The Image Biomarker Standardization Initiative: Standardized Convolutional Filters for Reproducible Radiomics and Enhanced Clinical Insights.

Filters are commonly used to enhance specific structures and patterns in images, such as vessels or peritumoral regions, to enable clinical insights beyond the visible image using radiomics. However, their lack of standardization restricts reproducibility and clinical translation of radiomics decision support tools. In this special report, teams of researchers who developed radiomics software participated in a three-phase study (September 2020 to December 2022) to establish a standardized set of filters. The first two phases focused on finding reference filtered images and reference feature values for commonly used convolutional filters: mean, Laplacian of Gaussian, Laws and Gabor kernels, separable and nonseparable wavelets (including decomposed forms), and Riesz transformations. In the first phase, 15 teams used digital phantoms to establish 33 reference filtered images of 36 filter configurations. In phase 2, 11 teams used a chest CT image to derive reference values for 323 of 396 features computed from filtered images using 22 filter and image processing configurations. Reference filtered images and feature values for Riesz transformations were not established. Reproducibility of standardized convolutional filters was validated on a public data set of multimodal imaging (CT, fluorodeoxyglucose PET, and T1-weighted MRI) in 51 patients with soft-tissue sarcoma. At validation, reproducibility of 486 features computed from filtered images using nine configurations × three imaging modalities was assessed using the lower bounds of 95% CIs of intraclass correlation coefficients. Out of 486 features, 458 were found to be reproducible across nine teams with lower bounds of 95% CIs of intraclass correlation coefficients greater than 0.75. In conclusion, eight filter types were standardized with reference filtered images and reference feature values for verifying and calibrating radiomics software packages. A web-based tool is available for compliance checking.

Outcomes and potential impact of a virtual hands-on training program on MRI staging confidence and performance in rectal cancer.

OBJECTIVES: To explore the potential impact of a dedicated virtual training course on MRI staging confidence and performance in rectal cancer. METHODS: Forty-two radiologists completed a stepwise virtual training course on rectal cancer MRI staging composed of a pre-course (baseline) test with 7 test cases (5 staging, 2 restaging), a 1-day online workshop, 1 month of individual case readings (n = 70 cases with online feedback), a live online feedback session supervised by two expert faculty members, and a post-course test. The ESGAR structured reporting templates for (re)staging were used throughout the course. Results of the pre-course and post-course test were compared in terms of group interobserver agreement (Krippendorf's alpha), staging confidence (perceived staging difficulty), and diagnostic accuracy (using an expert reference standard). RESULTS: Though results were largely not statistically significant, the majority of staging variables showed a mild increase in diagnostic accuracy after the course, ranging between + 2% and + 17%. A similar trend was observed for IOA which improved for nearly all variables when comparing the pre- and post-course. There was a significant decrease in the perceived difficulty level (p = 0.03), indicating an improved diagnostic confidence after completion of the course. CONCLUSIONS: Though exploratory in nature, our study results suggest that use of a dedicated virtual training course and web platform has potential to enhance staging performance, confidence, and interobserver agreement to assess rectal cancer on MRI virtual training and could thus be a good alternative (or addition) to in-person training. CLINICAL RELEVANCE STATEMENT: Rectal cancer MRI reporting quality is highly dependent on radiologists' expertise, stressing the need for dedicated training/teaching. This study shows promising results for a virtual web-based training program, which could be a good alternative (or addition) to in-person training. KEY POINTS: • Rectal cancer MRI reporting quality is highly dependent on radiologists' expertise, stressing the need for dedicated training and teaching. • Using a dedicated virtual training course and web-based platform, encouraging first results were achieved to improve staging accuracy, diagnostic confidence, and interobserver agreement. • These exploratory results suggest that virtual training could thus be a good alternative (or addition) to in-person training.

External validation of an MR-based radiomic model predictive of locoregional control in oropharyngeal cancer.

