The feasibility of MRI-based radiomics model in presurgical evaluation of tumor budding in locally advanced rectal cancer

MRI-based radiomic models for the preoperative prediction of extramural venous invasion in rectal cancer: A systematic review and meta-analysis

AIM

To estimate the diagnostic value of magnetic resonance imaging (MRI)-based radiomic models in detecting the extramural venous invasion (EMVI) of rectal cancer.

MATERIALS AND METHODS

Appropriate studies in multiple electronic databases were systematically retrieved. The Quality Assessment of Diagnostic Accuracy Studies 2 and Radiomics Quality Score (RQS) were used to evaluate the eligible studies' methodology quality. Summary accuracy metrics were calculated, and the publication bias was detected using Deek's funnel plot. The sensitivity and meta-regression analysis were performed to investigate the causes of heterogeneity.

RESULTS

For the seven eligible studies, which included 1175 patients, the pooled sensitivity, specificity, positive likelihood ratio, negative likelihood ratio, and diagnostic odds ratio were 0.80 (95 % CI, 0.70-0.88), 0.89 (95 % CI, 0.84-0.92), 7.0 (95 % CI, 4.7, 10.4), 0.22 (95 % CI, 0.14, 0.34), and 32 (95 % CI, 16, 65), respectively. The area under the receiver operating characteristic curve (AUC) was 0.91 (95 % CI, 0.88, 0.93). Moderate heterogeneity was found due to I2 values of 38.63 % and 32.29 % in sensitivity and specificity, respectively. Meta-regression analysis suggested that the patient enrollment, number of patients, segmentation method, and RQS score were the source of the heterogeneity. The head-to-head analysis suggested that radiomics model had a higher sensitivity for detection of EMVI than subjective evaluation by radiologist (0.47 vs. 0.73, p ≤ 0.001).

CONCLUSION

Our study suggests that MRI-based radiomic models have good diagnostic value in detecting EMVI for rectal cancer patients. Nevertheless, more prospective and high-quality studies with larger sample sizes are needed in the future to validate these results.

Preoperative Noninvasive Evaluation of Tumor Budding in Rectal Cancer Using Multiparameter MRI Radiomics

RATIONALE AND OBJECTIVES

To assess the value of a multiparametric magnetic resonance imaging (MRI)-based model integrating radiomics features with clinical and MRI semantic features for preoperative evaluation of tumor budding (TB) in rectal cancer.

MATERIALS AND METHODS

A total of 120 patients with pathologically confirmed rectal cancer were retrospectively analyzed. The patients were randomized into training and validation cohorts in a 6:4 ratio. Radiomics features were extracted and selected from preoperative T2-weighted imaging (T2WI), diffusion-weighted imaging (DWI), and contrast-enhanced T1-weighted imaging (T1CE) sequences, after which the corresponding radiomics score (RS) was calculated, and the radiomics models (T2WI model, DWI model, and T1CE model) were constructed. Logistic regression analysis was selected to develop a combined model integrated RST2WI, RSDWI, RST1CE, and clinical and MRI semantic features. The efficacy of each model in diagnosing TB grade was observed by the receiver operating characteristic (ROC) curve. Decision curve analysis (DCA) was used to assess the clinical benefits of the models.

RESULTS

Seven features were extracted and selected from each T2WI, DWI, and T1CE sequence to calculate the corresponding RS and construct the corresponding radiomics model. MRI reported N stage was an independent risk factor for TB. The area under the ROC curve of the combined model was 0.961 and 0.891 in the training and validation cohorts, respectively. The combined model showed better performance than the other models. DCA showed that the net benefit of the combined model was better than that of the other models in the vast majority of threshold probabilities.

CONCLUSION

A combined model integrating radiomics features and MRI semantic features allows for noninvasive preoperative evaluation of TB grading in patients with rectal cancer.

The radiomorphological appearance of the invasive margin in pancreatic cancer is associated with tumor budding

PURPOSE

Pancreatic cancer (PDAC) is characterized by infiltrative, spiculated tumor growth into the surrounding non-neoplastic tissue. Clinically, its diagnosis is often established by magnetic resonance imaging (MRI). At the invasive margin, tumor buds can be detected by histology, an established marker associated with poor prognosis in different types of tumors.

