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

Prediction of pathological response and lymph node metastasis after neoadjuvant therapy in rectal cancer through tumor and mesorectal MRI radiomic features

Establishing predictive models for the pathological response and lymph node metastasis in locally advanced rectal cancer (LARC) treated with neoadjuvant chemoradiotherapy (nCRT) based on MRI radiomic features derived from the tumor and mesorectal compartment (MC). This study included 209 patients with LARC who underwent rectal MRI both before and after nCRT. The patients were divided into a training set (n = 146) and a test set (n = 63). Regions of interest (ROIs) for the tumor and MC were delineated on both pre- and post-nCRT MRI images. Radiomic features were extracted, and delta radiomic features were computed. The predictive endpoints were pathological complete response (pCR), pathological good response (pGR), and lymph node metastasis (LNM). Feature selection for various models involved sequentially removing features with a correlation coefficient > 0.9, and features with P-values ≥ 0.05 in univariate analysis, followed by LASSO regression on the remaining features. Logistic regression models were developed, and their performance was evaluated using the area under the receiver operating characteristic curve (AUC). Among the 209 LARC patients, the number of patients achieving pCR, pGR, and LNM were 44, 118, and 40, respectively. The optimal model for predicting each endpoint is the combined model that incorporates pre- and delta-radiomics features for both the tumor and MC. These models exhibited superior performance with AUC values of 0.874 (for pCR), 0.801 (for pGR), and 0.826 (for LNM), outperforming the MRI tumor regression grade (mrTRG) which yielded AUC values of 0.800, 0.715, and 0.603, respectively. The results demonstrate the potential utility of the tumor and MC radiomics features, in predicting treatment efficacy among LARC patients undergoing nCRT.

Contrast Enhanced CT Radiogenomics in a Retrospective NSCLC Cohort: Models, Attempted Validation of a Published Model and the Relevance of the Clinical Context

RATIONALE AND OBJECTIVE

To develop a radiogenomic predictive model for non-small cell lung cancer (NSCLC) patients studied through contrast enhanced chest computed tomography (CE-CT) targeting the most frequent gene alterations. M&M: A retrospective study of patients with NSCLC imaged with CE-CT before treatment and had their tumor genomics sequenced at our institution was performed. Data was gathered from their imaging studies, their electronic medical records and a web-based database search (cBioPortal.ca). All of the patient data was tabulated for analysis. Two predictive models (M1 & M2) were created using different approaches and a third model was extracted from the literature to also be tested in our population.

RESULTS

Out of 157 patients, eighty were male (51%) and 124 (79%) had a history of smoking. The three most prevalent genes were KRAS, TP53 and EGFR. The M1 radiomics-only model median AUC were 0.61 (TP53), 0.53 (KRAS) and 0.64 (EGFR) and for M1 radiomics + clinical were 0.61 (TP53), 0.61 (KRAS) and 0.80 (EGFR). The M2 radiomics-only model median AUC were 0.63 (TP53), 0.60 (KRAS) and 0.65 (EGFR) and for M2 radiomics + clinical were 0.64 (TP53), 0.62 (KRAS) and 0.81 (EGFR). The external EGFR radiomic model showed an AUC of 0.69 and 0.86 for the radiomics-only and combined radiomics + clinical respectively.

CONCLUSION

Our study was able to provide robust predictive radiomics model evaluation for the detection of TP53, KRAS and EGFR. We also compared our performance with an already published model and observed how impactful clinical variables can be on models' performance.

CLINICAL RELEVANCE STATEMENT

Identifying tumor mutations in patients that can't undergo biopsy is critical for their outcomes.

KEYPOINTS

• Tumor genomic profiling is critical for treatment selection • CE-CT radiomics produce robust predictive models comparable to those already published • Clinical variables should be considered/included in predictive models.

Multiparametric magnetic resonance imaging (MRI)-based radiomics model explained by the Shapley Additive exPlanations (SHAP) method for predicting complete response to neoadjuvant chemoradiotherapy in locally advanced rectal cancer: a multicenter retrospective study

Background

Predicting the response to neoadjuvant chemoradiotherapy (nCRT) before initiating treatment is essential for tailoring therapeutic strategies and monitoring prognosis in locally advanced rectal cancer (LARC). In this study, we aimed to develop and validate radiomic-based models to predict clinical and pathological complete responses (cCR and pCR, respectively) by incorporating the Shapley Additive exPlanations (SHAP) method for model interpretation.