OBJECTIVES: To externally validate a pre-treatment MR-based radiomics model predictive of locoregional control in oropharyngeal squamous cell carcinoma (OPSCC) and to assess the impact of differences between datasets on the predictive performance. METHODS: Radiomic features, as defined in our previously published radiomics model, were extracted from the primary tumor volumes of 157 OPSCC patients in a different institute. The developed radiomics model was validated using this cohort. Additionally, parameters influencing performance, such as patient subgroups, MRI acquisition, and post-processing steps on prediction performance will be investigated. For this analysis, matched subgroups (based on human papillomavirus (HPV) status of the tumor, T-stage, and tumor subsite) and a subgroup with only patients with 4-mm slice thickness were studied. Also the influence of harmonization techniques (ComBat harmonization, quantile normalization) and the impact of feature stability across observers and centers were studied. Model performances were assessed by area under the curve (AUC), sensitivity, and specificity. RESULTS: Performance of the published model (AUC/sensitivity/specificity: 0.74/0.75/0.60) drops when applied on the validation cohort (AUC/sensitivity/specificity: 0.64/0.68/0.60). The performance of the full validation cohort improves slightly when the model is validated using a patient group with comparable HPV status of the tumor (AUC/sensitivity/specificity: 0.68/0.74/0.60), using patients acquired with a slice thickness of 4 mm (AUC/sensitivity/specificity: 0.67/0.73/0.57), or when quantile harmonization was performed (AUC/sensitivity/specificity: 0.66/0.69/0.60). CONCLUSION: The previously published model shows its generalizability and can be applied on data acquired from different vendors and protocols. Harmonization techniques and subgroup definition influence performance of predictive radiomics models. KEY POINTS: • Radiomics, a noninvasive quantitative image analysis technique, can support the radiologist by enhancing diagnostic accuracy and/or treatment decision-making. • A previously published model shows its generalizability and could be applied on data acquired from different vendors and protocols.

Development and multicenter validation of a multiparametric imaging model to predict treatment response in rectal cancer.

OBJECTIVES: To develop and validate a multiparametric model to predict neoadjuvant treatment response in rectal cancer at baseline using a heterogeneous multicenter MRI dataset. METHODS: Baseline staging MRIs (T2W (T2-weighted)-MRI, diffusion-weighted imaging (DWI) / apparent diffusion coefficient (ADC)) of 509 patients (9 centres) treated with neoadjuvant chemoradiotherapy (CRT) were collected. Response was defined as (1) complete versus incomplete response, or (2) good (Mandard tumor regression grade (TRG) 1-2) versus poor response (TRG3-5). Prediction models were developed using combinations of the following variable groups: (1) Non-imaging: age/sex/tumor-location/tumor-morphology/CRT-surgery interval (2) Basic staging: cT-stage/cN-stage/mesorectal fascia involvement, derived from (2a) original staging reports, or (2b) expert re-evaluation (3) Advanced staging: variables from 2b combined with cTN-substaging/invasion depth/extramural vascular invasion/tumor length (4) Quantitative imaging: tumour volume + first-order histogram features (from T2W-MRI and DWI/ADC) Models were developed with data from 6 centers (n = 412) using logistic regression with the Least Absolute Shrinkage and Selector Operator (LASSO) feature selection, internally validated using repeated (n = 100) random hold-out validation, and externally validated using data from 3 centers (n = 97). RESULTS: After external validation, the best model (including non-imaging and advanced staging variables) achieved an area under the curve of 0.60 (95%CI=0.48-0.72) to predict complete response and 0.65 (95%CI=0.53-0.76) to predict a good response. Quantitative variables did not improve model performance. Basic staging variables consistently achieved lower performance compared to advanced staging variables. CONCLUSIONS: Overall model performance was moderate. Best results were obtained using advanced staging variables, highlighting the importance of good-quality staging according to current guidelines. Quantitative imaging features had no added value (in this heterogeneous dataset). CLINICAL RELEVANCE STATEMENT: Predicting tumour response at baseline could aid in tailoring neoadjuvant therapies for rectal cancer. This study shows that image-based prediction models are promising, though are negatively affected by variations in staging quality and MRI acquisition, urging the need for harmonization. KEY POINTS: This multicenter study combining clinical information and features derived from MRI rendered disappointing performance to predict response to neoadjuvant treatment in rectal cancer. Best results were obtained with the combination of clinical baseline information and state-of-the-art image-based staging variables, highlighting the importance of good quality staging according to current guidelines and staging templates. No added value was found for quantitative imaging features in this multicenter retrospective study. This is likely related to acquisition variations, which is a major problem for feature reproducibility and thus model generalizability.

Sources of variation in multicenter rectal MRI data and their effect on radiomics feature reproducibility.

