Feasibility of an ADC-based radiomics model for predicting pelvic lymph node metastases in patients with stage IB-IIA cervical squamous cell carcinoma

Radiomics analysis based on multiparametric magnetic resonance imaging for differentiating early stage of cervical cancer

Objective

To investigate the performance of multiparametric magnetic resonance imaging (MRI)-based radiomics models in differentiating early stage of cervical cancer (Stage I-IIa vs. IIb-IV).

Methods

One hundred patients with cervical cancer who underwent preoperative MRI between June 2020 and March 2022 were retrospectively enrolled. Training (n = 70) and testing cohorts (n = 30) were assigned by stratified random sampling. The clinical and pathological features, including age, histological subtypes, tumor grades, and node status, were compared between the two cohorts by t-test or chi-square test. Radiomics features were extracted from each volume of interest (VOI) on T2-weighted images (T2WI) and apparent diffusion coefficient (ADC) maps. The data balance of the training cohort was resampled by synthesizing minority oversampling techniques. Subsequently, the adiomics signatures were constructed by the least absolute shrinkage and selection operator algorithm and minimum-redundancy maximum-relevance with 10-fold cross-validation. Logistic regression was applied to predict the cervical cancer stages (low [I-IIa]) and (high [IIb-IV] FIGO stages). The receiver operating characteristic curve (area under the curve [AUC]) and decision curve analysis were used to assess the performance of the radiomics model.

Results

The characteristics of age, histological subtypes, tumor grades, and node status were not significantly different between the low [I-IIa] and high [IIb-IV] FIGO stages (p > 0.05 for both the training and test cohorts). Three models based on T2WI, ADC maps, and the combined were developed based on six radiomics features from T2WI and three radiomics features from ADC maps, with AUCs of 0.855 (95% confidence interval [CI], 0.777-0.934) and 0.823 (95% CI, 0.727-0.919), 0.861 (95% CI, 0.785-0.936) and 0.81 (95% CI, 0.701-0.918), 0.934 (95% CI, 0.884-0.984) and 0.902 (95% CI, 0.832-0.972) in the training and test cohorts.

Conclusion

The radiomics models combined T2W and ADC maps had good predictive performance in differentiating the early stage from locally advanced cervical cancer.

A contrast-enhanced CT radiomics-based model to identify candidates for deintensified chemoradiotherapy in locoregionally advanced nasopharyngeal carcinoma patients

OBJECTIVES

To develop a contrast-enhanced CT (CECT) radiomics-based model to identify locoregionally advanced nasopharyngeal carcinoma (LA-NPC) patients who would benefit from deintensified chemoradiotherapy.

METHODS

LA-NPC patients who received low-dose concurrent cisplatin therapy (cumulative: 150 mg/m2), were randomly divided into training and validation groups. 107 radiomics features based on the primary nasopharyngeal tumor were extracted from each pre-treatment CECT scan. Through Cox regression analysis, a radiomics model and patients' corresponding radiomics scores were created with predictive independent radiomics features. T stage (T) and radiomics score (R) were compared as predictive factors. Combining the N stage (N), a clinical model (T + N), and a substitution model (R + N) were constructed.

RESULTS

Training and validation groups consisted of 66 and 33 patients, respectively. Three significant independent radiomics features (flatness, mean, and gray level non-uniformity in gray level dependence matrix (GLDM-GLN)) were found. The radiomics score showed better predictive ability than the T stage (concordance index (C-index): 0.67 vs. 0.61, AUC: 0.75 vs. 0.60). The R + N model had better predictive performance and more effective risk stratification than the T + N model (C-index: 0.77 vs. 0.68, AUC: 0.80 vs. 0.70). The R + N model identified a low-risk group as deintensified chemoradiotherapy candidates in which no patient developed progression within 3 years, with 5-year progression-free survival (PFS) and overall survival (OS) both 90.7% (hazard ratio (HR) = 4.132, p = 0.018).

CONCLUSION

Our radiomics-based model combining radiomics score and N stage can identify specific LA-NPC candidates for whom de-escalation therapy can be performed without compromising therapeutic efficacy.

CLINICAL RELEVANCE STATEMENT

Our study shows that the radiomics-based model (R + N) can accurately stratify patients into different risk groups, with satisfactory prognosis in the low-risk group when treated with low-dose concurrent chemotherapy, providing new options for individualized de-escalation strategies.

KEY POINTS

• A radiomics score, consisting of 3 predictive radiomics features (flatness, mean, and GLDM-GLN) integrated with the N stage, can identify specific LA-NPC populations for deintensified treatment. • In the selection of LA-NPC candidates for de-intensified treatment, radiomics score extracted from primary nasopharyngeal tumors based on CECT can be superior to traditional T stage classification as a predictor.

Radiomics systematic review in cervical cancer: gynecological oncologists' perspective

OBJECTIVE

Radiomics is the process of extracting quantitative features from radiological images, and represents a relatively new field in gynecological cancers. Cervical cancer has been the most studied gynecological tumor for what concerns radiomics analysis. The aim of this study was to report on the clinical applications of radiomics combined and/or compared with clinical-pathological variables in patients with cervical cancer.

METHODS

A systematic review of the literature from inception to February 2023 was performed, including studies on cervical cancer analysing a predictive/prognostic radiomics model, which was combined and/or compared with a radiological or a clinical-pathological model.

RESULTS

A total of 57 of 334 (17.1%) screened studies met inclusion criteria. The majority of studies used magnetic resonance imaging (MRI), but positron emission tomography (PET)/computed tomography (CT) scan, CT scan, and ultrasound scan also underwent radiomics analysis. In apparent early-stage disease, the majority of studies (16/27, 59.3%) analysed the role of radiomics signature in predicting lymph node metastasis; six (22.2%) investigated the prediction of radiomics to detect lymphovascular space involvement, one (3.7%) investigated depth of stromal infiltration, and one investigated (3.7%) parametrial infiltration. Survival prediction was evaluated both in early-stage and locally advanced settings. No study focused on the application of radiomics in metastatic or recurrent disease.

CONCLUSION

Radiomics signatures were predictive of pathological and oncological outcomes, particularly if combined with clinical variables. These may be integrated in a model using different clinical-pathological and translational characteristics, with the aim to tailor and personalize the treatment of each patient with cervical cancer.

Intra- and peri-tumoral MRI radiomics features for preoperative lymph node metastasis prediction in early-stage cervical cancer

BACKGROUND

Noninvasive and accurate prediction of lymph node metastasis (LNM) is very important for patients with early-stage cervical cancer (ECC). Our study aimed to investigate the accuracy and sensitivity of radiomics models with features extracted from both intra- and peritumoral regions in magnetic resonance imaging (MRI) with T2 weighted imaging (T2WI) and diffusion weighted imaging (DWI) for predicting LNM.

METHODS

A total of 247 ECC patients with confirmed lymph node status were enrolled retrospectively and randomly divided into training (n = 172) and testing sets (n = 75). Radiomics features were extracted from both intra- and peritumoral regions with different expansion dimensions (3, 5, and 7 mm) in T2WI and DWI. Radiomics signature and combined radiomics models were constructed with selected features. A nomogram was also constructed by combining radiomics model with clinical factors for predicting LNM.