Methods

A total of 285 patients with complete pretreatment clinical characteristics and T1-weighted (T1W) and T2-weighted (T2W) magnetic resonance imaging (MRI) at 3 centers were retrospectively recruited. The features of tumor lesions were extracted by PyRadiomics and selected using least absolute shrinkage and selection operator (LASSO) algorithm. The selected features were used to build multilayer perceptron (MLP) models alone or combined with clinical features. Area under the receiver operating characteristic curve (AUC), decision curve, and calibration curve were applied to evaluate performance of models. The SHAP method was adopted to explain the prediction models.

Results

The radiomic-based models all showed better performances than clinical models. The clinical-radiomic models showed the best differentiation on cCR and pCR with mean AUCs of 0.718 and 0.810 in the validation set, respectively. The decision curves of the clinical-radiomic models showed its values in clinical application. The SHAP method powerfully interpreted the prediction models both at a holistic and individual levels.

Conclusions

Our study highlights that the radiomic-based prediction models have more excellent abilities than clinical models and can effectively predict treatment response and optimize therapeutic strategies for patients with LARCs.

Solving the Pervasive Problem of Protocol Non-Compliance in MRI using an Open-Source tool mrQA

Pooling data across diverse sources acquired by multisite consortia requires compliance with a predefined reference protocol i.e., ensuring different sites and scanners for a given project have used identical or compatible MR physics parameter values. Traditionally, this has been an arduous and manual process due to difficulties in working with the complicated DICOM standard and lack of resources allocated towards protocol compliance. Moreover, issues of protocol compliance is often overlooked for lack of realization that parameter values are routinely improvised/modified locally at various sites. The inconsistencies in acquisition protocols can reduce SNR, statistical power, and in the worst case, may invalidate the results altogether. An open-source tool, mrQA was developed to automatically assess protocol compliance on standard dataset formats such as DICOM and BIDS, and to study the patterns of non-compliance in over 20 open neuroimaging datasets, including the large ABCD study. The results demonstrate that the lack of compliance is rather pervasive. The frequent sources of non-compliance include but are not limited to deviations in Repetition Time, Echo Time, Flip Angle, and Phase Encoding Direction. It was also observed that GE and Philips scanners exhibited higher rates of non-compliance relative to the Siemens scanners in the ABCD dataset. Continuous monitoring for protocol compliance is strongly recommended before any pre/post-processing, ideally right after the acquisition, to avoid the silent propagation of severe/subtle issues. Although, this study focuses on neuroimaging datasets, the proposed tool mrQA can work with any DICOM-based datasets.

Progress of MRI in predicting the circumferential resection margin of rectal cancer: A narrative review

Rectal cancer (RC) is the third most frequently diagnosed cancer worldwide, and the status of its circumferential resection margin (CRM) is of paramount significance for treatment strategies and prognosis. CRM involvement is defined as tumor touching or within 1 mm from the outermost part of tumor or outer border of the mesorectal or lymph node deposits to the resection margin. The incidence of involved CRM varied from 5.4 % to 36 %, which may associate with an in consistent definition of CRM, the quality of surgeries, and the different examination modalities. Although T and N status are essential factors in determining whether a patient should receive neoadjuvant therapy before surgery, CRM status is a powerful predictor of local and distant recurrence as well as survival rate. This review explores the significance of CRM, the various assessment methods, and the role of magnetic resonance imaging (MRI) and artificial intelligence-based MRI in predicting CRM status. MRI showed potential advantage in predicting CRM status with a high sensitivity and specificity compared to computed tomography (CT). We also discuss MRI advancements in RC imaging, including conventional MRI with body coil, high-resolution MRI with phased-array coil, and endorectal MRI. Along with a discussion of artificial intelligence-based MRI techniques to predict the CRM status of RCs before and after treatments.