METHODS

We analyzed PDAC by determining the degree of tumor spiculation on T2-weighted MRI using a 3-tier grading system. The grade of spiculation was correlated with the density of tumor buds quantified in histological sections of the respective surgical specimen according to the guidelines of the International Tumor Budding Consensus Conference (n = 28 patients).

RESULTS

64% of tumors revealed intermediate to high spiculation on MRI. In over 90% of cases, tumor buds were detected. We observed a significant positive rank correlation between the grade of radiological tumor spiculation and the histopathological number of tumor buds (rs = 0.745, p < 0.001). The number of tumor buds was not significantly associated with tumor stage, presence of lymph node metastases, or histopathological grading (p ≥ 0.352).

CONCLUSION

Our study identifies a readily available radiological marker for non-invasive estimation of tumor budding, as a correlate for infiltrative tumor growth. This finding could help to identify PDAC patients who might benefit from more extensive peripancreatic soft tissue resection during surgery or stratify patients for personalized therapy concepts.

Predicting rectal cancer tumor budding grading based on MRI and CT with multimodal deep transfer learning: A dual-center study

Objective

To investigate the effectiveness of a multimodal deep learning model in predicting tumor budding (TB) grading in rectal cancer (RC) patients.

Materials and methods

A retrospective analysis was conducted on 355 patients with rectal adenocarcinoma from two different hospitals. Among them, 289 patients from our institution were randomly divided into an internal training cohort (n = 202) and an internal validation cohort (n = 87) in a 7:3 ratio, while an additional 66 patients from another hospital constituted an external validation cohort. Various deep learning models were constructed and compared for their performance using T1CE and CT-enhanced images, and the optimal models were selected for the creation of a multimodal fusion model. Based on single and multiple factor logistic regression, clinical N staging and fecal occult blood were identified as independent risk factors and used to construct the clinical model. A decision-level fusion was employed to integrate these two models to create an ensemble model. The predictive performance of each model was evaluated using the area under the curve (AUC), DeLong's test, calibration curve, and decision curve analysis (DCA). Model visualization Gradient-weighted Class Activation Mapping (Grad-CAM) was performed for model interpretation.

Results

The multimodal fusion model demonstrated superior performance compared to single-modal models, with AUC values of 0.869 (95% CI: 0.761-0.976) for the internal validation cohort and 0.848 (95% CI: 0.721-0.975) for the external validation cohort. N-stage and fecal occult blood were identified as clinically independent risk factors through single and multivariable logistic regression analysis. The final ensemble model exhibited the best performance, with AUC values of 0.898 (95% CI: 0.820-0.975) for the internal validation cohort and 0.868 (95% CI: 0.768-0.968) for the external validation cohort.

Conclusion

Multimodal deep learning models can effectively and non-invasively provide individualized predictions for TB grading in RC patients, offering valuable guidance for treatment selection and prognosis assessment.

Prognostic prediction value of the clinical-radiomics tumour-stroma ratio in locally advanced rectal cancer

PURPOSE

To develop and validate a radiomics model based on high-resolution T2WI and a clinical-radiomics model for tumour-stroma ratio (TSR) evaluation with a gold standard of TSR evaluated by rectal specimens without therapeutic interference and further apply them in prognosis prediction of locally advanced rectal cancer (LARC) patients who received neoadjuvant chemoradiotherapy.

METHODS

A total of 178 patients (mean age: 59.35, range 20-85 years; 65 women and 113 men) with rectal cancer who received surgery alone from January 2016 to October 2020 were enrolled and randomly separated at a ratio of 7:3 into training and validation sets. A senior radiologist reviewed after 2 readers manually delineated the whole tumour in consensus on preoperative high-resolution T2WI in the training set. A total of 1046 features were then extracted, and recursive feature elimination embedded with leave-one-out cross validation was applied to select features, with which an MR-TSR evaluation model was built containing 6 filtered features via a support vector machine classifier trained by comparing patients' pathological TSR. Stepwise logistic regression was employed to integrate clinical factors with the radiomics model (Fusion-TSR) in the training set. Later, the MR-TSR and Fusion-TSR models were replicated in the validation set for diagnostic effectiveness evaluation. Subsequently, 243 patients (mean age: 53.74, range 23-74 years; 63 women and 180 men) with LARC from October 2012 to September 2017 who were treated with NCRT prior to surgery and underwent standard pretreatment rectal MR examination were enrolled. The MR-TSR and Fusion-TSR were applied, and the Kaplan-Meier method and log-rank test were used to compare the survival of patients with different MR-TSR and Fusion-TSR. Cox proportional hazards regression was used to calculate the hazard ratio (HR).