RESULTS

The area under curves (AUCs) of radiomics signature with features from tumors in T2WI and DWI were 0.841 vs. 0.791 and 0.820 vs. 0.771 in the training and testing sets, respectively. Combining radiomics features from tumors in the T2WI, DWI and peritumoral 3 mm expansion in T2WI achieved the best performance with an AUC of 0.868 and 0.846 in the training and testing sets, respectively. A nomogram combining age and maximum tumor diameter (MTD) with radiomics signature achieved a C-index of 0.884 in the prediction of LNM for ECC.

CONCLUSIONS

 Radiomics features extracted from both intra- and peritumoral regions in T2WI and DWI are feasible and promising for the preoperative prediction of LNM for patients with ECC.

Prediction Model of Residual Neural Network for Pathological Confirmed Lymph Node Metastasis of Ovarian Cancer

Purpose. We want to develop a model for predicting lymph node status based on positron emission computed tomography (PET) images of untreated ovarian cancer patients. We use the feature map formed by wavelet transform and the parameters obtained by image segmentation to build the model. The model is expected to help clinicians and provide additional information about what to do with first-visit patients. Materials and Methods. Our study included 224 patients with ovarian cancer. We have chosen two main methods to extract information from images. On the one hand, we segmented the image to extract the parameters to evaluate the clustering effect. On the other hand, we used wavelet transform to extract the image’s texture information to form the image’s feature map. Based on the above two kinds of information, we used residual neural network and support vector machine for modeling. Results. We established a model to predict lymph node metastasis in patients with primary ovarian cancer using PET images. On the training set, our accuracy was 0.8854, AUC: 0.9472, CI: 0.9098-0.9752, sensitivity was 0.9865, and specificity was 0.7952. On the test set, our accuracy was 0.9104, AUC: 0.9259, CI: 0.8417-0.9889, sensitivity was 0.8125, and specificity was 1.0000. Conclusions. We used wavelet transform to process the preoperative medical images of ovarian cancer patients, and the residual neural network can effectively predict the lymph node metastasis of ovarian cancer patients, which is undoubted of great significance for patients’ staging and treatment options.

Radiomics Based on Nomogram Predict Pelvic Lymphnode Metastasis in Early-Stage Cervical Cancer

The accurate prediction of the status of PLNM preoperatively plays a key role in treatment strategy decisions in early-stage cervical cancer. The aim of this study was to develop and validate a radiomics-based nomogram for the preoperative prediction of pelvic lymph node metastatic status in early-stage cervical cancer. One hundred fifty patients were enrolled in this study. Radiomics features were extracted from T2-weighted MRI imaging (T2WI). Based on the selected features, a support vector machine (SVM) algorithm was used to build the radiomics signature. The radiomics-based nomogram was developed incorporating radiomics signature and clinical risk factors. In the training cohort (AUC = 0.925, accuracy = 81.6%, sensitivity = 70.3%, and specificity = 92.0%) and the testing cohort (AUC = 0.839, accuracy = 74.2%, sensitivity = 65.7%, and specificity = 82.8%), clinical models that combine stromal invasion depth, FIGO stage, and MTD perform poorly. The combined model had the highest AUC in the training cohort (AUC = 0.988, accuracy = 95.9%, sensitivity = 92.0%, and specificity = 100.0%) and the testing cohort (AUC = 0.922, accuracy = 87.1%, sensitivity = 85.7%, and specificity = 88.6%) when compared to the radiomics and clinical models. The study may provide valuable guidance for clinical physicians regarding the treatment strategies for early-stage cervical cancer patients.

MRI outcome evaluation in patients with IB2 and IIA2 squamous cervical cancer stages: preliminary results

Objectives

To evaluate the therapeutic effect of neoadjuvant therapy (NAT) followed by radical hysterectomy and concurrent chemoradiotherapy (CCRT) in stage IB2 and IIA2 squamous cervical cancer (SCC) and investigate the value of apparent diffusion coefficient (ADC) in outcome evaluation of different treatment strategies in the patients.

Methods

A total of 149 patients with IB2 and IIA2 SCC who underwent pretreatment MRI and DWI scan were included. Patients were treated with NAT + RH or CCRT. Clinical indices and pathological factors were recorded. The imaging indices were measured including tumor size and tumor ADC values. Intraclass correlation coefficient was employed to evaluate the consistency of the indices measured by two observers. ROC curves were used to evaluate the cutoff values of clinical and imaging indices. Kaplan–Meier and Cox proportional hazard model were used to analyze the independent factors of disease-free survival (DFS).

Results

The median follow-up period was 42.3 months. SCC-Ag, ADCmax and ADCmin were independent factors for DFS in the entire cohort. SCC-Ag, ADCmin and vascular invasion were independent factors for DFS in NAT + RH group. ADCmax and ADCmin were independent factors for DFS in CCRT group. ADCmin was the strongest independent factor for DFS in NAT + RH group, while ADCmax was that in CCRT group.

Conclusion

The NAT + RH patients had similar DFS to that of CCRT in IB2 and IIA2 SCC, which could be a potential feasible alternative treatment. ADCmin and ADCmax were more valuable in evaluating the outcome of patients who underwent NAT + RH or CCRT, respectively.

Prognostic Significance of Pseudotime from Texture Parameters of FDG PET/CT in Locally Advanced Non-Small-Cell Lung Cancer with Tri-Modality Therapy

Simple Summary

Although texture parameters of F-18 fluorodeoxyglucose positron emission tomography/computed tomography images were known to associate tumor biology and clinical features, the types and implications of parameters are too various and complicated. To overcome the limitation of texture parameter, we attempted to produce a new simplified parameter from texture parameters of F-18 fluorodeoxyglucose positron emission tomography/computed tomography images in lung cancer patients using pseudotime analysis. Pseudotime analysis is a recently developed method to explore changes in cell or tissue characteristics based on transcriptomic expression. It is the first study to apply pseudotime analysis into radiomics dataset other than transcriptomics data. Herein, we demonstrated that pseudotime can be successfully estimated from texture parameters. In the aspect of prognostic prediction, pseudotime was an independent prognostic factor for overall survival in contrast to conventional parameters such as metabolic tumor volume and total lesion glycolysis. This study showed possibility of integrating various texture parameters into single parameter which reflects disease progression status. Pseudotime, as a concrete value of disease progression, is expected to be used in clinical field to evaluate disease and predict prognosis.