OBJECTIVES: To investigate sources of variation in a multicenter rectal cancer MRI dataset focusing on hardware and image acquisition, segmentation methodology, and radiomics feature extraction software. METHODS: T2W and DWI/ADC MRIs from 649 rectal cancer patients were retrospectively acquired in 9 centers. Fifty-two imaging features (14 first-order/6 shape/32 higher-order) were extracted from each scan using whole-volume (expert/non-expert) and single-slice segmentations using two different software packages (PyRadiomics/CapTk). Influence of hardware, acquisition, and patient-intrinsic factors (age/gender/cTN-stage) on ADC was assessed using linear regression. Feature reproducibility was assessed between segmentation methods and software packages using the intraclass correlation coefficient. RESULTS: Image features differed significantly (p < 0.001) between centers with more substantial variations in ADC compared to T2W-MRI. In total, 64.3% of the variation in mean ADC was explained by differences in hardware and acquisition, compared to 0.4% by patient-intrinsic factors. Feature reproducibility between expert and non-expert segmentations was good to excellent (median ICC 0.89-0.90). Reproducibility for single-slice versus whole-volume segmentations was substantially poorer (median ICC 0.40-0.58). Between software packages, reproducibility was good to excellent (median ICC 0.99) for most features (first-order/shape/GLCM/GLRLM) but poor for higher-order (GLSZM/NGTDM) features (median ICC 0.00-0.41). CONCLUSIONS: Significant variations are present in multicenter MRI data, particularly related to differences in hardware and acquisition, which will likely negatively influence subsequent analysis if not corrected for. Segmentation variations had a minor impact when using whole volume segmentations. Between software packages, higher-order features were less reproducible and caution is warranted when implementing these in prediction models. KEY POINTS: • Features derived from T2W-MRI and in particular ADC differ significantly between centers when performing multicenter data analysis. • Variations in ADC are mainly (> 60%) caused by hardware and image acquisition differences and less so (< 1%) by patient- or tumor-intrinsic variations. • Features derived using different image segmentations (expert/non-expert) were reproducible, provided that whole-volume segmentations were used. When using different feature extraction software packages with similar settings, higher-order features were less reproducible.

Studying local tumour heterogeneity on MRI and FDG-PET/CT to predict response to neoadjuvant chemoradiotherapy in rectal cancer.

OBJECTIVE: To investigate whether quantifying local tumour heterogeneity has added benefit compared to global tumour features to predict response to chemoradiotherapy using pre-treatment multiparametric PET and MRI data. METHODS: Sixty-one locally advanced rectal cancer patients treated with chemoradiotherapy and staged at baseline with MRI and FDG-PET/CT were retrospectively analyzed. Whole-tumour volumes were segmented on the MRI and PET/CT scans from which global tumour features (T2Wvolume/T2Wentropy/ADCmean/SUVmean/TLG/CTmean-HU) and local texture features (histogram features derived from local entropy/mean/standard deviation maps) were calculated. These respective feature sets were combined with clinical baseline parameters (e.g. age/gender/TN-stage) to build multivariable prediction models to predict a good (Mandard TRG1-2) versus poor (Mandard TRG3-5) response to chemoradiotherapy. Leave-one-out cross-validation (LOOCV) with bootstrapping was performed to estimate performance in an 'independent' dataset. RESULTS: When using only imaging features, local texture features showed an AUC = 0.81 versus AUC = 0.74 for global tumour features. After internal cross-validation (LOOCV), AUC to predict a good response was the highest for the combination of clinical baseline variables + global tumour features (AUC = 0.83), compared to AUC = 0.79 for baseline + local texture and AUC = 0.76 for all combined (baseline + global + local texture). CONCLUSION: In imaging-based prediction models, local texture analysis has potential added value compared to global tumour features to predict response. However, when combined with clinical baseline parameters such as cTN-stage, the added value of local texture analysis appears to be limited. The overall performance to predict response when combining baseline variables with quantitative imaging parameters is promising and warrants further research. KEY POINTS: • Quantification of local tumour texture on pre-therapy FDG-PET/CT and MRI has potential added value compared to global tumour features to predict response to chemoradiotherapy in rectal cancer. • However, when combined with clinical baseline parameters such as cTN-stage, the added value of local texture over global tumour features is limited. • Predictive performance of our optimal model-combining clinical baseline variables with global quantitative tumour features-was encouraging (AUC 0.83), warranting further research in this direction on a larger scale.

Value of combined multiparametric MRI and FDG-PET/CT to identify well-responding rectal cancer patients before the start of neoadjuvant chemoradiation.