RESULTS

Both the MR-TSR and Fusion-TSR models were validated with favourable diagnostic power: the AUC of the MR-TSR was 0.77 (p = 0.01; accuracy = 69.8 %, sensitivity = 88.9 %, specificity = 65.9 %, PPV = 34.8 %, NPV = 96.7 %), while the AUC of the Fusion-TSR was 0.76 (p = 0.014; accuracy = 67.9 %, sensitivity = 88.9 %, specificity = 63.6 %, PPV = 33.3 %, NPV = 96.6 %), outperforming their effectiveness in the training set: the AUC of the MR-TSR was 0.65 (p = 0.035; accuracy = 66.4 %, sensitivity = 61.9 %, specificity = 67.3 %, PPV = 27.7 %, NPV = 90.0 %), while the AUC of the Fusion-TSR was 0.73 (p = 0.001; accuracy = 73.6 %, sensitivity = 71.4 %, specificity = 74.0 %, PPV = 35.73 %, NPV = 92.8 %). With further prognostic analysis, the MR-TSR was validated as a significant prognostic factor for DFS in LARC patients treated with NCRT (p = 0.020, HR = 1.662, 95 % CI = 1.077-2.565), while the Fusion-TSR was a significant prognostic factor for OS (p = 0.005, HR = 2.373, 95 % CI = 1.281-4.396).

CONCLUSIONS

We developed and validated a radiomics TSR and a clinical-radiomics TSR model and successfully applied them to better risk stratification for LARC patients receiving NCRT and for better decision making.

Preoperative prediction of tumor budding in rectal cancer using multiple machine learning algorithms based on MRI T2WI radiomics

Objective

This study aimed to explore the radiomics model based on magnetic resonance imaging (MRI) T2WI and compare the value of different machine algorithms in preoperatively predicting tumor budding (TB) grading in rectal cancer.

Methods

A retrospective study was conducted on 266 patients with preoperative rectal MRI examinations, who underwent complete surgical resection and confirmed pathological diagnosis of rectal cancer. Among them, patients from Qingdao West Coast Hospital were assigned as the training group (n=172), while patients from other hospitals were assigned as the external validation group (n=94). Regions of interest (ROIs) were delineated, and image features were extracted and dimensionally reduced using the Least Absolute Shrinkage and Selection Operator (LASSO). Eight machine algorithms were used to construct the models, and the diagnostic performance of the models was evaluated and compared using receiver operating characteristic (ROC) curves and the area under the curve (AUC), as well as clinical utility assessment using decision curve analysis (DCA).

Results

A total of 1197 features were extracted, and after feature selection and dimension reduction, 11 image features related to TB grading were obtained. Among the eight algorithm models, the support vector machine (SVM) algorithm achieved the best diagnostic performance, with accuracy, sensitivity, and specificity of 0.826, 0.949, and 0.723 in the training group, and 0.713, 0.579, and 0.804 in the validation group, respectively. DCA demonstrated the clinical utility of this radiomics model.

Conclusion

The radiomics model based on MR T2WI can provide an effective and noninvasive method for preoperative TB grading assessment in patients with rectal cancer.

Artificial intelligence and radiomics in magnetic resonance imaging of rectal cancer: a review

Rectal cancer (RC) is one of the most common tumours worldwide in both males and females, with significant morbidity and mortality rates, and it accounts for approximately one-third of colorectal cancers (CRCs). Magnetic resonance imaging (MRI) has been demonstrated to be accurate in evaluating the tumour location and stage, mucin content, invasion depth, lymph node (LN) metastasis, extramural vascular invasion (EMVI), and involvement of the mesorectal fascia (MRF). However, these features alone remain insufficient to precisely guide treatment decisions. Therefore, new imaging biomarkers are necessary to define tumour characteristics for staging and restaging patients with RC. During the last decades, RC evaluation via MRI-based radiomics and artificial intelligence (AI) tools has been a research hotspot. The aim of this review was to summarise the achievement of MRI-based radiomics and AI for the evaluation of staging, response to therapy, genotyping, prediction of high-risk factors, and prognosis in the field of RC. Moreover, future challenges and limitations of these tools that need to be solved to favour the transition from academic research to the clinical setting will be discussed.