Image-based artificial intelligence for the prediction of pathological complete response to neoadjuvant chemoradiotherapy in patients with rectal cancer: a systematic review and meta-analysis

OBJECTIVE

Artificial intelligence (AI) holds enormous potential for noninvasively identifying patients with rectal cancer who could achieve pathological complete response (pCR) following neoadjuvant chemoradiotherapy (nCRT). We aimed to conduct a meta-analysis to summarize the diagnostic performance of image-based AI models for predicting pCR to nCRT in patients with rectal cancer.

METHODS

This study followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. A literature search of PubMed, Embase, Cochrane Library, and Web of Science was performed from inception to July 29, 2023. Studies that developed or utilized AI models for predicting pCR to nCRT in rectal cancer from medical images were included. The Quality Assessment of Diagnostic Accuracy Studies-AI was used to appraise the methodological quality of the studies. The bivariate random-effects model was used to summarize the individual sensitivities, specificities, and areas-under-the-curve (AUCs). Subgroup and meta-regression analyses were conducted to identify potential sources of heterogeneity. Protocol for this study was registered with PROSPERO (CRD42022382374).

RESULTS

Thirty-four studies (9933 patients) were identified. Pooled estimates of sensitivity, specificity, and AUC of AI models for pCR prediction were 82% (95% CI: 76-87%), 84% (95% CI: 79-88%), and 90% (95% CI: 87-92%), respectively. Higher specificity was seen for the Asian population, low risk of bias, and deep-learning, compared with the non-Asian population, high risk of bias, and radiomics (all P < 0.05). Single-center had a higher sensitivity than multi-center (P = 0.001). The retrospective design had lower sensitivity (P = 0.012) but higher specificity (P < 0.001) than the prospective design. MRI showed higher sensitivity (P = 0.001) but lower specificity (P = 0.044) than non-MRI. The sensitivity and specificity of internal validation were higher than those of external validation (both P = 0.005).

CONCLUSIONS

Image-based AI models exhibited favorable performance for predicting pCR to nCRT in rectal cancer. However, further clinical trials are warranted to verify the findings.

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.

The Role of Radiomics in Rectal Cancer

PURPOSE

Radiomics is a promising method for advancing imaging assessment in rectal cancer. This review aims to describe the emerging role of radiomics in the imaging assessment of rectal cancer, including various applications of radiomics based on CT, MRI, or PET/CT.

METHODS

We conducted a literature review to highlight the progress of radiomic research to date and the challenges that need to be addressed before radiomics can be implemented clinically.

RESULTS

The results suggest that radiomics has the potential to provide valuable information for clinical decision-making in rectal cancer. However, there are still challenges in terms of standardization of imaging protocols, feature extraction, and validation of radiomic models. Despite these challenges, radiomics holds great promise for personalized medicine in rectal cancer, with the potential to improve diagnosis, prognosis, and treatment planning. Further research is needed to validate the clinical utility of radiomics and to establish its role in routine clinical practice.

CONCLUSION

Overall, radiomics has emerged as a powerful tool for improving the imaging assessment of rectal cancer, and its potential benefits should not be underestimated.

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.

Radiomics for predicting survival in patients with locally advanced rectal cancer: a systematic review and meta-analysis

Background

Radiomics has recently received considerable research attention for providing potential prognostic biomarkers for locally advanced rectal cancer (LARC). We aimed to comprehensively evaluate the methodological quality and prognostic prediction value of radiomic studies for predicting survival outcomes in patients with LARC.

Methods

The Cochrane, Embase, Medline, and Web of Science databases were searched. The radiomics quality score (RQS), Transparent Reporting of a Multivariable Prediction Model for Individual Prognosis or Diagnosis (TRIPOD) checklist, the Image Biomarkers Standardization Initiative (IBSI) guideline, and the Prediction Model Risk of Bias Assessment Tool were used to assess the quality of the selected studies. A further meta-analysis of hazard ratio (HR) regarding disease-free survival (DFS) and overall survival (OS) was performed.

Results

Among the 358 studies reported, 15 studies were selected for our review. The mean RQS score was 7.73±4.61 (21.5% of the ideal score of 36). The overall TRIPOD adherence rate was 64.4% (251/390). Most of the included studies (60%) were assessed as having a high risk of bias (ROB) overall. The pooled estimates of the HRs were 3.14 [95% confidence interval (CI): 2.12-4.64, P<0.01] for DFS and 3.36 (95% CI: 1.74-6.49, P<0.01) for OS.