OBJECTIVES: To explore the value of multiparametric MRI combined with FDG-PET/CT to identify well-responding rectal cancer patients before the start of neoadjuvant chemoradiation. METHODS: Sixty-one locally advanced rectal cancer patients who underwent a baseline FDG-PET/CT and MRI (T2W + DWI) and received long-course neoadjuvant chemoradiotherapy were retrospectively analysed. Tumours were delineated on MRI and PET/CT from which the following quantitative parameters were calculated: T2W volume and entropy, ADC mean and entropy, CT density (mean-HU), SUV maximum and mean, metabolic tumour volume (MTV42%) and total lesion glycolysis (TLG). These features, together with sex, age, mrTN-stage ("baseline parameters") and the CRT-surgery interval were analysed using multivariable stepwise logistic regression. Outcome was a good (TRG 1-2) versus poor histopathological response. Performance (AUC) to predict response was compared for different combinations of baseline ± quantitative imaging parameters and performance in an 'independent' dataset was estimated using bootstrapped leave-one-out cross-validation (LOOCV). RESULTS: The optimal multivariable prediction model consisted of a combination of baseline + quantitative imaging parameters and included mrT-stage (OR 0.004, p < 0.001), T2W-signal entropy (OR 7.81, p = 0.0079) and T2W volume (OR 1.028, p = 0.0389) as the selected predictors. AUC in the study dataset was 0.88 and 0.83 after LOOCV. No PET/CT features were selected as predictors. CONCLUSIONS: A multivariable model incorporating mrT-stage and quantitative parameters from baseline MRI can aid in identifying well-responding patients before the start of treatment. Addition of FDG-PET/CT is not beneficial. KEY POINTS: • A multivariable model incorporating the mrT-stage and quantitative features derived from baseline MRI can aid in identifying well-responding patients before the start of neoadjuvant chemoradiotherapy. • mrT-stage was the strongest predictor in the model and was complemented by the tumour volume and signal entropy calculated from T2W-MRI. • Adding quantitative features derived from pre-treatment PET/CT or DWI did not contribute to the model's predictive performance.

A Pattern-Based Approach Combining Tumor Morphology on MRI With Distinct Signal Patterns on Diffusion-Weighted Imaging to Assess Response of Rectal Tumors After Chemoradiotherapy.

BACKGROUND: Diffusion-weighted imaging is increasingly used in rectal cancer MRI to assess response after chemoradiotherapy. Certain pitfalls (eg, artefacts) may hamper diffusion-MRI assessment, leading to suboptimal diagnostic performance. Combining diffusion-weighted MRI with the underlying morphology on standard (T2-weighted) MRI may help overcome these pitfalls. OBJECTIVE: The purpose of this study was to evaluate the diagnostic performance of a pattern-based approach combining tumor morphology on T2-weighted MRI with distinct diffusion-weighted imaging signal patterns to assess response after chemoradiotherapy in rectal cancer. DESIGN: Response to chemoradiotherapy was scored according to 4 patterns: 1) cases with either a clear residual mass with corresponding high-diffusion signal (A+) or completely normalized wall without diffusion signal (A-); 2) cases with circular and/or irregular fibrosis with (B+) or without (B-) small foci of diffusion signal scattered throughout the fibrosis; 3) cases with semicircular fibrosis with (C+) or without (C-) high diffusion signal at the inner margin of the fibrosis; and 4) polypoid tumors showing regression of the polyp and fibrosis at the site of the stalk with (D+) or without (D-) focal high-diffusion signal in the stalk. A total of 75 cases were rescored by an independent second reader to study interobserver variations. Standard of reference was histopathology or long-term outcome. SETTINGS: The study was conducted at a single tertiary referral center. PATIENTS: A total of 222 patients with locally advanced rectal cancer undergoing chemoradiotherapy were included. MAIN OUTCOME MEASURES: Diagnostic performance to discriminate between a complete response and residual tumor was measured. RESULTS: The pattern-based approach resulted in a sensitivity of 94%, specificity of 77%, positive predictive value of 88%, negative predictive value of 87%, and overall accuracy of 88% to differentiate between tumor versus complete response. Accuracies per pattern were 100% (A), 74% (B), 86% (C), and 92% (D). Interobserver agreement was good (κ = 0.75). LIMITATIONS: The study included no comparison with routine (nonpattern) diffusion-MRI assessment. CONCLUSIONS: A pattern-based approach combining tumor morphology with distinct diffusion-weighted imaging patterns results in good diagnostic performance to assess response. See Video Abstract at http://links.lww.com/DCR/A433.

Multi-sequence MRI radiomics of colorectal liver metastases: Which features are reproducible across readers?

