Radiomics workflow definition & challenges - German priority program 2177 consensus statement on clinically applied radiomics.

OBJECTIVES: Achieving a consensus on a definition for different aspects of radiomics workflows to support their translation into clinical usage. Furthermore, to assess the perspective of experts on important challenges for a successful clinical workflow implementation. MATERIALS AND METHODS: The consensus was achieved by a multi-stage process. Stage 1 comprised a definition screening, a retrospective analysis with semantic mapping of terms found in 22 workflow definitions, and the compilation of an initial baseline definition. Stages 2 and 3 consisted of a Delphi process with over 45 experts hailing from sites participating in the German Research Foundation (DFG) Priority Program 2177. Stage 2 aimed to achieve a broad consensus for a definition proposal, while stage 3 identified the importance of translational challenges. RESULTS: Workflow definitions from 22 publications (published 2012-2020) were analyzed. Sixty-nine definition terms were extracted, mapped, and semantic ambiguities (e.g., homonymous and synonymous terms) were identified and resolved. The consensus definition was developed via a Delphi process. The final definition comprising seven phases and 37 aspects reached a high overall consensus (> 89% of experts "agree" or "strongly agree"). Two aspects reached no strong consensus. In addition, the Delphi process identified and characterized from the participating experts' perspective the ten most important challenges in radiomics workflows. CONCLUSION: To overcome semantic inconsistencies between existing definitions and offer a well-defined, broad, referenceable terminology, a consensus workflow definition for radiomics-based setups and a terms mapping to existing literature was compiled. Moreover, the most relevant challenges towards clinical application were characterized. CRITICAL RELEVANCE STATEMENT: Lack of standardization represents one major obstacle to successful clinical translation of radiomics. Here, we report a consensus workflow definition on different aspects of radiomics studies and highlight important challenges to advance the clinical adoption of radiomics. KEY POINTS: Published radiomics workflow terminologies are inconsistent, hindering standardization and translation. A consensus radiomics workflow definition proposal with high agreement was developed. Publicly available result resources for further exploitation by the scientific community.

A Machine Learning Framework Reduces the Manual Workload for Systematic Reviews of the Diagnostic Performance of Prostate Magnetic Resonance Imaging.

Prostate magnetic resonance imaging has become the imaging standard for prostate cancer in various clinical settings, with interpretation standardized according to the Prostate Imaging Reporting and Data System (PI-RADS). Each year, hundreds of scientific studies that report on the diagnostic performance of PI-RADS are published. To keep up with this ever-increasing evidence base, systematic reviews and meta-analyses are essential. As systematic reviews are highly resource-intensive, we investigated whether a machine learning framework can reduce the manual workload and speed up the screening process (title and abstract). We used search results from a living systematic review of the diagnostic performance of PI-RADS (1585 studies, of which 482 were potentially eligible after screening). A naïve Bayesian classifier was implemented in an active learning environment for classification of the titles and abstracts. Our outcome variable was the percentage of studies that can be excluded after 95% of relevant studies have been identified by the classifier (work saved over sampling: WSS@95%). In simulation runs of the entire screening process (controlling for classifier initiation and the frequency of classifier updating), we obtained a WSS@95% value of 28% (standard error of the mean ±0.1%). Applied prospectively, our classification framework would translate into a significant reduction in manual screening effort. Patient summary: Systematic reviews of scientific evidence are labor-intensive and take a lot of time. For example, many studies on prostate cancer diagnosis via MRI (magnetic resonance imaging) are published every year. We describe the use of machine learning to reduce the manual workload in screening search results. For a review of MRI for prostate cancer diagnosis, this approach reduced the screening workload by about 28%.

Prediction of upgrade to clinically significant prostate cancer in patients under active surveillance: Performance of a fully automated AI-algorithm for lesion detection and classification.