Conclusions

Radiomics has potential to noninvasively predict outcome in patients with LARC. However, the overall methodological quality of radiomics studies was low, and the adherence to the TRIPOD statement was moderate. Future radiomics research should put a greater focus on enhancing the methodological quality and considering the influence of higher-order features on reproducibility in radiomics.

Pearls and pitfalls of structured staging and reporting of rectal cancer on MRI: an international multireader study

OBJECTIVES

To investigate uniformity and pitfalls in structured radiological staging of rectal cancer.

METHODS

Twenty-one radiologists (12 countries) staged 75 rectal cancers on MRI using a structured reporting template. Interobserver agreement (IOA) was calculated as the percentage agreement between readers (categorical variables) and Krippendorff's α (continuous variables). Agreement with an expert consensus served as a surrogate standard of reference to estimate diagnostic accuracy. Polychoric correlation coefficients were used to assess correlations between diagnostic confidence and accuracy (=agreement with expert consensus).

RESULTS

Uniformity to diagnose high-risk (≥cT3 ab) versus low-risk (≤cT3 cd) cT-stage, cN0 versus cN+, lateral nodes and tumour deposits, MRF and sphincter involvement, and solid versus mucinous tumours was high with IOA > 80% in the majority of cases (and >80% agreement with expert consensus). Results for assessing extramural vascular invasion, cT-stage (cT1-2/cT3/cT4a/cT4b), cN-stage (cN0/N1/N2), relation to the peritoneal reflection, extent of sphincter involvement (internal/intersphincteric/external) and morphology (solid/annular/semi-annular) were considerably poorer. IOA was high (α = 0.72-0.84) for tumour height/length and extramural invasion depth, but low for tumour-MRF distance and number of (suspicious) nodes (α = 0.05-0.55). There was a significant positive correlation between diagnostic confidence and accuracy (=agreement with expert consensus) (p < 0.001-p = 0.003).

CONCLUSIONS

- Several staging items lacked sufficient reproducibility.- Results for cT- and N-staging g improved when using a dichotomized stratification.- Considering the significant correlation between diagnostic confidence and accuracy, a confidence level may be incorporated into structured reporting for specific items with low reproducibility.

ADVANCES IN KNOWLEDGE

Although structured reporting aims to achieve uniformity in reporting, several items lack sufficient reproducibility and might benefit from dichotomized assessment and incorporating confidence levels.

Sense and nonsense of yT-staging on MRI after chemoradiotherapy in rectal cancer

AIM

The aim of this work was to investigate the value of rectal cancer T-staging on MRI after chemoradiotherapy (ymrT-staging) in relation to the degree of fibrotic transformation of the tumour bed as assessed using the pathological tumour regression grade (pTRG) of Mandard as a standard of reference.

METHOD

Twenty two radiologists, including five rectal MRI experts and 17 'nonexperts' (general/abdominal radiologists), evaluated the ymrT stage on the restaging MRIs of 90 rectal cancer patients after chemoradiotherapy. The ymrT stage was compared with the final ypT stage at histopathology; the percentages of correct staging (ymrT = ypT), understaging (ymrT < ypT) and overstaging (ymrT > ypT) were calculated and compared between patients with predominant tumour at histopathology (pTRG4-5) and patients with predominant fibrosis (pTRG1-3). Interobserver agreement (IOA) was computed using Krippendorff's alpha.

RESULTS

Average ymrT/ypT stage concordance was 48% for the experts and 43% for the nonexperts; ymrT/ypT stage concordance was significantly higher in the pTRG4-5 subgroup (58% vs. 41% for the pTRG1-3 group; p = 0.01), with the best results for the MRI experts. Overstaging was the main source of error, especially in the pTRG1-3 subgroup (average overstaging rate 38%-44% vs. 13%-55% in the pTRG4-5 subgroup). IOA was higher for the expert versus nonexpert readers (α = 0.67 vs. α = 0.39).

CONCLUSIONS

ymrT-staging is moderately accurate; accuracy is higher in poorly responding patients with predominant tumour but low in good responders with predominant fibrosis, resulting in significant overstaging. Radiologists should shift their focus from ymrT-staging to detecting gross residual (and progressive) disease, and identifying potential candidates for organ preservation who would benefit from further clinical and endoscopic evaluation to guide final treatment planning.