PURPOSE: To assess the inter-reader reproducibility of radiomics features on multiple MRI sequences after segmentations of colorectal liver metastases (CRLM). METHOD: 30 CRLM (in 23 patients) were manually delineated by three readers on MRI before the start of chemotherapy on the contrast enhanced T1-weighted images (CE-T1W) in the portal venous phase, T2-weighted images (T2W) and b800 diffusion weighted images (DWI). DWI delineations were copied to the ADC-maps. 107 radiomics features were extracted per sequence. The intraclass correlation coefficient (ICC) was calculated per feature. Features were considered reproducible if ICC > 0.9. RESULTS: 90% of CE-T1W features were reproducible with a median ICC of 0.98 (range 0.76-1.00). 81% of DWI features were robust with median ICC = 0.97 (range 0.38-1.00). The T2W features had a median ICC of 0.96 (range 0.55-0.99) and were reproducible in 80%. ADC showed the lowest number of reproducible features with 58% and median ICC = 0.91 (range 0.38-0.99) When considering the lower bound of the ICC 95% confidence intervals, 58%, 66%, 54% and 29% reached 0.9 for the CE-T1W, DWI, T2W and ADC features, respectively. The feature class with the best reproducibility differed per sequence. CONCLUSIONS: The majority of MRI radiomics features from CE-T1W, T2W, DWI and ADC in colorectal liver metastases were robust for segmentation variability between readers. The CE-T1W yielded slightly better reproducibility results compared to DWI and T2W. The ADC features seem more susceptible to reader differences compared to the other three sequences.

Evolutions in rectal cancer MRI staging and risk stratification in The Netherlands.

PURPOSE: To analyze how the MRI reporting of rectal cancer has evolved (following guideline updates) in The Netherlands. METHODS: Retrospective analysis of 712 patients (2011-2018) from 8 teaching hospitals in The Netherlands with available original radiological staging reports that were re-evaluated by a dedicated MR expert using updated guideline criteria. Original reports were classified as "free-text," "semi-structured," or "template" and completeness of reporting was documented. Patients were categorized as low versus high risk, first based on the original reports (high risk = cT3-4, cN+, and/or cMRF+) and then based on the expert re-evaluations (high risk = cT3cd-4, cN+, MRF+, and/or EMVI+). Evolutions over time were studied by splitting the inclusion period in 3 equal time periods. RESULTS: A significant increase in template reporting was observed (from 1.6 to 17.6-29.6%; p < 0.001), along with a significant increase in the reporting of cT-substage, number of N+ and extramesorectal nodes, MRF invasion and tumor-MRF distance, EMVI, anal sphincter involvement, and tumor morphology and circumference. Expert re-evaluation changed the risk classification from high to low risk in 18.0% of cases and from low to high risk in 1.7% (total 19.7%). In the majority (17.9%) of these cases, the changed risk classification was likely (at least in part) related to use of updated guideline criteria, which mainly led to a reduction in high-risk cT-stage and nodal downstaging. CONCLUSION: Updated concepts of risk stratification have increasingly been adopted, accompanied by an increase in template reporting and improved completeness of reporting. Use of updated guideline criteria resulted in considerable downstaging (of mainly high-risk cT-stage and nodal stage).

A decade of multi-modality PET and MR imaging in abdominal oncology.

OBJECTIVES: To investigate trends observed in a decade of published research on multimodality PET(/CT)+MR imaging in abdominal oncology, and to explore how these trends are reflected by the use of multimodality imaging performed at our institution. METHODS: First, we performed a literature search (2009-2018) including all papers published on the multimodality combination of PET(/CT) and MRI in abdominal oncology. Retrieved papers were categorized according to a structured labelling system, including study design and outcome, cancer and lesion type under investigation and PET-tracer type. Results were analysed using descriptive statistics and evolutions over time were plotted graphically. Second, we performed a descriptive analysis of the numbers of MRI, PET/CT and multimodality PET/CT+MRI combinations (performed within a ≤14 days interval) performed during a similar time span at our institution. RESULTS: Published research papers involving multimodality PET(/CT)+MRI combinations showed an impressive increase in numbers, both for retrospective combinations of PET/CT and MRI, as well as hybrid PET/MRI. Main areas of research included new PET-tracers, visual PET(/CT)+MRI assessment for staging, and (semi-)quantitative analysis of PET-parameters compared to or combined with MRI-parameters as predictive biomarkers. In line with literature, we also observed a vast increase in numbers of multimodality PET/CT+MRI imaging in our institutional data. CONCLUSIONS: The tremendous increase in published literature on multimodality imaging, reflected by our institutional data, shows the continuously growing interest in comprehensive multivariable imaging evaluations to guide oncological practice. ADVANCES IN KNOWLEDGE: The role of multimodality imaging in oncology is rapidly evolving. This paper summarizes the main applications and recent developments in multimodality imaging, with a specific focus on the combination of PET+MRI in abdominal oncology.