BACKGROUND: Multiparametric MRI (mpMRI) improves the detection of aggressive prostate cancer (PCa) subtypes. As cases of active surveillance (AS) increase and tumor progression triggers definitive treatment, we evaluated whether an AI-driven algorithm can detect clinically significant PCa (csPCa) in patients under AS. METHODS: Consecutive patients under AS who received mpMRI (PI-RADSv2.1 protocol) and subsequent MR-guided ultrasound fusion (targeted and extensive systematic) biopsy between 2017 and 2020 were retrospectively analyzed. Diagnostic performance of an automated clinically certified AI-driven algorithm was evaluated on both lesion and patient level regarding the detection of csPCa. RESULTS: Analysis of 56 patients resulted in 93 target lesions. Patient level sensitivity and specificity of the AI algorithm was 92.5%/31% for the detection of ISUP ≥ 1 and 96.4%/25% for the detection of ISUP ≥ 2, respectively. The only case of csPCa missed by the AI harbored only 1/47 Gleason 7a core (systematic biopsy; previous and subsequent biopsies rendered non-csPCa). CONCLUSIONS: AI-augmented lesion detection and PI-RADS scoring is a robust tool to detect progression to csPCa in patients under AS. Integration in the clinical workflow can serve as reassurance for the reader and streamline reporting, hence improve efficiency and diagnostic confidence.

Living systematic review and meta-analysis of the prostate MRI diagnostic test with Prostate Imaging Reporting and Data System (PI-RADS) assessment for the detection of prostate cancer: study protocol.

INTRODUCTION: The Prostate Imaging Reporting and Data System (PI-RADS) standardises reporting of prostate MRI for the detection of clinically significant prostate cancer. We provide the protocol of a planned living systematic review and meta-analysis for (1) diagnostic accuracy (sensitivity and specificity), (2) cancer detection rates of assessment categories and (3) inter-reader agreement. METHODS AND ANALYSIS: Retrospective and prospective studies reporting on at least one of the outcomes of interest are included. Each step that requires literature evaluation and data extraction is performed by two independent reviewers. Since PI-RADS is intended as a living document itself, a 12-month update cycle of the systematic review and meta-analysis is planned.This protocol is in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses-Protocols statement. The search strategies including databases, study eligibility criteria, index and reference test definitions, outcome definitions and data analysis processes are detailed. A full list of extracted data items is provided.Summary estimates of sensitivity and specificity (for PI-RADS ≥3 and PI-RADS ≥4 considered positive) are derived with bivariate binomial models. Summary estimates of cancer detection rates are calculated with random intercept logistic regression models for single proportions. Summary estimates of inter-reader agreement are derived with random effects models. ETHICS AND DISSEMINATION: No original patient data are collected, ethical review board approval, therefore, is not necessary. Results are published in peer-reviewed, open-access scientific journals. We make the collected data accessible as supplemental material to guarantee transparency of results. PROSPERO REGISTRATION NUMBER: CRD42022343931.

Quantitative Analysis of Diffusion Weighted Imaging May Improve Risk Stratification of Prostatic Transition Zone Lesions.

BACKGROUND/AIM: To investigate whether quantitative analysis of diffusion weighted images allows for improved risk stratification of transition zone lesions in prostate magnetic resonance imaging (MRI) evaluated according to PI-RADSv2.1 [Prostate Imaging Reporting and Data System, target variable: clinically significant prostate cancer (csPCa)]. PATIENTS AND METHODS: Consecutive patients with transition zone lesions in 3T prostate MRI were enrolled in the study. All lesions on MRI were histopathologically verified by transperineal MRI-TRUS fusion biopsy. Two blinded radiologists re-evaluated all lesions according to PI-RADSv2.1. A consensus reading was performed after reading of all cases. Additionally, mean apparent diffusion coefficient values (mADC) were derived from blinded lesion segmentation. ROC analysis was performed for PI-RADS categories and PI-RADS categories with separate subcategories and diffusion coefficient values (ADC). Data were examined for optimal mADC cut-off values that improve stratification of csPCa and benign lesions. RESULTS: Among 85 patients (mean age=66.2 years), 98 transition zone lesions were detected. Biopsy confirmed csPCa in 24/98 cases. Area under the curve (AUC) was 0.89/0.90 for reader 1, 0.92/0.91 for reader 2 and 0.92/0.91 for the consensus reading (5 category analysis/analysis with subcategories separately). Inter-reader agreement was substantial, with lower PI-RADS categories assigned by the more experienced reader (p<0.05). AUC for mADC alone was 0.81. When a cut-off threshold of 950 µm2/s mADC is used to downgrade PI-RADS 3 lesions to PI-RADS 2, biopsy could be avoided in all benign PI-RADS 3 cases. CONCLUSION: Quantitative analysis of diffusion weighted images may help avoid unnecessary biopsies of transition zone PI-RADS 3 lesions.