Psychophysical Evaluation of Visual vs. Computer-Aided Detection of Brain Lesions on Magnetic Resonance Images

BACKGROUND

Magnetic resonance imaging (MRI) diagnosis is usually performed by analyzing contrast-weighted images, where pathology is detected once it reached a certain visual threshold. Computer-aided diagnosis (CAD) has been proposed as a way for achieving higher sensitivity to early pathology.

PURPOSE

To compare conventional (i.e., visual) MRI assessment of artificially generated multiple sclerosis (MS) lesions in the brain's white matter to CAD based on a deep neural network.

STUDY TYPE

Prospective.

POPULATION

A total of 25 neuroradiologists (15 males, age 39 ± 9, 9 ± 9.8 years of experience) independently assessed all synthetic lesions.

FIELD STRENGTH/SEQUENCE

A 3.0 T, T₂ -weighted multi-echo spin-echo (MESE) sequence.

ASSESSMENT

MS lesions of varying severity levels were artificially generated in healthy volunteer MRI scans by manipulating T₂ values. Radiologists and a neural network were tasked with detecting these lesions in a series of 48 MR images. Sixteen images presented healthy anatomy and the rest contained a single lesion at eight increasing severity levels (6%, 9%, 12%, 15%, 18%, 21%, 25%, and 30% elevation in T₂ ). True positive (TP) rates, false positive (FP) rates, and odds ratios (ORs) were compared between radiological diagnosis and CAD across the range lesion severity levels.

STATISTICAL TESTS

Diagnostic performance of the two approaches was compared using z-tests on TP rates, FP rates, and the logarithm of ORs across severity levels. A P-value <0.05 was considered statistically significant.

RESULTS

ORs of identifying pathology were significantly higher for CAD vis-à-vis visual inspection for all lesions' severity levels. For a 6% change in T₂ value (lowest severity), radiologists' TP and FP rates were not significantly different (P = 0.12), while the corresponding CAD results remained statistically significant.

DATA CONCLUSION

CAD is capable of detecting the presence or absence of more subtle lesions with greater precision than the representative group of 25 radiologists chosen in this study.

LEVEL OF EVIDENCE

1 TECHNICAL EFFICACY: Stage 3.

ComBat Harmonization of Myocardial Radiomic Features Sensitive to Cardiac MRI Acquisition Parameters

Purpose

To investigate the effect of ComBat harmonization methods on the robustness of cardiac MRI-derived radiomic features to variations in imaging parameters.

Materials and Methods

This Health Insurance Portability and Accountability Act-compliant retrospective study used a publicly available data set of 11 healthy controls (mean age, 33 years ± 16 [SD]; six men) and five patients (mean age, 52 years ± 16; four men). A single midventricular short-axis section was acquired with 3-T MRI using cine balanced steady-state free precision, T1-weighted, T2-weighted, T1 mapping, and T2 mapping imaging sequences. Each sequence was acquired using baseline parameters and after variations in flip angle, spatial resolution, section thickness, and parallel imaging. Image registration was performed for all sequences at a per-individual level. Manual myocardial contouring was performed, and 1652 radiomic features per sequence were extracted using baseline and variations in imaging parameters. Radiomic feature stability to change in imaging parameters was assessed using Cohen d sensitivity. The stability of radiomic features was assessed both without and after ComBat harmonization of radiomic features. Three ComBat methods were studied: parametric, nonparametric, and Gaussian mixture model (GMM).

Results

For all sequences combined, 51.4% of features were robust to changes in imaging parameters when no ComBat method was applied. ComBat harmonization substantially increased the number of stable features to 95.1% (95% CI: 94.9, 95.3) when parametric ComBat was used and 90.9% (95% CI: 90.6, 91.2) when nonparametric ComBat was used. GMM combat resulted in only 52.6% stable features.

Conclusion

ComBat harmonization improved the stability of radiomic features to changes in imaging parameters across all cardiac MRI sequences.Keywords: Cardiac MRI, Radiomics, ComBat, Harmonization Supplemental material is available for this article. © RSNA, 2023.