Radiomics nomogram for the prediction of Ki-67 index in advanced non-small cell lung cancer based on dual-phase enhanced computed tomography

PURPOSE

To develop a radiomics nomogram based on dual-phase enhanced computed tomography (CT) for predicting the Ki-67 index status in patients with advanced non-small cell lung cancer (NSCLC).

METHODS

137 patients with NSCLC who had undergone dual-phase enhanced CT scans and Ki-67 examination within 2 weeks were retrospectively enrolled between January 2020 and December 2022. Clinical and laboratory data were collected, and the patients were categorized based on low or high expression of Ki-67 index, with a cut-off value of 40%. The cohort was randomly divided into a training group (n = 95) and a testing group (n = 42) at a ratio of 7:3. The least absolute shrinkage and selection operator (LASSO) algorithm was employed to select the most valuable radiomics features from the dual-phase enhanced CT images. Subsequently, a nomogram that incorporated the radiomics score and clinical factors associated with Ki-67 index status was established through univariate and multivariate logistic regression analyses. The predictive performance of the nomogram was evaluated using the area under the curve (AUC).

RESULTS

The AUC values of the radiomics features of artery phase and vein phase CT in the testing group were 0.748 and 0.758, respectively. The AUC of the dual-phase enhanced CT was 0.785, and the AUC of the developed nomogram was 0.859, which was higher than those of the radiomics (AUC, 0.785) and clinical models (AUC, 0.736).

CONCLUSIONS

The radiomics nomogram based on dual-phase enhanced CT images provides a promising method for predicting the Ki-67 index status in patients with advanced NSCLC.

Normalization Strategies in Multi-Center Radiomics Abdominal MRI: Systematic Review and Meta-Analyses

Goal: Artificial intelligence applied to medical image analysis has been extensively used to develop non-invasive diagnostic and prognostic signatures. However, these imaging biomarkers should be largely validated on multi-center datasets to prove their robustness before they can be introduced into clinical practice. The main challenge is represented by the great and unavoidable image variability which is usually addressed using different pre-processing techniques including spatial, intensity and feature normalization. The purpose of this study is to systematically summarize normalization methods and to evaluate their correlation with the radiomics model performances through meta-analyses. This review is carried out according to the PRISMA statement: 4777 papers were collected, but only 74 were included. Two meta-analyses were carried out according to two clinical aims: characterization and prediction of response. Findings of this review demonstrated that there are some commonly used normalization approaches, but not a commonly agreed pipeline that can allow to improve performance and to bridge the gap between bench and bedside.

MRI-Based Artificial Intelligence in Rectal Cancer

Rectal cancer (RC) accounts for approximately one-third of colorectal cancer (CRC), with death rates increasing in patients younger than 50 years old. Magnetic resonance imaging (MRI) is routinely performed for tumor evaluation. However, the semantic features from images alone remain insufficient to guide treatment decisions. Functional MRIs are useful for revealing microstructural and functional abnormalities and nevertheless have low or modest repeatability and reproducibility. Therefore, during the preoperative evaluation and follow-up treatment of patients with RC, novel noninvasive imaging markers are needed to describe tumor characteristics to guide treatment strategies and achieve individualized diagnosis and treatment. In recent years, the development of artificial intelligence (AI) has created new tools for RC evaluation based on MRI. In this review, we summarize the research progress of AI in the evaluation of staging, prediction of high-risk factors, genotyping, response to therapy, recurrence, metastasis, prognosis, and segmentation with RC. We further discuss the challenges of clinical application, including improvement in imaging, model performance, and the biological meaning of features, which may also be major development directions in the future. EVIDENCE LEVEL: 5 TECHNICAL EFFICACY: Stage 2.

Role of radiomics in the diagnosis and treatment of gastrointestinal cancer

Gastrointestinal cancer (GIC) has high morbidity and mortality as one of the main causes of cancer death. Preoperative risk stratification is critical to guide patient management, but traditional imaging studies have difficulty predicting its biological behavior. The emerging field of radiomics allows the conversion of potential pathophysiological information in existing medical images that cannot be visually recognized into high-dimensional quantitative image features. Tumor lesion characterization, therapeutic response evaluation, and survival prediction can be achieved by analyzing the relationships between these features and clinical and genetic data. In recent years, the clinical application of radiomics to GIC has increased dramatically. In this editorial, we describe the latest progress in the application of radiomics to GIC and discuss the value of its potential clinical applications, as well as its limitations and future directions.