Abstract

Texture analysis provides image parameters from F-18 fluorodeoxyglucose positron emission tomography/computed tomography (FDG PET/CT). Although some parameters are associated with tumor biology and clinical features, the types and implications of these parameters are complicated. We applied pseudotime analysis, which has recently been used to estimate changes in individual sample characteristics, to texture parameters from FDG PET/CT images of locally advanced non-small-cell lung cancer (NSCLC) patients undergoing neoadjuvant concurrent chemoradiation therapy (CCRT) followed by surgery. Our subjects were 303 NSCLC patients who underwent pretherapeutic FDG PET/CT and tri-modality therapy. Texture parameters of the primary tumor were calculated from FDG PET/CT images acquired before neoadjuvant CCRT. Pseudotime analysis was performed using the PhenoPath tool. Clinicopathologic features including survival data were collected and survival analysis was performed to compare the prognostic significances of pseudotime parameters with those of conventional PET parameters. Pseudotime was successfully estimated from texture parameters. Normalized co-occurrence homogeneity, normalized co-occurrence inverse difference moment, and black–white symmetry showed positive correlations with pseudotime, short run emphasis, normalized co-occurrence dissimilarity, and short zone emphasis negative correlation. The maximum standardized uptake value (SUV) and mean SUV were not associated with overall survival. Pseudotime, metabolic tumor volume (MTV), and total lesion glycolysis (TLG) showed significant associations with overall survival. In contrast to MTV and TLG, pseudotime was an independent prognostic factor for overall survival. Various metabolic texture parameters can be integrated into a single parameter using pseudotime analysis. Pseudotime of the primary tumor, estimated from FDG PET/CT images, better predicts overall survival in locally advanced NSCLC patients treated with tri-modality therapy than conventional PET parameters.

Feasibility of Predicting Pelvic Lymph Node Metastasis Based on IVIM-DWI and Texture Parameters of the Primary Lesion and Lymph Nodes in Patients with Cervical Cancer

RATIONALE AND OBJECTIVES

To investigate the feasibility and value of intravoxel incoherent motion diffusion weighted imaging (IVIM-DWI) and texture parameters of primary lesions and lymph nodes for predicting pelvic lymph node metastasis in patients with cervical cancer.

MATERIALS AND METHODS

A total of 143 patients with cervical cancer confirmed by surgical pathology were analyzed retrospectively and 125 patients were enrolled in primary lesions study, 83 patients and 134 lymph nodes were enrolled in lymph nodes study. Patients and lymph nodes were randomly divided into training group and test group at a ratio of 2: 1. The IVIM-DWI parameters and 3D texture features of primary lesions and lymph nodes of all patients were measured. The least absolute shrinkage and selection operator algorithm, spearman's correlation analysis, independent two-sample t-test and Mann-Whitney U-test were used to select texture parameters. Multivariate Logistic regression analysis and receiver operating characteristic curves were used to model and evaluate diagnostic performances.

RESULTS

In primary lesions study, model 1 was constructed by combining f value, original_shape_Sphericity and original_firstorder_Mean of primary lesions. In lymph nodes study, model 2 was constructed by combining short diameter, circular enhancement and rough margin of lymph nodes. Model 3 was constructed by combining ADC, f value and original_glszm_Small Area Emphasis of lymph nodes. The areas under curve of model 1, 2 and 3 in training group and test group were 0.882, 0.798, 0.907 and 0.862, 0.771, 0.937 respectively.

CONCLUSION

Models based on IVIM-DWI and texture parameters of primary lesions and lymph nodes both performed well in diagnosing pelvic lymph node metastasis of cervical cancer and were superior to morphological features of lymph nodes. Especially, parameters of lymph nodes showed higher diagnostic efficiency and clinical significance.

Classifying early stages of cervical cancer with MRI-based radiomics

This study aims to establish a MRI-based classifier to distinguish early stages of cervical cancer with improved diagnostic performance to assist clinical diagnosis and treatment. 57 patients with pathological diagnosis of cervical cancer from January 2018 to May 2019 were enrolled in this study. MRI examinations, including T1-weighted image(T1WI), T2-weighted image(T2W), diffusion weighted imaging (DWI) and dynamic contrast enhanced (DCE), were performed before surgery. MR images from patients of stage Ib or IIa cervical cancer with tumor segmented were used as input. Feature extraction process extracted first-order statistics and texture and applied filters. The dimensionality of the radiomic features was reduced using the least absolute shrinkage and selection operator (LASSO). Models were trained by three machine-learning (k-nearest neighbor (KNN), support vector machine (SVM), and logistic regression (LR)) and diagnostic performance in differentiating stage Ib and stage IIa cases was evaluated. A total of 27 features were extracted to establish models, including 2 features from T1WI, 5 features from T2WI, 5 features from DWI (b = 50), 4 features from DWI (b = 800), 5 features from DCE, and 6 features from ADC. For each machine learning (ML) classifier, six sequences of training set and testing set are modeled and analyzed. Among all the models, the training set and testing set of T2WI model built by SVM classifier were the best (Area under the curve (AUC) 0.915) / (AUC 0.907). Radiomic analysis of ML-based texture features and first-order statistics features can be used to stage the early cervical cancer pre-operatively.

Artificial Intelligence in Cervical Cancer Screening and Diagnosis

Cervical cancer remains a leading cause of cancer death in women, seriously threatening their physical and mental health. It is an easily preventable cancer with early screening and diagnosis. Although technical advancements have significantly improved the early diagnosis of cervical cancer, accurate diagnosis remains difficult owing to various factors. In recent years, artificial intelligence (AI)-based medical diagnostic applications have been on the rise and have excellent applicability in the screening and diagnosis of cervical cancer. Their benefits include reduced time consumption, reduced need for professional and technical personnel, and no bias owing to subjective factors. We, thus, aimed to discuss how AI can be used in cervical cancer screening and diagnosis, particularly to improve the accuracy of early diagnosis. The application and challenges of using AI in the diagnosis and treatment of cervical cancer are also discussed.

Correlation between Tissue Cellularity and Metabolism Represented by Diffusion-Weighted Imaging (DWI) and 18F-FDG PET/MRI in Head and Neck Cancer (HNC)

Simple Summary

We report on the correlation between the diffusion-weighted imaging (DWI) and the metabolic volume parameters derived from a PET scan, to determine the correlation between these parameters and the tumor cellularity in head and neck primary tumors. Our findings implied that there was no correlation between the information derived from the DWI and the information derived from the FDG metabolic parameters. Thus, both imaging techniques might play a complementary role in HNC diagnosis and assessment. This is significant because the treatment plan of patients with HNC should be well evaluated by using all the available diagnosis techniques, for a better understanding of how the tumor will react.

Abstract

Background: This study aimed to assess the association of 18F-Fluorodeoxyglucose positron-emission-tomography (18F-FDG/PET) and DWI imaging parameters from a primary tumor and their correlations with clinicopathological factors. Methods: We retrospectively analyzed primary tumors in 71 patients with proven HNC. Primary tumor radiological parameters: DWI and FDG, as well as pathological characteristics were analyzed. Spearman correlation coefficient was used to assess the correlation between DWI and FDG parameters, ANOVA or Kruskal–Wallis, independent sample t-test, Mann–Whitney test, and multiple regression were performed on the clinicopathological features that may affect the 18F- FDG and apparent-diffusion coefficient (ADC) of the tumor. Results: No significant correlations were observed between DWI and any of the 18F-FDG parameters (p > 0.05). SUVmax correlated with N-stages (p = 0.023), TLG and MTV correlated with T-stages (p = 0.006 and p = 0.001), and ADC correlated with tumor grades (p = 0.05). SUVmax was able to differentiate between N+ and N− groups (p = 0.004). Conclusions: Our results revealed a non-significant correlation between the FDG-PET and ADC-MR parameters. FDG-PET-based glucose metabolic and DWI-MR-derived cellularity data may represent different biological aspects of HNC.