Baseline and early digital [ 18 F]FDG PET/CT and multiparametric MRI contain promising features to predict response to neoadjuvant therapy in locally advanced rectal cancer patients: a pilot study

OBJECTIVE

In this pilot study, we investigated the feasibility of response prediction using digital [ 18 F]FDG PET/computed tomography (CT) and multiparametric MRI before, during, and after neoadjuvant chemoradiation therapy in locally advanced rectal cancer (LARC) patients and aimed to select the most promising imaging modalities and timepoints for further investigation in a larger trial.

METHODS

Rectal cancer patients scheduled to undergo neoadjuvant chemoradiation therapy were prospectively included in this trial, and underwent multiparametric MRI and [ 18 F]FDG PET/CT before, 2 weeks into, and 6-8 weeks after chemoradiation therapy. Two groups were created based on pathological tumor regression grade, that is, good responders (TRG1-2) and poor responders (TRG3-5). Using binary logistic regression analysis with a cutoff value of P  ≤ 0.2, promising predictive features for response were selected.

RESULTS

Nineteen patients were included. Of these, 5 were good responders, and 14 were poor responders. Patient characteristics of these groups were similar at baseline. Fifty-seven features were extracted, of which 13 were found to be promising predictors of response. Baseline [T2: volume, diffusion-weighted imaging (DWI): apparent diffusion coefficient (ADC) mean, DWI: difference entropy], early response (T2: volume change, DWI: ADC mean change) and end-of-treatment presurgical evaluation MRI (T2: gray level nonuniformity, DWI: inverse difference normalized, DWI: gray level nonuniformity normalized), as well as baseline (metabolic tumor volume, total lesion glycolysis) and early response PET/CT (Δ maximum standardized uptake value, Δ peak standardized uptake value corrected for lean body mass), were promising features.

CONCLUSION

Both multiparametric MRI and [ 18 F]FDG PET/CT contain promising imaging features to predict response to neoadjuvant chemoradiotherapy in LARC patients. A future larger trial should investigate baseline, early response, and end-of-treatment presurgical evaluation MRI and baseline and early response PET/CT.

Artificial intelligence-driven radiomics study in cancer: the role of feature engineering and modeling

Modern medicine is reliant on various medical imaging technologies for non-invasively observing patients' anatomy. However, the interpretation of medical images can be highly subjective and dependent on the expertise of clinicians. Moreover, some potentially useful quantitative information in medical images, especially that which is not visible to the naked eye, is often ignored during clinical practice. In contrast, radiomics performs high-throughput feature extraction from medical images, which enables quantitative analysis of medical images and prediction of various clinical endpoints. Studies have reported that radiomics exhibits promising performance in diagnosis and predicting treatment responses and prognosis, demonstrating its potential to be a non-invasive auxiliary tool for personalized medicine. However, radiomics remains in a developmental phase as numerous technical challenges have yet to be solved, especially in feature engineering and statistical modeling. In this review, we introduce the current utility of radiomics by summarizing research on its application in the diagnosis, prognosis, and prediction of treatment responses in patients with cancer. We focus on machine learning approaches, for feature extraction and selection during feature engineering and for imbalanced datasets and multi-modality fusion during statistical modeling. Furthermore, we introduce the stability, reproducibility, and interpretability of features, and the generalizability and interpretability of models. Finally, we offer possible solutions to current challenges in radiomics research.

Radiomic features derived from pretherapeutic MRI predict chemoradiation response in locally advanced rectal cancer

BACKGROUND

The standard treatment for locally advanced rectal cancer (LARC) is neoadjuvant concurrent chemoradiotherapy (CRT) followed by surgical excision. Current evidence suggests a favorable prognosis for those with pathological complete response (pCR), and surgery may be spared for them. We trained and validated regression models for CRT response prediction with selected radiomic features extracted from pretreatment magnetic resonance (MR) images to recruit potential candidates for this watch-and-wait strategy.

METHODS

We retrospectively enrolled patients with LARC who underwent pre-CRT MR imaging between 2010 and 2019. Pathological complete response in surgical specimens after CRT was defined as the ground truth. Quantitative features derived from both unfiltered and filtered images were extracted from manually segmented region of interests on T2-weighted images and selected using variance threshold, univariate statistical tests, and cross-validation least absolute shrinkage and selection operator (Lasso) regression. Finally, a regression model using selected features with high coefficients was optimized and evaluated. Model performance was measured by classification accuracies and area under the receiver operating characteristic (AUROC).