Deep-learning-based 3D super-resolution MRI radiomics model: superior predictive performance in preoperative T-staging of rectal cancer

OBJECTIVES

To investigate the feasibility and efficacy of a deep-learning (DL)-based three-dimensional (3D) super-resolution (SR) MRI radiomics model for preoperative T-staging prediction in rectal cancer (RC).

METHODS

Seven hundred six eligible RC patients (T1/2 = 287, T3/4 = 419) were retrospectively enrolled in this study and chronologically allocated into a training cohort (n = 565) and a validation cohort (n = 141). We conducted a deep-transfer-learning network on high-resolution (HR) T2-weighted imaging (T2WI) to enhance the z-resolution of the images and acquired the preoperative SRT2WI. The radiomics models named model_HRT2 and model_SRT2 were respectively constructed with high-dimensional quantitative features extracted from manually segmented volume of interests of HRT2WI and SRT2WI through the Least Absolute Shrinkage and Selection Operator method. The performances of the models were evaluated by ROC, calibration, and decision curves.

RESULTS

Model_SRT2 outperformed model_HRT2 (AUC 0.869, sensitivity 71.1%, specificity 93.1%, and accuracy 83.3% vs. AUC 0.810, sensitivity 89.5%, specificity 70.1%, and accuracy 77.3%) in distinguishing T1/2 and T3/4 RC with significant difference (p < 0.05). Both radiomics models achieved higher AUCs than the expert radiologists (0.685, 95% confidence interval 0.595-0.775, p < 0.05). The calibration curves confirmed high goodness of fit, and the decision curve analysis revealed the clinical value.

CONCLUSIONS

Model_SRT2 yielded superior predictive performance in preoperative RC T-staging by comparison with model_HRT2 and expert radiologists' visual assessments.

KEY POINTS

• For the first time, DL-based 3D SR images were applied in radiomics analysis for clinical utility. • Compared with the visual assessment of expert radiologists and the conventional radiomics model based on HRT2WI, the SR radiomics model showed a more favorable capability in helping clinicians assess the invasion depth of RC preoperatively. • This is the largest radiomics study for T-staging prediction in RC.

Radiomics Based on Digital Mammography Helps to Identify Mammographic Masses Suspicious for Cancer

Objectives

This study aims to build radiomics model of Breast Imaging Reporting and Data System (BI-RADS) category 4 and 5 mammographic masses extracted from digital mammography (DM) for mammographic masses characterization by using a sensitivity threshold similar to that of biopsy.

Materials and Methods

This retrospective study included 288 female patients (age, 52.41 ± 10.31) who had BI-RADS category 4 or 5 mammographic masses with an indication for biopsy. The patients were divided into two temporal set (training set, 82 malignancies and 110 benign lesions; independent test set, 48 malignancies and 48 benign lesions). A total of 188 radiomics features were extracted from mammographic masses on the combination of craniocaudal (CC) position images and mediolateral oblique (MLO) position images. For the training set, Pearson’s correlation and the least absolute shrinkage and selection operator (LASSO) were used to select non-redundant radiomics features and useful radiomics features, respectively, and support vector machine (SVM) was applied to construct a radiomics model. The receiver operating characteristic curve (ROC) analysis was used to evaluate the classification performance of the radiomics model and to determine a threshold value with a sensitivity higher than 98% to predict the mammographic masses malignancy. For independent test set, identical threshold value was used to validate the classification performance of the radiomics model. The stability of the radiomics model was evaluated by using a fivefold cross-validation method, and two breast radiologists assessed the diagnostic agreement of the radiomics model.

Results

In the training set, the radiomics model obtained an area under the receiver operating characteristic curve (AUC) of 0.934 [95% confidence intervals (95% CI), 0.898–0.971], a sensitivity of 98.8% (81/82), a threshold of 0.22, and a specificity of 60% (66/110). In the test set, the radiomics model obtained an AUC of 0.901 (95% CI, 0.835–0.961), a sensitivity of 95.8% (46/48), and a specificity of 66.7% (32/48). The radiomics model had relatively stable sensitivities in fivefold cross-validation (training set, 97.39% ± 3.9%; test set, 98.7% ± 4%).

Conclusion

The radiomics method based on DM may help reduce the temporarily unnecessary invasive biopsies for benign mammographic masses over-classified in BI-RADS category 4 and 5 while providing similar diagnostic performance for malignant mammographic masses as biopsies.