Combination of Estrogen Receptor Alpha and Histological Type Helps to Predict Lymph Node Metastasis in Patients with Stage IA2 to IIA2 Cervical Cancer

Objective

This study aimed to identify a subset of patients with stage IA2 to IIA2 cervical cancer who are at low risk of lymph node metastasis (LNM) using pathological parameters including estrogen receptor alpha (ERα) and progesterone receptor (PR).

Methods

The clinical data of patients with stage IA2 to IIA2 cervical cancer who underwent radical surgery between 2014 and 2015 were retrospectively reviewed. Immunohistochemical staining was used to determine the expression of ERα and PR. A low-risk criterion for LNM was identified using logistic regression analysis, and its performance was estimated through receiver-operating characteristic curve analysis.

Results

Of 263 patients, 57 (21.7%) had pathological LNM. ERα (adjusted odds ratio [aOR], 7.582; 95% confidence interval [CI], 2.991–19.222; P < 0.001) and squamous cell carcinoma (aOR, 3.520; 95% CI, 1.887–6.568; P < 0.001) were identified as independent predictors for no LNM by multivariate logistic regression analysis, while PR had no effect on LNM. The rate of LNM was 1.4% for low-risk patients (n = 73) identified as ERα positive with squamous cell carcinoma. The 5-year disease-free survival in low-risk patients was significantly greater than in those negative for ERα and/or those with non-squamous cell carcinoma (96.9% vs 80.1%, P = 0.002).

Conclusion

ERα positivity and squamous cell carcinoma are associated with a low risk of LNM in patients with stage IA2 to IIA2 cervical cancer. Hence, those patients without a low risk of LNM could be considered for definitive chemoradiotherapy to avoid unnecessary surgery.

MRI-based peritumoral radiomics analysis for preoperative prediction of lymph node metastasis in early-stage cervical cancer: A multi-center study

PURPOSE

To evaluate intra- and preitumoral radiomics on the contrast-enhanced T1-weighted (CE-T1) and T2-weighted (T2W) MRI for predicting the LNM, and develop a nomogram for potential clinical uses.

METHODS

We enrolled 169 cervical cancer cases who underwent CE-T1 and T2W MR scans from two hospitals between Dec. 2015 and Sep. 2021. Intra- and peritumoral features were extracted separately and selected by the least absolute shrinkage and selection operator (LASSO) regression. Radiomics signatures were built using the selected features from different regions. Clinical parameters were evaluated by statistical analysis. The nomogram was developed combining the multi-regional radiomics signature and the most predictive clinical parameters.

RESULTS

Five radiomics features were finally selected from the peritumoral regions with 1 and 3 mm distances in the CE-T1 and T2W MRI, respectively. The nomogram incorporating multi-regional combined radiomics signature, MR-reported LN status and tumor diameter achieved the highest AUCs in the training (nomogram vs. combined radiomics signature vs. clinical model, 0.891 vs. 0.830 vs. 0.812), internal validation (nomogram vs. combined radiomics signature vs. clinical model, 0.863 vs. 0.853 vs. 0.816) and external validation (nomogram vs. combined radiomics signature vs. clinical model, 0.804 vs. 0.701 vs. 0.787) cohort. DCA suggested good clinical usefulness of our developed models.

CONCLUSION

The current work suggested clinical potential for intra- and peritumoral radiomics with multi-modal MRI for preoperative predicting LNM.

Deep learning nomogram for predicting lymph node metastasis using computed tomography image in cervical cancer

BACKGROUND

Deep learning (DL) has been used on medical images to grade, differentiate, and predict prognosis in many tumors.

PURPOSE

To explore the effect of computed tomography (CT)-based deep learning nomogram (DLN) for predicting cervical cancer lymph node metastasis (LNM) before surgery.

MATERIAL AND METHODS

In total, 418 patients with stage IB-IIB cervical cancer were retrospectively enrolled for model exploration (n = 296) and internal validation (n = 122); 62 patients from another independent institution were enrolled for external validation. A convolutional neural network (CNN) was used for DL features extracting from all lesions. The least absolute shrinkage and selection operator (Lasso) logistic regression was used to develop a deep learning signature (DLS). A DLN incorporating the DLS and clinical risk factors was proposed to predict LNM individually. The performance of the DLN was evaluated on internal and external validation cohorts.

RESULTS

Stage, CT-reported pelvic lymph node status, and DLS were found to be independent predictors and could be used to construct the DLN. The combination showed a better performance than the clinical model and DLS. The proposed DLN had an area under the curve (AUC) of 0.925 in the training cohort, 0.771 in the internal validation cohort, and 0.790 in the external validation cohort. Decision curve analysis and stratification analysis suggested that the DLN has potential ability to generate a personalized probability of LNM in cervical cancer.

CONCLUSION

The proposed CT-based DLN could be used as a personalized non-invasive tool for preoperative prediction of LNM in cervical cancer, which could facilitate the choice of clinical treatment methods.

Combination of Radiomics and Machine Learning with Diffusion-Weighted MR Imaging for Clinical Outcome Prognostication in Cervical Cancer

Objectives: To explore the potential of Radiomics alone and in combination with a diffusion-weighted derived quantitative parameter, namely the apparent diffusion co-efficient (ADC), using supervised classification algorithms in the prediction of outcomes and prognosis. Materials and Methods: Retrospective evaluation of the imaging was conducted for a study cohort of uterine cervical cancer, candidates for radical treatment with chemo radiation. ADC values were calculated from the darkest part of the tumor, both before (labeled preADC) and post treatment (labeled postADC) with chemo radiation. Post extraction of 851 Radiomics features and feature selection analysis—by taking the union of the features that had Pearson correlation >0.35 for recurrence, >0.49 for lymph node and >0.40 for metastasis—was performed to predict clinical outcomes. Results: The study enrolled 52 patients who presented with variable FIGO stages in the age range of 28–79 (Median = 53 years) with a median follow-up of 26.5 months (range: 7–76 months). Disease recurrence occurred in 12 patients (23%). Metastasis occurred in 15 patients (28%). A model generated with 24 radiomics features and preADC using a monotone multi-layer perceptron neural network to predict the recurrence yields an AUC of 0.80 and a Kappa value of 0.55 and shows that the addition of radiomics features to ADC values improves the statistical metrics by approximately 40% for AUC and approximately 223% for Kappa. Similarly, the neural network model for prediction of metastasis returns an AUC value of 0.84 and a Kappa value of 0.65, thus exceeding performance expectations by approximately 25% for AUC and approximately 140% for Kappa. There was a significant input of GLSZM features (SALGLE and LGLZE) and GLDM features (SDLGLE and DE) in correlation with clinical outcomes of recurrence and metastasis. Conclusions: The study is an effort to bridge the unmet need of translational predictive biomarkers in the stratification of uterine cervical cancer patients based on prognosis.