Radiomics performs comparable to morphologic assessment by expert radiologists for prediction of response to neoadjuvant chemoradiotherapy on baseline staging MRI in rectal cancer.

PURPOSE: To compare the performance of advanced radiomics analysis to morphological assessment by expert radiologists to predict a good or complete response to chemoradiotherapy in rectal cancer using baseline staging MRI. MATERIALS AND METHODS: We retrospectively assessed the primary staging MRIs [prior to chemoradiotherapy (CRT)] of 133 rectal cancer patients from 2 centers. First, two expert radiologists subjectively estimated the likelihood of achieving a "complete response" (ypT0) and "good response" (TRG 1-2), using a 5-point score (based on TN-stage, MRF/EMVI-status, size/signal/shape). Next, tumor volumes were segmented on high b value DWI (semi-automated, corrected by 2 non-expert and 2-expert readers, resulting in 5 segmentations), copied to the remaining sequences after which a total of 2505 radiomic features were extracted from T2W, low and high b value DWI and ADC. Stability of features for noise due to inter-reader and inter-scanner and protocol variations was assessed using intraclass correlation (ICC) and the Kruskal-Wallis test. Using data from center 1 (n = 86; training set), top 9 features were selected using minimum Redundancy Maximum Relevance and combined in a logistic regression model. Finally, diagnostic performance of the fitted models was assessed on data from center 2 (n = 47; validation set) and compared to the performance of the radiologists. RESULTS: The Radiomic models resulted in AUCs of 0.69-0.79 (with similar results for the segmentations performed by expert/non-expert readers) to predict response, results similar to the morphologic prediction by the expert radiologists (AUC 0.67-0.83). Radiomics using semi-automatically generated segmentations (without manual input) did not result in significant predictive performance. CONCLUSIONS: Radiomics could predict response to therapy with comparable diagnostic performance as expert radiologists, regardless of whether image segmentation was performed by non-expert or expert readers, indicating that expert input is not required in order for the radiomics workflow to produce significant predictive performance.

Peritumoral radiomics features predict distant metastasis in locally advanced NSCLC.

PURPOSE: Radiomics provides quantitative tissue heterogeneity profiling and is an exciting approach to developing imaging biomarkers in the context of precision medicine. Normal-appearing parenchymal tissues surrounding primary tumors can harbor microscopic disease that leads to increased risk of distant metastasis (DM). This study assesses whether computed-tomography (CT) imaging features of such peritumoral tissues can predict DM in locally advanced non-small cell lung cancer (NSCLC). MATERIAL AND METHODS: 200 NSCLC patients of histological adenocarcinoma were included in this study. The investigated lung tissues were tumor rim, defined to be 3mm of tumor and parenchymal tissue on either side of the tumor border and the exterior region extended from 3 to 9mm outside of the tumor. Fifteen stable radiomic features were extracted and evaluated from each of these regions on pre-treatment CT images. For comparison, features from expert-delineated tumor contours were similarly prepared. The patient cohort was separated into training and validation datasets for prognostic power evaluation. Both univariable and multivariable analyses were performed for each region using concordance index (CI). RESULTS: Univariable analysis reveals that six out of fifteen tumor rim features were significantly prognostic of DM (p-value < 0.05), as were ten features from the visible tumor, and only one of the exterior features was. Multivariablely, a rim radiomic signature achieved the highest prognostic performance in the independent validation sub-cohort (CI = 0.64, p-value = 2.4×10-5) significantly over a multivariable clinical model (CI = 0.53), a visible tumor radiomics model (CI = 0.59), or an exterior tissue model (CI = 0.55). Furthermore, patient stratification by the combined rim signature and clinical predictor led to a significant improvement on the clinical predictor alone and also outperformed stratification using the combined tumor signature and clinical predictor. CONCLUSIONS: We identified peritumoral rim radiomic features significantly associated with DM. This study demonstrated that peritumoral imaging characteristics may provide additional valuable information over the visible tumor features for patient risk stratification due to cancer metastasis.