RESULTS

We extracted 1223 radiomic features from each MRI study of 133 enrolled patients. After tumor excision, 34 (26 %) of 133 patients had pCR in resected specimens. When 25 image-derived features were selected from univariate analysis, classification AUROC was 0.86 and 0.79 with the addition of six clinical features on the hold-out internal validation dataset. When 11 image-derived features were used, the optimized linear regression model had an AUROC value of 0.79 and 0.65 with the addition of six clinical features on the hold-out dataset. Among the radiomic features, texture features including gray level variance, strength, and cluster prominence had the highest coefficient by Lasso regression.

CONCLUSION

Radiomic features derived from pretreatment MR images demonstrated promising efficacy in predicting pCR after CRT. However, radiomic features combined with clinical features did not result in remarkable improvement in model performance.

Radiomics in clinical trials: perspectives on standardization

The term biomarker is used to describe a biological measure of the disease behavior. The existing imaging biomarkers are associated with the known tissue biological characteristics and follow a well-established roadmap to be implemented in routine clinical practice. Recently, a new quantitative imaging analysis approach named radiomics has emerged. It refers to the extraction of a large number of advanced imaging features with high-throughput computing. Extensive research has demonstrated its value in predicting disease behavior, progression, and response to therapeutic options. However, there are numerous challenges to establishing it as a clinically viable solution, including lack of reproducibility and transparency. The data-driven nature also does not offer insights into the underpinning biology of the observed relationships. As such, additional effort is needed to establish it as a qualified biomarker to inform clinical decisions. Here we review the technical difficulties encountered in the clinical applications of radiomics and current effort in addressing some of these challenges in clinical trial designs. By addressing these challenges, the true potential of radiomics can be unleashed.

Repeatability and reproducibility of magnetic resonance imaging-based radiomic features in rectal cancer

Purpose: Radiomics of magnetic resonance images (MRIs) in rectal cancer can non-invasively characterize tumor heterogeneity with potential to discover new imaging biomarkers. However, for radiomics to be reliable, the imaging features measured must be stable and reproducible. The aim of this study is to quantify the repeatability and reproducibility of MRI-based radiomic features in rectal cancer. Approach: An MRI radiomics phantom was used to measure the longitudinal repeatability of radiomic features and the impact of post-processing changes related to image resolution and noise. Repeatability measurements in rectal cancers were also quantified in a cohort of 10 patients with test-retest imaging among two observers. Results: We found that many radiomic features, particularly from texture classes, were highly sensitive to changes in image resolution and noise. About 49% of features had coefficient of variations ≤ 10 % in longitudinal phantom measurements. About 75% of radiomic features in in vivo test-retest measurements had an intraclass correlation coefficient of ≥ 0.8 . We saw excellent interobserver agreement with mean Dice similarity coefficient of 0.95 ± 0.04 for test and retest scans. Conclusions: The results of this study show that even when using a consistent imaging protocol many radiomic features were unstable. Therefore, caution must be taken when selecting features for potential imaging biomarkers.

Are We There Yet? The Value of Deep Learning in a Multicenter Setting for Response Prediction of Locally Advanced Rectal Cancer to Neoadjuvant Chemoradiotherapy