Preoperative magnetic resonance imaging criteria for predicting lymph node metastasis in patients with stage IB1-IIA2 cervical cancer

Objective

This study aimed to identify patients with stage IB1‐IIA2 cervical cancer at low risk for lymph node metastasis (LNM) using preoperative magnetic resonance imaging (MRI) parameters.

Methods

Clinical and MRI data of patients with stage IB1‐IIA2 cervical cancer who underwent radical surgery between 2010 and 2015 were retrospectively reviewed. Clinical stage IB1‐IIA2 cervical cancer was diagnosed according to the 2009 International Federation of Gynecology and Obstetrics staging system. The low‐risk criteria for LNM were identified using logistic regression analysis. The performance of the logistic regression analysis was estimated through receiver operating characteristic curve analysis.

Results

Of 453 patients, 105 (23.2%) exhibited pathological LNM (p‐LNM). The maximal tumor diameter (adjusted odds ratio [aOR], 1.586; 95% confidence interval [CI], 1.312–1.916; p < 0.001) and LNM (aOR, 2.384; 95% CI, 1.418–4.007; p = 0.001) on preoperative MRI (m‐LNM) were identified as independent risk factors for p‐LNM using a multivariate logistic analysis. The p‐LNM rate was 4.0% for low‐risk patients (n = 124) identified using the current criteria (maximal tumor diameter <3.0 cm and no sign of m‐LNM). The 5‐year disease‐free survival rate of low‐risk patients was significantly greater than the rate of patients with a maximal tumor diameter ˃3.0 cm and/or signs of m‐LNM (90.4% vs. 82.1%; p = 0.033).

Conclusions

The low‐risk criteria for p‐LNM were a maximal tumor diameter <3.0 cm and no sign of m‐LNM. Patients with stage IB1‐IIA2 cervical cancer at low risk for m‐LNM could be candidates for radical surgery; hence, they have a lesser need for adjuvant chemoradiotherapy, thus avoiding the severe comorbidities it causes.

Preoperative Prediction of Lymphovascular Space Invasion in Cervical Cancer With Radiomics -Based Nomogram

Purpose

To build and evaluate a radiomics-based nomogram that improves the predictive performance of the LVSI in cervical cancer non-invasively before the operation.

Method

This study involved 149 patients who underwent surgery with cervical cancer from February 2017 to October 2019. Radiomics features were extracted from T2 weighted imaging (T2WI). The radiomic features were selected by logistic regression with the least absolute shrinkage and selection operator (LASSO) penalty in the training cohort. Based on the selected features, support vector machine (SVM) algorithm was used to build the radiomics signature on the training cohort. Incorporating radiomics signature and clinical risk factors, the radiomics-based nomogram was developed. The sensitivity, specificity, accuracy, and area under the curve (AUC) and Receiver operating characteristic (ROC) curve were calculated to assess these models.

Result

The radiomics model performed much better than the clinical model in both training (AUCs 0.925 vs. 0.786, accuracies 87.5% vs. 70.5%, sensitivities 83.6% vs. 41.7% and specificities 90.9% vs. 94.7%) and testing (AUCs 0.911 vs. 0.706, accuracies 84.0% vs. 71.3%, sensitivities 81.1% vs. 43.4% and specificities 86.4% vs. 95.0%). The combined model based on the radiomics signature and tumor stage, tumor infiltration depth and tumor pathology yielded the best performance (training cohort, AUC = 0.943, accuracies 89.5%, sensitivities 85.4% and specificities 92.9%; testing cohort, AUC = 0.923, accuracies 84.6%, sensitivities 84.0% and specificities 85.1%).

Conclusion

Radiomics-based nomogram was a useful tool for predicting LVSI of cervical cancer. This would aid the selection of the optimal therapeutic strategy and clinical decision-making for individuals.

The Value of MR-DWI and T1 Mapping in Indicating Radiation-Induced Soft Tissue Injury

Objective

To explore the value of MR-DWI and T1 mapping in predicting radiation-induced soft tissue fibrosis and its correlation with radiation inflammation.

Methods

① a total of 30 C57BL/6 mice were randomly divided into a control group (Nor group), irradiation group (IR group) and irradiation plus glycyrrhetinic acid group (GA group). The IR group and GA group were treated with 6MV X-rays to irradiate the right hind limbs of mice for 30 Gy in a single shot. MRI examinations were performed before and on the 7th day after irradiation to measure the apparent diffusion coefficient (ADC) value and the longitudinal relaxation time (T1) value of the hind limb muscles of the mice. On the 90th day after irradiation, the hind limb contracture was measured, and the right hind limb muscle was taken for HE staining, masson staining, immunohistochemical staining and Western blot analysis to detect the expression of a-SMA and Fibronectin. ② The other 30 mice were grouped randomly as above. On the 7th day after irradiation, the right hind limbs of the mice were examined by MRI to measure the ADC value and T1 value of the thigh muscles, and then the right hind thigh muscles were immediately sacrificed to detect IL-1β, IL-6, TNF-a and TGF-β1 expression with ELISA.

Results

On the 7th day after irradiation, the ADC values ​​of right hind thigh muscles of mice in Nor group, IR group and GA group were (1.35 ± 0.11)*10^-3mm2/s, (1.48 ± 0.07) *10^-3mm2/s and (1.36 ± 0.13)*10^-3mm2/s, respectively, by which the differences between the IR group and Nor group (P=0.008) and that between IR group and GA group (P=0.013) were statistically significant; T1 values ​​were (1369.7 ± 62.7)ms, (1483.7 ± 127.7)ms and (1304.1 ± 82.3)ms, respectively, with which the differences in the T1 value between the IR group and Nor group (P=0.012) and between IR group and GA group (P<0.001) were also statistically significant. On the 90th day after irradiation, the contracture lengths of the right hind limbs of the three groups of mice were (0.00 ± 0.07)cm, (2.08 ± 0.32)cm, and (1.49 ± 0.70) cm, respectively. There were statistically significant differences in the IR group compared with the Nor group (P<0.001) and the GA group (P=0.030). The ADC value (r=0.379, P=0.039) and T1 value (r=0.377, P=0.040) of the mice's hindlimbs on Day 7 after irradiation were correlated with the degree of contracture on Day 90 after irradiation; the ADC value (r=0.496, P=0.036) and T1 value (r=0.52, P=0.027) were positively correlated with the Masson staining results and with the expression of α-SMA and Fibronectin. While the ADC value was positively correlated with IL-6 (r=0.553, P=0.002), there was no obvious correlation with IL-1β, TNF-a and TGF-β1; the T1 value was positively correlated with IL-1β (r=0.419, P=0.021), IL-6 (r=0.535, P=0.002) and TNF-a (r=0.540, P=0.002) but not significantly related to TGF-β1 (r=0.155, P=0.413).

Conclusion

The MR-DWI and T1 mapping values on the 7th day after irradiation can reflect the early condition of tissue inflammation after the soft tissue is irradiated, and the values have a certain correlation with the degree of radiofibrosis of the soft tissue in the later period and may be used as an index to predict radiofibrosis.