Biochemical recurrence prediction after radiotherapy for prostate cancer with T2w magnetic resonance imaging radiomic features.

BACKGROUND AND PURPOSE: High-risk prostate cancer patients are frequently treated with external-beam radiotherapy (EBRT). Of all patients receiving EBRT, 15-35% will experience biochemical recurrence (BCR) within five years. Magnetic resonance imaging (MRI) is commonly acquired as part of the diagnostic procedure and imaging-derived features have shown promise in tumour characterisation and biochemical recurrence prediction. We investigated the value of imaging features extracted from pre-treatment T2w anatomical MRI to predict five year biochemical recurrence in high-risk patients treated with EBRT. MATERIALS AND METHODS: In a cohort of 120 high-risk patients, imaging features were extracted from the whole-prostate and a margin surrounding it. Intensity, shape and textural features were extracted from the original and filtered T2w-MRI scans. The minimum-redundancy maximum-relevance algorithm was used for feature selection. Random forest and logistic regression classifiers were used in our experiments. The performance of a logistic regression model using the patient's clinical features was also investigated. To assess the prediction accuracy we used stratified 10-fold cross validation and receiver operating characteristic analysis, quantified by the area under the curve (AUC). RESULTS: A logistic regression model built using whole-prostate imaging features obtained an AUC of 0.63 in the prediction of BCR, outperforming a model solely based on clinical variables (AUC = 0.51). Combining imaging and clinical features did not outperform the accuracy of imaging alone. CONCLUSIONS: These results illustrate the potential of imaging features alone to distinguish patients with an increased risk of recurrence, even in a clinically homogeneous cohort.

Author Correction: Deep Learning for Fully-Automated Localization and Segmentation of Rectal Cancer on Multiparametric MR.

A correction to this article has been published and is linked from the HTML version of this paper. The error has been fixed in the paper.

Gas-induced susceptibility artefacts on diffusion-weighted MRI of the rectum at 1.5 T - Effect of applying a micro-enema to improve image quality.

PURPOSE: Assess whether application of a micro-enema can reduce gas-induced susceptibility artefacts in Single-shot Echo Planar Imaging (EPI) Diffusion-weighted imaging of the rectum at 1.5 T. MATERIALS AND METHODS: Retrospective analysis of n = 50 rectal cancer patients who each underwent multiple DWI-MRIs (1.5 T) from 2012 to 2016 as part of routine follow-up during a watch-and-wait approach after chemoradiotherapy. From March 2014 DWI-MRIs were routinely acquired after application of a preparatory micro-enema (Microlax®; 5 ml; self-administered shortly before acquisition); before March 2014 no bowel preparation was given. In total, 335 scans were scored by an experienced reader for the presence/severity of gas-artefacts (on b1000 DWI), ranging from 0 (no artefact) to 5 (severe artefact). A score ≥3 (moderate-severe) was considered a clinically relevant artefact. A random sample of 100 scans was re-assessed by a second independent reader to study inter-observer effects. Scores were compared between the scans performed without and with a preparatory micro-enema using univariable and multivariable logistic regression taking into account potential confounding factors (age/gender, acquisition parameters, MRI-hardware, rectoscopy prior to MRI). RESULTS: Clinically relevant gas-artefacts were seen in 24.3% (no micro-enema) vs. 3.7% (micro-enema), odds ratios were 0.118 in univariable and 0.230 in multivariable regression (P = 0.0005 and 0.0291). Mean severity score (±SD) was 1.19 ±â€¯1.71 (no-enema) vs 0.32 ±â€¯0.77 (micro-enema), odds ratios were 0.321 (P < 0.0001) and 0.489 (P = 0.0461) in uni- and multivariable regression, respectively. Inter-observer agreement was excellent (κ0.85). CONCLUSION: Use of a preparatory micro-enema shortly before rectal EPI-DWI examinations performed at 1.5 T MRI significantly reduces both the incidence and severity of gas-induced artefacts, compared to examinations performed without bowel preparation.

Computational Radiomics System to Decode the Radiographic Phenotype.