This retrospective study aims to evaluate the generalizability of a promising state-of-the-art multitask deep learning (DL) model for predicting the response of locally advanced rectal cancer (LARC) to neoadjuvant chemoradiotherapy (nCRT) using a multicenter dataset. To this end, we retrained and validated a Siamese network with two U-Nets joined at multiple layers using pre- and post-therapeutic T2-weighted (T2w), diffusion-weighted (DW) images and apparent diffusion coefficient (ADC) maps of 83 LARC patients acquired under study conditions at four different medical centers. To assess the predictive performance of the model, the trained network was then applied to an external clinical routine dataset of 46 LARC patients imaged without study conditions. The training and test datasets differed significantly in terms of their composition, e.g., T-/N-staging, the time interval between initial staging/nCRT/re-staging and surgery, as well as with respect to acquisition parameters, such as resolution, echo/repetition time, flip angle and field strength. We found that even after dedicated data pre-processing, the predictive performance dropped significantly in this multicenter setting compared to a previously published single- or two-center setting. Testing the network on the external clinical routine dataset yielded an area under the receiver operating characteristic curve of 0.54 (95% confidence interval [CI]: 0.41, 0.65), when using only pre- and post-therapeutic T2w images as input, and 0.60 (95% CI: 0.48, 0.71), when using the combination of pre- and post-therapeutic T2w, DW images, and ADC maps as input. Our study highlights the importance of data quality and harmonization in clinical trials using machine learning. Only in a joint, cross-center effort, involving a multidisciplinary team can we generate large enough curated and annotated datasets and develop the necessary pre-processing pipelines for data harmonization to successfully apply DL models clinically.

Rectal MRI radiomics inter- and intra-reader reliability: should we worry about that?

PURPOSE

The aim of this review paper is to summarize the current literature regarding inter- and intra-reader reliability of radiomics on rectal MRI.

METHODS

Original studies examining treatment response prediction in patients with rectal cancer following neoadjuvant therapy using rectal MRI-based radiomics between January 2010 and December 2021 were identified via a PubMed/Medline search. Studies in which intra- and/or inter-reader reliability had been reported were included in this review.

RESULTS

Thirteen studies were selected, with an average number of patients of 145 (range, 20-649). All included studies evaluated T2-weighted imaging (T2WI) and/or diffusion-weighted imaging (DWI) sequences, while 3/13 (23%) also evaluated the contrast-enhanced T1-weighted imaging (T1WI) sequence. Most of the selected studies involved two readers (10/13, 77%), 6/13 (46%) studies used baseline MRI only, 1/13 (8%) study used restaging MRI only, and 6/13 (46%) used both. Segmentation was performed manually in 10/13 (77%) studies, and in a slight majority of studies (7/13, 54%), the entire tumor volume (3D VOI) was segmented, while 4/13 (31%) studies segmented the 2D ROI and 2/13 (15%) segmented both. Intraclass correlation coefficient (ICC) on intra-reader agreement varied from 0.73 to 0.93. ICC to assess inter-reader varied from 0.60 to 0.99. Overall, features obtained from baseline rectal MRI, using 3D VOI and first-order features, had higher agreement.

CONCLUSION

Based on our qualitative assessment of a small number of non-dedicated studies, there seems to be good reliability, particularly among low-order features extracted from the entire tumor volume using baseline MRI; however, direct evidence remains scarce. More targeted research in this area is required to quantitatively verify reliability, and before these novel radiomic techniques can be clinically adopted.

MRI radiomics independent of clinical baseline characteristics and neoadjuvant treatment modalities predicts response to neoadjuvant therapy in rectal cancer

Background

To analyse the performance of multicentre pre-treatment MRI-based radiomics (MBR) signatures combined with clinical baseline characteristics and neoadjuvant treatment modalities to predict complete response to neoadjuvant (chemo)radiotherapy in locally advanced rectal cancer (LARC).

Methods

Baseline MRI and clinical characteristics with neoadjuvant treatment modalities at four centres were collected. Decision tree, support vector machine and five-fold cross-validation were applied for two non-imaging and three radiomics-based models’ development and validation.

Results

We finally included 674 patients. Pre-treatment CEA, T stage, and histologic grade were selected to generate two non-imaging models: C model (clinical baseline characteristics alone) and CT model (clinical baseline characteristics combining neoadjuvant treatment modalities). The prediction performance of both non-imaging models were poor. The MBR signatures comprising 30 selected radiomics features, the MBR signatures combining clinical baseline characteristics (CMBR), and the CMBR incorporating neoadjuvant treatment modalities (CTMBR) all showed good discrimination with mean AUCs of 0.7835, 0.7871 and 0.7916 in validation sets, respectively. The three radiomics-based models had insignificant discrimination in performance.

Conclusions

The performance of the radiomics-based models were superior to the non-imaging models. MBR signatures seemed to reflect LARC’s true nature more accurately than clinical parameters and helped identify patients who can undergo organ preservation strategies.