Radiomics in radiation oncology for gynecological malignancies: a review of literature

Radiomics is the extraction of a significant number of quantitative imaging features with the aim of detecting information in correlation with useful clinical outcomes. Features are extracted, after delineation of an area of interest, from a single or a combined set of imaging modalities (including X-ray, US, CT, PET/CT and MRI). Given the high dimensionality, the analytical process requires the use of artificial intelligence algorithms. Firstly developed for diagnostic performance in radiology, it has now been translated to radiation oncology mainly to predict tumor response and patient outcome but other applications have been developed such as dose painting, prediction of side-effects, and quality assurance. In gynecological cancers, most studies have focused on outcomes of cervical cancers after chemoradiation. This review highlights the role of this new tool for the radiation oncologists with particular focus on female GU oncology.

Effect of the Number of Removed Lymph Nodes on Survival in Patients with FIGO Stage IB-IIA Cervical Squamous Cell Carcinoma following Open Radical Hysterectomy with Pelvic Lymphadenectomy: A Retrospective Cohort Study

Objective

To determine whether the number of removed lymph nodes (RLN) is associated with survival in patients with International Federation of Gynecology and Obstetrics (FIGO) stage IB-IIA cervical squamous cell carcinoma (CSCC).

Methods

We reviewed the medical records of FIGO stage IB-IIA CSCC patients who underwent standardized radical hysterectomy with pelvic lymphadenectomy (RHPL) in our center between 2006 and 2014. The X-tile software was performed to calculate the optimal grouping of cutoff points for RLN. The impact of RLN on progression-free survival (PFS) and overall survival (OS) was analyzed using Cox regression analysis.

Results

Among 3,127 patients, the mean number of RLN was 22, and positive lymph node (LN) was found in 668 (21.4%) patients. X-tile plots identified “21” and “16” as the optimal cutoff value of RLN to divide the patients into two groups in terms of PFS and OS separately. In all patients, the number of RLN was not associated with PFS (P=0.182) or OS (P=0.193). Moreover, in both LN positive and negative patients, the number of RLN was not associated with either PFS (P=0.212 and P=0.540, respectively) or OS (P=0.173 and P=0.497, respectively). Cox regression analysis showed that the number of RLN was not an independent prognostic factor for PFS or OS.

Conclusion

If standardized RHPL was performed, the number of RLN was not an independent prognostic factor for survival of patients with FIGO stage IB-IIA CSCC.

Feasibility of T2WI-MRI-based radiomics nomogram for predicting normal-sized pelvic lymph node metastasis in cervical cancer patients

OBJECTIVE

To investigate the feasibility of T₂WI-based radiomics nomogram analysis to non-invasively predict normal-sized pelvic lymph node (LN) metastasis (LNM) in cervical cancer patients.

METHODS

Preoperative images of 219 normal-sized pathologically confirmed LNs from 132 cervical cancer patients admitted to our hospital between January 2013 and March 2020 were retrospectively reviewed. Regions of interests (ROIs) were separately delineated on whole LNs and tumors. The maximum-relevance and minimum-redundancy (mRMR) and least absolute shrinkage and selection operator (LASSO) methods were used for the construction of radiomics signature. Logistic regression modeling was employed to build models based on clinical features on LN T₂WI (model 1), model 1 combined with LN radiomics features (model 2), and model 2 combined with tumor score (model 3). Diagnostic performance was assessed and compared.

RESULTS

Both model 2 and model 3 showed higher diagnostic accuracy (training: model 2 0.75, model 3 0.78, model 1 0.72; validation: model 2 0.77, model 3 0.69, model 1 0.66) and AUC (training: model 2 0.77, model 3 0.82, model 1 0.74; validation: model 2 0.75, model 3 0.74, model 1 0.70) than clinical model 1. Diagnostic performance of model 3 was improved compared with model 2 in primary cohort, but reduced in validation cohort. However, the differences did not show obvious statistical difference (p = 0.05 and p = 0.15).

CONCLUSIONS

T₂WI-based radiomics nomogram incorporating the LN radiomics signature with the clinical morphological LN features is promising for predicting the normal-sized pelvic LNM in cervical cancer patients. The original tumor radiomics analysis did not significantly improve the differential diagnosis of LNM.

KEY POINTS

• The combination of LN radiomics signature with LN clinical morphological features on T₂WI could discriminate LNM relatively well. • The tumor radiomics analysis did not significantly improve the differential diagnosis of LNM.

Machine learning-based MRI texture analysis to predict occult lymph node metastasis in early-stage oral tongue squamous cell carcinoma

OBJECTIVES

To develop and compare several machine learning models to predict occult cervical lymph node (LN) metastasis in early-stage oral tongue squamous cell cancer (OTSCC) from preoperative MRI texture features.

MATERIALS AND METHODS

We retrospectively enrolled 116 patients with early-stage OTSCC (cT1-2N0) who had been surgically treated by tumor excision and elective neck dissection (END). For each patient, we extracted 86 texture features from T2-weighted imaging (T2WI) and contrast-enhanced T1-weighted imaging (ceT1WI), respectively. Dimension reduction was performed in three consecutive steps: reproducibility analysis, collinearity analysis, and information gain algorithm. Models were created using six machine learning methods, including logistic regression (LR), random forest (RF), naïve Bayes (NB), support vector machine (SVM), AdaBoost, and neural network (NN). Their performance was assessed using tenfold cross-validation.

RESULTS

Occult LN metastasis was pathologically detected in 42.2% (49/116) of the patients. No significant association was identified between node status and patients' gender, age, or clinical T stage. Dimension reduction steps selected 6 texture features. The NB model gave the best overall performance, which correctly classified the nodal status in 74.1% (86/116) of the carcinomas, with an AUC of 0.802.

CONCLUSION

Machine learning-based MRI texture analysis offers a feasible tool for preoperative prediction of occult cervical node metastasis in early-stage OTSCC.

KEY POINTS

• A machine learning-based MRI texture analysis approach was adopted to predict occult cervical node metastasis in early-stage OTSCC with no evidence of node involvement on conventional images. • Six texture features from T2WI and ceT1WI of preoperative MRI were selected to construct the predictive model. • After comparing six machine learning methods, naïve Bayes (NB) achieved the best performance by correctly identifying the node status in 74.1% of the patients, using tenfold cross-validation.

Feasibility of MRI-based radiomics features for predicting lymph node metastases and VEGF expression in cervical cancer

PURPOSE

To investigate the predictive value of MRI-based radiomics features for lymph node metastasis (LNM) and vascular endothelial growth factor (VEGF) expression in patients with cervical cancer.

METHOD

A total of 163 patients with cervical cancer were enrolled in this study. A total of 134 patients were included for LNM differentiation, and 118 were included for VEGF expression discrimination. The patients were randomly assigned to the training group or test group at a ratio of 2:1. Radiomics features were extracted from T1WI enhanced and T2WI MRI scans of each patient, and tumor stage was also documented according to the International Federation of Gynecology and Obstetrics (FIGO) guidelines. The least absolute shrinkage and selection operator algorithm was used for feature selection. The results of 5-fold cross validation were applied to select the best classification models. The performances of the constructed models were further evaluated with the test group.