Radiomics aims to quantify phenotypic characteristics on medical imaging through the use of automated algorithms. Radiomic artificial intelligence (AI) technology, either based on engineered hard-coded algorithms or deep learning methods, can be used to develop noninvasive imaging-based biomarkers. However, lack of standardized algorithm definitions and image processing severely hampers reproducibility and comparability of results. To address this issue, we developed PyRadiomics, a flexible open-source platform capable of extracting a large panel of engineered features from medical images. PyRadiomics is implemented in Python and can be used standalone or using 3D Slicer. Here, we discuss the workflow and architecture of PyRadiomics and demonstrate its application in characterizing lung lesions. Source code, documentation, and examples are publicly available at www.radiomics.io With this platform, we aim to establish a reference standard for radiomic analyses, provide a tested and maintained resource, and to grow the community of radiomic developers addressing critical needs in cancer research. Cancer Res; 77(21); e104-7. ©2017 AACR.

Deep Learning for Fully-Automated Localization and Segmentation of Rectal Cancer on Multiparametric MR.

Multiparametric Magnetic Resonance Imaging (MRI) can provide detailed information of the physical characteristics of rectum tumours. Several investigations suggest that volumetric analyses on anatomical and functional MRI contain clinically valuable information. However, manual delineation of tumours is a time consuming procedure, as it requires a high level of expertise. Here, we evaluate deep learning methods for automatic localization and segmentation of rectal cancers on multiparametric MR imaging. MRI scans (1.5T, T2-weighted, and DWI) of 140 patients with locally advanced rectal cancer were included in our analysis, equally divided between discovery and validation datasets. Two expert radiologists segmented each tumor. A convolutional neural network (CNN) was trained on the multiparametric MRIs of the discovery set to classify each voxel into tumour or non-tumour. On the independent validation dataset, the CNN showed high segmentation accuracy for reader1 (Dice Similarity Coefficient (DSC = 0.68) and reader2 (DSC = 0.70). The area under the curve (AUC) of the resulting probability maps was very high for both readers, AUC = 0.99 (SD = 0.05). Our results demonstrate that deep learning can perform accurate localization and segmentation of rectal cancer in MR imaging in the majority of patients. Deep learning technologies have the potential to improve the speed and accuracy of MRI-based rectum segmentations.

Automated and Semiautomated Segmentation of Rectal Tumor Volumes on Diffusion-Weighted MRI: Can It Replace Manual Volumetry?

PURPOSE: Diffusion-weighted imaging (DWI) tumor volumetry is promising for rectal cancer response assessment, but an important drawback is that manual per-slice tumor delineation can be highly time consuming. This study investigated whether manual DWI-volumetry can be reproduced using a (semi)automated segmentation approach. METHODS AND MATERIALS: Seventy-nine patients underwent magnetic resonance imaging (MRI) that included DWI (highest b value [b1000 or b1100]) before and after chemoradiation therapy (CRT). Tumor volumes were assessed on b1000 (or b1100) DWI before and after CRT by means of (1) automated segmentation (by 2 inexperienced readers), (2) semiautomated segmentation (manual adjustment of the volumes obtained by method 1 by 2 radiologists), and (3) manual segmentation (by 2 radiologists); this last assessment served as the reference standard. Intraclass correlation coefficients (ICC) and Dice similarity indices (DSI) were calculated to evaluate agreement between different methods and observers. Measurement times (from a radiologist's perspective) were recorded for each method. RESULTS: Tumor volumes were not significantly different among the 3 methods, either before or after CRT (P=.08 to .92). ICCs compared to manual segmentation were 0.80 to 0.91 and 0.53 to 0.66 before and after CRT, respectively, for the automated segmentation and 0.91 to 0.97 and 0.61 to 0.75, respectively, for the semiautomated method. Interobserver agreement (ICC) pre and post CRT was 0.82 and 0.59 for automated segmentation, 0.91 and 0.73 for semiautomated segmentation, and 0.91 and 0.75 for manual segmentation, respectively. Mean DSI between the automated and semiautomated method were 0.83 and 0.58 pre-CRT and post-CRT, respectively; DSI between the automated and manual segmentation were 0.68 and 0.42 and 0.70 and 0.41 between the semiautomated and manual segmentation, respectively. Median measurement time for the radiologists was 0 seconds (pre- and post-CRT) for the automated method, 41 to 69 seconds (pre-CRT) and 60 to 67 seconds (post-CRT) for the semiautomated method, and 180 to 296 seconds (pre-CRT) and 84 to 91 seconds (post-CRT) for the manual method. CONCLUSIONS: DWI volumetry using a semiautomated segmentation approach is promising and a potentially time-saving alternative to manual tumor delineation, particularly for primary tumor volumetry. Once further optimized, it could be a helpful tool for tumor response assessment in rectal cancer.