No significant effect of number of included patients, image sequences, or reference standard on the reported performance was found. Three studies described an external validation and none of the papers described a validation in a prospective clinical trial.

CONCLUSIONS

To unlock the promising potential of machine and deep learning approaches, validation studies and clinical prospective studies should be performed with an established protocol to assess the added value in decision-making.

Multiparametric MRI-Based Radiomics Nomogram for Predicting Lymphovascular Space Invasion in Endometrial Carcinoma

BACKGROUND

Lymphovascular space invasion (LVSI) of endometrial carcinoma (EMC) is one of the important prognostic factors, which is not usually visible subjectively. Therefore, clinicians lack imaging-based evidence about LVSI for preoperative treatment decision-making.

PURPOSE

To develop a multiparametric MRI (mpMRI)-based radiomics nomogram for predicting LVSI in EMC and provide decision-making support to clinicians.

STUDY TYPE

Retrospective.

POPULATION

In all, 144 patients with histologically confirmed EMC, 101 patients in a training cohort, and 43 patients in a test cohort.

FIELD STRENGTH/SEQUENCE

T₂ WI, contrast enhanced-T₁ WI, and diffusion-weighted imaging (DWI) at 3.0T MRI.

ASSESSMENT

Tumors were independently segmented images by two radiologists. Two pathologists reviewed the tissue specimens of the tumors to identify the existence of LVSI in consensus.

STATISTICAL TESTS

The intraclass correlation coefficient was used to test the reliability and least absolute shrinkage and selection operator (LASSO) regression for features selection and then developed a radiomics signature named Rad-score. A nomogram was developed in the training cohort. The diagnostic performance of the nomogram model was assessed by area under the curve (AUC) of the receiver operator characteristic (ROC) in the training and test cohort, respectively.

RESULTS

LVSI was identified in 32 patients (22.2%). Older age and high grade were correlated with LVSI at univariate analysis. There were five radiomics features that were identified as independent risk factors for LVSI by LASSO regression. Based on age, grade, and Rad-score, the AUC values of the nomogram model to predict LVSI in the training and test cohort were 0.820 (95% confidence interval [CI]: 0.725, 0.916; sensitivity: 82.6%, specificity: 72.9%), 0.807 (95% CI: 0.673, 0.941; sensitivity: 77.8%, specificity: 78.6%), respectively.

DATA CONCLUSION

The radiomic-based machine-learning model using a nomogram algorithm achieved high diagnostic performance for predicting LVSI of EMC preoperatively, which could enhance risk stratification and provide support for therapeutic decision-making.

LEVEL OF EVIDENCE

2.

TECHNICAL EFFICACY STAGE

3. J. Magn. Reson. Imaging 2020;52:1257-1262.

Predicting clinically significant prostate cancer from quantitative image features including compressed sensing radial MRI of prostate perfusion using machine learning: comparison with PI-RADS v2 assessment scores

Background

To investigate if supervised machine learning (ML) classifiers would be able to predict clinically significant cancer (sPC) from a set of quantitative image-features and to compare these results with established PI-RADS v2 assessment scores.

Methods

We retrospectively included 201, histopathologically-proven, peripheral zone (PZ) prostate cancer lesions. Gleason scores ≤3+3 were considered as clinically insignificant (inPC) and Gleason scores ≥3+4 as sPC and were encoded in a binary fashion, serving as ground-truth. MRI was performed at 3T with high spatiotemporal resolution DCE using Golden-angle RAdial SParse (GRASP) MRI. Perfusion maps (Ktrans, Kep, Ve), apparent diffusion coefficient (ADC), and absolute T2-signal intensities (SI) were determined in all lesions and served as input parameters for four supervised ML models: Gradient Boosting Machines (GBM), Neural Networks (NNet), Random Forest (RF) and Support Vector Machines (SVM). ML results and PI-RADS scores were compared with the ground-truth. Next ROC-curves and AUC values were calculated.

Results

All ML models outperformed PI-RADS v2 assessment scores in the prediction of sPC (RF, GBM, NNet and SVM vs. PI-RADS: AUC 0.899, 0.864, 0.884 and 0.874 vs. 0.595, all P<0.001).

Conclusions

Using quantitative imaging parameters as input, supervised ML models outperformed PI-RADS v2 assessment scores in the prediction of sPC. These results indicate that quantitative imagining parameters contain relevant information for the prediction of sPC from image features.

[Artificial intelligence and radiomics in MRI-based prostate diagnostics]

CLINICAL/METHODICAL ISSUE

In view of the diagnostic complexity and the large number of examinations, modern radiology is challenged to identify clinically significant prostate cancer (PCa) with high sensitivity and specificity. Meanwhile overdiagnosis and overtreatment of clinically nonsignificant carcinomas need to be avoided.

STANDARD RADIOLOGICAL METHODS

Increasingly, international guidelines recommend multiparametric magnetic resonance imaging (mpMRI) as first-line investigation in patients with suspected PCa.

METHODICAL INNOVATIONS

Image interpretation according to the PI-RADS criteria is limited by interobserver variability. Thus, rapid developments in the field of automated image analysis tools, including radiomics and artificial intelligence (AI; machine learning, deep learning), give hope for further improvement in patient care.

PERFORMANCE

AI focuses on the automated detection and classification of PCa, but it also attempts to stratify tumor aggressiveness according to the Gleason score. Recent studies present good to very good results in radiomics or AI-supported mpMRI diagnosis. Nevertheless, these systems are not widely used in clinical practice.

ACHIEVEMENTS AND PRACTICAL RECOMMENDATIONS

In order to apply these innovative technologies, a growing awareness for the need of structured data acquisition, development of robust systems and an increased acceptance of AI as diagnostic support are needed. If AI overcomes these obstacles, it may play a key role in the quantitative and reproducible image-based diagnosis of ever-increasing prostate MRI examination volumes.