RESULTS

Sixteen radiomics features and the FIGO stage were selected to construct the LNM discrimination model. The LNM prediction model achieved the best diagnostic performance, with areas under the receiver operating curve (AUCs) of 0.95 and 0.88 in the training group and test group, respectively. Nine radiomics characteristics were screened to build the VEGF prediction model, with AUCs of 0.82 and 0.70 in the training group and test group, respectively. Decision curve analysis confirmed their clinical usefulness.

CONCLUSIONS

The presented radiomics prediction models demonstrated potential to noninvasively differentiate LNM and VEGF expression in cervical cancer.

The Role of Multiparametric Magnetic Resonance Imaging in the Study of Primary Tumor and Pelvic Lymph Node Metastasis in Stage IB1-IIA1 Cervical Cancer

OBJECTIVE

The aim of this study was to investigate the value of multiparametric magnetic resonance imaging (MRI) in demonstrating the metastatic potential of primary tumor and differentiating metastatic lymph nodes (MLNs) from nonmetastatic lymph nodes (non-MLNs) in stage IB1-IIA1 cervical cancer.

METHODS

Fifty-seven stage IB1-IIA1 subjects were included. The apparent diffusion coefficient (ADC) values and dynamic contrast-enhanced MRI (DCE-MRI) parameters of primary tumors and lymph nodes and the conventional imaging features of the lymph nodes were measured and analyzed. Mann-Whitney test and χ test were used to analyze statistically significant parameters, logistic regression was used for multivariate analysis, and receiver operating characteristic analysis was used to compare the diagnostic performance of the MLNs.

RESULTS

Nineteen subjects had lymph node metastasis. A total of 94 lymph nodes were evaluated, including 30 MLNs and 64 non-MLNs. There were no significant difference in ADC and DCE-MRI parameters between metastatic and nonmetastatic primary tumors. The heterogeneous signal was more commonly seen in MLNs than in non-MLNs (P = 0.001). The values of ADCmean, ADCmin, and ADCmax of MLNs were lower than those of non-MLNs (P < 0.001). The values of short-axis diameter, K, Kep, and Ve of MLNs were higher than those of non-MLNs (P < 0.05). Compared with individual MRI parameters, the combined evaluation of short-axis diameter, ADCmean, and K showed the highest area under the curve of 0.930.

CONCLUSIONS

Diffusion-weighted imaging and DCE-MRI could not demonstrate the metastatic potential of primary tumor in stage IB1-IIA1 cervical cancer. Compared with individual MRI parameters, the combination of multiparametric MRI could improve the diagnostic performance of lymph node metastasis.

Radiomics Analysis of Multiparametric MRI for the Preoperative Prediction of Lymph Node Metastasis in Cervical Cancer

Objective: To develop and validate a radiomics predictive model based on multiparameter MR imaging features and clinical features to predict lymph node metastasis (LNM) in patients with cervical cancer.

Material and Methods: A total of 168 consecutive patients with cervical cancer from two centers were enrolled in our retrospective study. A total of 3,930 imaging features were extracted from T2-weighted (T2W), ADC, and contrast-enhanced T1-weighted (cT1W) images for each patient. Four-step procedures, mainly minimum redundancy maximum relevance (MRMR) and least absolute shrinkage and selection operator (LASSO) regression, were applied for feature selection and radiomics signature building in the training set from center I (n = 115). Combining clinical risk factors, a radiomics nomogram was then constructed. The models were then validated in the external validation set comprising 53 patients from center II. The predictive performance was determined by its calibration, discrimination, and clinical usefulness.

Results: The radiomics signature derived from the combination of T2W, ADC, and cT1W images, composed of six LN-status-related features, was significantly associated with LNM and showed better predictive performance than signatures derived from either of them alone in both sets. Encouragingly, the radiomics signature also showed good discrimination in the MRI-reported LN-negative subgroup, with AUC of 0.825 (95% CI: 0.732–0.919). The radiomics nomogram that incorporated radiomics signature and MRI-reported LN status also showed good calibration and discrimination in both sets, with AUCs of 0.865 (95% CI: 0.794–0.936) and 0.861 (95% CI: 0.733–0.990), respectively. Decision curve analysis confirmed its clinical usefulness.

Conclusion: The proposed MRI-based radiomics nomogram has good performance for predicting LN metastasis in cervical cancer and may be useful for improving clinical decision making.

The invasion status of lymphovascular space and lymph nodes in cervical cancer assessed by mono-exponential and bi-exponential DWI-related parameters

AIM

To investigate whether mono-exponential and bi-exponential diffusion-weighted imaging (DWI)-related parameters of the primary tumour can evaluate the status of lymphovascular space invasion (LVSI) and lymph node metastasis (LNM) in patients with cervical carcinoma preoperatively.

MATERIALS AND METHODS

Eighty patients with cervical carcinoma were enrolled, who underwent preoperative multi b-value DWI and radical hysterectomy. They were classified into LVSI(+) versus LVSI(-) and LNM(+) versus LNM(-) according to postoperative pathology. The apparent diffusion coefficient (ADC), pure molecular diffusion (D), pseudo-diffusion coefficient (D∗), and perfusion fraction (f) were calculated from the whole tumour (__whole) and tumour margin (__margin). All parameters were compared between LVSI(+) and LVSI(-) and between LNM(+) and LNM(-). Logistic regression analysis and receiver operating characteristic (ROC) curve analysis were performed to evaluate the diagnostic performance of these parameters.

RESULTS

f__margin and D∗__whole showed significant differences in differentiating LVSI(+) from LVSI(-) tumours (p=0.002, 0.008, respectively), while LNM(+) tumours presented with significantly higher ADC__margin than that of LNM(-) tumours (p=0.009). The other parameters were not independent related factors with the status of LVSI or LNM according to logistic regression analysis (p>0.05). The area under the ROC curve of f__margin combined with D∗__whole in discriminating LVSI(+) from LVSI(-) was 0.826 (95% confidence interval [CI]: 0.691-0.961), while ADC__margin in differentiating LNM(+) from LNM(-) was 0.788 (95% CI: 0.648-0.928).

CONCLUSIONS

The parameters generated from mono-exponential and bi-exponential DWI of the primary cervical carcinoma could help discriminate its status regarding LVSI (f__margin and D∗__whole) and LNM (ADC__margin).

Development of a Deep Learning Model to Identify Lymph Node Metastasis on Magnetic Resonance Imaging in Patients With Cervical Cancer

IMPORTANCE

Accurate identification of lymph node metastasis preoperatively and noninvasively in patients with cervical cancer can avoid unnecessary surgical intervention and benefit treatment planning.

OBJECTIVE

To develop a deep learning model using preoperative magnetic resonance imaging for prediction of lymph node metastasis in cervical cancer.