Explainable AI for CNN-based prostate tumor segmentation in multi-parametric MRI correlated to whole mount histopathology.

Automatic prostate tumor segmentation is often unable to identify the lesion even if multi-parametric MRI data is used as input, and the segmentation output is difficult to verify due to the lack of clinically established ground truth images. In this work we use an explainable deep learning model to interpret the predictions of a convolutional neural network (CNN) for prostate tumor segmentation. The CNN uses a U-Net architecture which was trained on multi-parametric MRI data from 122 patients to automatically segment the prostate gland and prostate tumor lesions. In addition, co-registered ground truth data from whole mount histopathology images were available in 15 patients that were used as a test set during CNN testing. To be able to interpret the segmentation results of the CNN, heat maps were generated using the Gradient Weighted Class Activation Map (Grad-CAM) method. The CNN achieved a mean Dice Sorensen Coefficient 0.62 and 0.31 for the prostate gland and the tumor lesions -with the radiologist drawn ground truth and 0.32 with whole-mount histology ground truth for tumor lesions. Dice Sorensen Coefficient between CNN predictions and manual segmentations from MRI and histology data were not significantly different. In the prostate the Grad-CAM heat maps could differentiate between tumor and healthy prostate tissue, which indicates that the image information in the tumor was essential for the CNN segmentation.

More Than Detection of Adenocarcinoma - Indications and Findings in Prostate MRI in Benign Prostatic Disorders.

BACKGROUND: Multiparametric MRI of the prostate has become a fundamental tool in the diagnostic pathway for prostate cancer and is recommended before (or after negative) biopsy to guide biopsy and increase accuracy, as a staging examination (high-risk setting), and prior to inclusion into active surveillance. Despite this main field of application, prostate MRI can be utilized to obtain information in a variety of benign disorders of the prostate. METHODS: Systematic bibliographical research with extraction of studies, national (German) as well as international guidelines (EAU, AUA), and consensus reports on MRI of benign disorders of the prostate was performed. Indications and imaging findings of prostate MRI were identified for a) imaging the enlarged prostate, b) prostate MRI in prostatic artery embolization, c) imaging in prostatitis and d) imaging in congenital anomalies. RESULTS AND CONCLUSIONS: Different phenotypes of the enlarged prostate that partly correlate with severity of symptoms are discussed. We provide an overview of the different types of prostatitis and possible imaging findings, highlighting abscesses as a severe complication. The most common congenital anomalies of the prostate are utricular cysts, whereas anomalies like aplasia, hypoplasia, and ectopia are rare disorders. Knowledge of indications for imaging and imaging appearance of these conditions may improve patient care and enhance differential diagnosis. KEY POINTS: · Current guidelines do not implement indications for mpMRI apart from prostate carcinoma.. · MRI can distinguish different anatomical phenotypes of prostatic enlargement.. · Prostatic artery embolization represents a valuable treatment option in cases of symptomatic benign prostatic enlargement.. · Different forms of prostatitis exist and may mimic prostate carcinoma in MRI.. · MRI can be used to evaluate anatomical prostate anomalies.. CITATION FORMAT: · Oerther B, Sigle A, Franiel T et al. More Than Detection of Adenocarcinoma - Indications and Findings in Prostate MRI in Benign Prostatic Disorders. Fortschr Röntgenstr 2022; 194: 481 - 490.

Cancer detection rates of the PI-RADSv2.1 assessment categories: systematic review and meta-analysis on lesion level and patient level.

BACKGROUND: The Prostate Imaging Reporting and Data System, version 2.1 (PI-RADSv2.1) standardizes reporting of multiparametric MRI of the prostate. Assigned assessment categories are a risk stratification algorithm, higher categories indicate a higher probability of clinically significant cancer compared to lower categories. PI-RADSv2.1 does not define these probabilities numerically. We conduct a systematic review and meta-analysis to determine the cancer detection rates (CDR) of the PI-RADSv2.1 assessment categories on lesion level and patient level. METHODS: Two independent reviewers screen a systematic PubMed and Cochrane CENTRAL search for relevant articles (primary outcome: clinically significant cancer, index test: prostate MRI reading according to PI-RADSv2.1, reference standard: histopathology). We perform meta-analyses of proportions with random-effects models for the CDR of the PI-RADSv2.1 assessment categories for clinically significant cancer. We perform subgroup analysis according to lesion localization to test for differences of CDR between peripheral zone lesions and transition zone lesions. RESULTS: A total of 17 articles meet the inclusion criteria and data is independently extracted by two reviewers. Lesion level analysis includes 1946 lesions, patient level analysis includes 1268 patients. On lesion level analysis, CDR are 2% (95% confidence interval: 0-8%) for PI-RADS 1, 4% (1-9%) for PI-RADS 2, 20% (13-27%) for PI-RADS 3, 52% (43-61%) for PI-RADS 4, 89% (76-97%) for PI-RADS 5. On patient level analysis, CDR are 6% (0-20%) for PI-RADS 1, 9% (5-13%) for PI-RADS 2, 16% (7-27%) for PI-RADS 3, 59% (39-78%) for PI-RADS 4, 85% (73-94%) for PI-RADS 5. Higher categories are significantly associated with higher CDR (P < 0.001, univariate meta-regression), no systematic difference of CDR between peripheral zone lesions and transition zone lesions is identified in subgroup analysis. CONCLUSIONS: Our estimates of CDR demonstrate that PI-RADSv2.1 stratifies lesions and patients as intended. Our results might serve as an initial evidence base to discuss management strategies linked to assessment categories.

Isotropic Expansion of the Intraprostatic Gross Tumor Volume of Primary Prostate Cancer Patients Defined in MRI-A Correlation Study With Whole Mount Histopathological Information as Reference.

INTRODUCTION: An accurate delineation of the intraprostatic gross tumor volume (GTV) is of importance for focal treatment in patients with primary prostate cancer (PCa). Multiparametric MRI (mpMRI) is the standard of care for lesion detection but has been shown to underestimate GTV. This study investigated how far the GTV has to be expanded in MRI in order to reach concordance with the histopathological reference and whether this strategy is practicable in clinical routine. PATIENTS AND METHODS: Twenty-two patients with planned prostatectomy and preceded 3 Tesla mpMRI were prospectively examined. After surgery, PCa contours delineated on histopathological slides (GTV-Histo) were superimposed on MRI using ex-vivo imaging as support for co-registration. According to the PI-RADSv2 classification, GTV was manually delineated in MRI (GTV-MRI) by two experts in consensus. For volumetric analysis, we compared GTV-MRI and GTV-Histo. Subsequently, we isotropically enlarged GTV-MRI in 1 mm increments within the prostate and also compared those with GTV-Histo regarding the absolute volumes. For evaluating the spatial accuracy, we considered the coverage ratio of GTV-Histo, the Sørensen-Dice coefficient (DSC), as well as the contact with the urethra. RESULTS: In 19 of 22 patients MRI underestimated the intraprostatic tumor volume compared to histopathological reference: median GTV-Histo (4.7 cm3, IQR: 2.5-18.8) was significantly (p<0.001) lager than median GTV-MRI (2.6 cm3, IQR: 1.2-6.9). A median expansion of 1 mm (range: 0-4 mm) adjusted the initial GTV-MRI to at least the volume of GTV-Histo (GTVexp-MRI). Original GTV-MRI and expansion with 1, 2, 3, and 4 mm covered in median 39% (IQR: 2%-78%), 62% (10%-91%), 70% (15%-95%), 80% (21-100), 87% (25%-100%) of GTV-Histo, respectively. Best DSC (median: 0.54) between GTV-Histo and GTV-MRI was achieved by median expansion of 2 mm. The urethra was covered by initial GTVs-MRI in eight patients (36%). After applying an expansion with 2 mm the urethra was covered in one more patient by GTV-MRI. CONCLUSION: Using histopathology as reference, we demonstrated that MRI underestimates intraprostatic tumor volume. A 2 mm-expansion may improve accurate GTV-delineation while respecting the balance between histological tumor coverage and overtreatment.