DESIGN, SETTING, AND PARTICIPANTS

This diagnostic study developed an end-to-end deep learning model to identify lymph node metastasis in cervical cancer using magnetic resonance imaging (MRI). A total of 894 patients with stage IB to IIB cervical cancer who underwent radical hysterectomy and pelvic lymphadenectomy were reviewed. All patients underwent radical hysterectomy and pelvic lymphadenectomy, received pelvic MRI within 2 weeks before the operations, had no concurrent cancers, and received no preoperative treatment. To achieve the optimal model, the diagnostic value of 3 MRI sequences was compared, and the outcomes in the intratumoral and peritumoral regions were explored. To mine tumor information from both image and clinicopathologic levels, a hybrid model was built and its prognostic value was assessed by Kaplan-Meier analysis. The deep learning model and hybrid model were developed on a primary cohort consisting of 338 patients (218 patients from Sun Yat-sen University Cancer Center, Guangzhou, China, between January 2011 and December 2017 and 120 patients from Henan Provincial People's Hospital, Zhengzhou, China, between December 2016 and June 2018). The models then were evaluated on an independent validation cohort consisting of 141 patients from Yunnan Cancer Hospital, Kunming, China, between January 2011 and December 2017.

MAIN OUTCOMES AND MEASURES

The primary diagnostic outcome was lymph node metastasis status, with the pathologic characteristics diagnosed by lymphadenectomy. The secondary primary clinical outcome was survival. The primary diagnostic outcome was assessed by receiver operating characteristic (area under the curve [AUC]) analysis; the primary clinical outcome was assessed by Kaplan-Meier survival analysis.

RESULTS

A total of 479 patients (mean [SD] age, 49.1 [9.7] years) fulfilled the eligibility criteria and were enrolled in the primary (n = 338) and validation (n = 141) cohorts. A total of 71 patients (21.0%) in the primary cohort and 32 patients (22.7%) in the validation cohort had lymph node metastais confirmed by lymphadenectomy. Among the 3 image sequences, the deep learning model that used both intratumoral and peritumoral regions on contrast-enhanced T1-weighted imaging showed the best performance (AUC, 0.844; 95% CI, 0.780-0.907). These results were further improved in a hybrid model that combined tumor image information mined by deep learning model and MRI-reported lymph node status (AUC, 0.933; 95% CI, 0.887-0.979). Moreover, the hybrid model was significantly associated with disease-free survival from cervical cancer (hazard ratio, 4.59; 95% CI, 2.04-10.31; P < .001).

CONCLUSIONS AND RELEVANCE

The findings of this study suggest that deep learning can be used as a preoperative noninvasive tool to diagnose lymph node metastasis in cervical cancer.

Multiparametric MRI-Based Radiomics Nomogram for Predicting Lymph Node Metastasis in Early-Stage Cervical Cancer

BACKGROUND

Lymph node metastasis (LNM) is a critical risk factor affecting treatment strategy and prognosis in patients with early-stage cervical cancer.

PURPOSE

To establish a multiparametric MRI (mpMRI)-based radiomics nomogram for preoperatively predicting LNM status.

STUDY TYPE

Retrospective.

POPULATION

Among 233 consecutive patients, 155 patients were randomly allocated to the primary cohort and 78 patients to the validation cohort.

FIELD STRENGTH

Radiomic features were extracted from a 1.5T mpMRI scan (T₁ -weighted imaging [T₁ WI], fat-saturated T₂ -weighted imaging [FS-T₂ WI], contrast-enhanced [CE], diffusion-weighted imaging [DWI], and apparent diffusion coefficient [ADC] maps).

ASSESSMENT

The performance of the nomogram was assessed with respect to its calibration, discrimination, and clinical usefulness. The area under the receiver operating characteristics curve (ROC AUC), accuracy, sensitivity, and specificity were also calculated.

STATISTICAL TESTS

The least absolute shrinkage and selection operator (LASSO) method was used for dimension reduction, feature selection, and radiomics signature building. Multivariable logistic regression analysis was used to develop the radiomics nomogram. An independent sample t-test and chi-squared test were used to compare the differences in continuous and categorical variables, respectively.

RESULTS

The radiomic signature allowed a good discrimination between the LNM and non-LNM groups, with a C-index of 0.856 (95% confidence interval [CI], 0.794-0.918) in the primary cohort and 0.883 (95% CI, 0.809-0.957) in the validation cohort. Additionally, the radiomics nomogram also had a good discriminating performance and yielded good calibration both in the primary and validation cohorts (C-index, 0.882 [95% CI, 0.827-0.937], C-index, 0.893 [95% CI, 0.822-0.964], respectively). Decision curve analysis demonstrated that the radiomics nomogram was clinically useful.

DATA CONCLUSION

A radiomics nomogram was developed by incorporating the radiomics signature with the MRI-reported LN status and FIGO stage. This nomogram might be used to facilitate the individualized prediction of LNM in patients with early-stage cervical cancer.

LEVEL OF EVIDENCE

3 TECHNICAL EFFICACY STAGE: 2 J. Magn. Reson. Imaging 2020;52:885-896.

Noninvasive prediction of lymph node status for patients with early-stage cervical cancer based on radiomics features from ultrasound images

OBJECTIVE

To investigate the feasibility of a noninvasive detection of lymph node metastasis (LNM) for early-stage cervical cancer (ECC) patients with radiomics methods based on the textural features from ultrasound images.

METHODS

One hundred seventy-two ECC patients between January 2014 and September 2018 with pathologically confirmed lymph node status (LNS) and preoperative ultrasound images were retrospectively reviewed. Regions of interest (ROIs) were delineated by a senior radiologist in the ultrasound images. LIFEx was applied to extract textural features for radiomics study. Least absolute shrinkage and selection operator (LASSO) regression was applied for dimension reduction and for selection of key features. A multivariable logistic regression analysis was adopted to build the radiomics signature. The Mann-Whitney U test was applied to investigate the correlation between radiomics and LNS for both training and validation cohorts. Receiver operating characteristic (ROC) curves were applied to evaluate the accuracy of the radiomics prediction models.

RESULTS

A total of 152 radiomics features were extracted from ultrasound images, in which 6 features were significantly associated with LNS (p < 0.05). The radiomics signatures demonstrated a good discrimination between patients with LNM and non-LNM groups. The best radiomics performance model achieved an area under the curve (AUC) of 0.79 (95% confidence interval (CI), 0.71-0.88) in the training cohort and 0.77 (95% CI, 0.65-0.88) in the validation cohort.

CONCLUSIONS

The feasibility of radiomics features from ultrasound images for the prediction of LNM in ECC was investigated. This noninvasive prediction method may be used to facilitate preoperative identification of LNS in patients with ECC.

KEY POINTS

• Few studied had investigated the feasibility of radiomics based on ultrasound images for cervical cancer, even though it is the most common practice for gynecological cancer diagnosis and treatment. • The radiomics signatures based on ultrasound images demonstrated a good discrimination between patients with and without lymph node metastasis with an area under the curve (AUC) of 0.79 and 0.77 in the training and validation cohorts, respectively. • The radiomics model based on preoperative ultrasound images has the potential ability to predict lymph node status noninvasively in patients with early-state cervical cancer, so as to reduce the impact of invasive examination and to optimize the treatment choices.