Predicting Biochemical Failure in Irradiated Patients With Prostate Cancer by Tumour Volume Measured by Multiparametric MRI.

BACKGROUND/AIM: We examined the prognostic value of intraprostatic gross tumour volume (GTV) as measured by multiparametric MRI (mpMRI) in patients with prostate cancer following (primary) external beam radiation therapy (EBRT). PATIENTS AND METHODS: In a retrospective monocentric study, we analysed patients with prostate cancer (PCa) after EBRT. GTV was delineated in pre-treatment mpMRI (GTV-MRI) using T2-weighted images. Cox-regression analyses were performed considering biochemical failure recurrence-free survival (BRFS) as outcome variable. RESULTS: Among 131 patients, after a median follow-up of 57 months, biochemical failure occurred in 27 (21%). GTV-MRI was not correlated with % of positive biopsy cores, Gleason score and initial PSA (all r<0.2) and only moderately correlated with cT stage (r=0.32). In univariate analysis, cT stage, Gleason score and GTV-MRI were higher in subjects with shorter BRFS (p<0.05). GTV-MRI remained a significant predictor for BRFS in multivariate analyses, independent of Gleason score and cT stage. CONCLUSION: GTV, defined using mpMRI, provides incremental prognostic value for BRFS, independent of established risk factors. This supports the implementation of imaging-based GTV for risk-stratification, although further validation is needed.

Immunohistochemistry and Radiomic Features for Survival Prediction in Small Cell Lung Cancer.

Background: The aim of the study was to evaluate the role of different immunohistochemical and radiomics features in patients with small cell lung cancer (SCLC). Methods: Consecutive patients with histologically proven SCLC with limited (n = 47, 48%) or extensive disease (n = 51, 52%) treated with radiotherapy and chemotherapy at our department were included in the analysis. The expression of different immunohistochemical markers from the initial tissue biopsy, such as CD56, CD44, chromogranin A, synaptophysin, TTF-1, GLUT-1, Hif-1 a, PD-1, and PD-L1, and MIB-1/KI-67 as well as LDH und NSE from the initial blood sample were evaluated. H-scores were additionally generated for CD44, Hif-1a, and GLUT-1. A total of 72 computer tomography (CT) radiomics texture features from a homogenous subgroup (n = 31) of patients were correlated with the immunohistochemistry, the survival (OS), and the progression-free survival (PFS). Results: The median OS, calculated from diagnosis, was 21 months for patients with limited disease and 13 months for patients with extensive disease. The expression of synaptophysin correlated with a better OS (HR 0.546 95% CI 0.308-0.966, p = 0.03). The expression of TTF-1 (HR 0.286, 95% CI: 0.117-0.698, p = 0.006) and a lower GLUT-1 H-score (median = 50, HR: 0.511, 95% CI: 0.260-1.003, p = 0.05) correlated with a better PFS. Patients without chromogranin A expression had a higher risk for developing cerebral metastases (p = 0.02) and patients with PD 1 expression were at risk for developing metastases (p = 0.02). Our radiomics analysis did not reveal a single texture feature that correlated highly with OS or PFS. Correlation coefficients ranged between -0.48 and 0.39 for OS and between -0.46 and 0.38 for PFS. Conclusions: The role of synaptophysin should be further evaluated as synaptophysin-negative patients might profit from treatment intensification. We report an, at most, moderate correlation of radiomics features with overall and progression free survival and no correlation with the expression of different immunohistochemical markers.