Using Radiomics and Machine Learning Applied to MRI to Predict Response to Neoadjuvant Chemotherapy in Locally Advanced Cervical Cancer

Assessing the Efficacy of 18F FDG PET-CT in Preoperative Staging of Early-Stage Cervical Cancer in Novi Sad, Serbia-A Pilot Study

Background and Objectives: This study primarily aims to evaluate the preoperative staging effectiveness of PET-CT in early-stage cervical cancer, particularly, its ability to detect primary tumors and micrometastases. Materials and Methods: In this retrospective study, cervical cancer patients who had undergone preoperative 18F FDG PET-CT scans and were treated at the Department of Gynecology, Institute of Oncology, Vojvodina, in Sremska Kamenica, during the period from 2016 to 2020 were analyzed. Results: The study included 62 patients (mean age, 49.3 ± 9.6 years). Squamous cell carcinoma was the predominant histological type (95.2%), with G2 differentiation (82.3%) and FIGO stage Ib1 (80.6%) being the most common. Assessed by 18F FDG PET-CT, the mean tumor size was 26.4 ± 10.8 mm, which is slightly lower than the 26.9 mm measured during clinical examination (p = 0.784), with a significant (r = 0.678, p < 0.001) correlation between these methods. 18F FDG PET-CT demonstrated an overall accuracy of 88.7% for identifying primary tumors, with a sensitivity of 86.8%, specificity of 100.0%, PPV of 100.0%, and NPV of 56.2%. An intraoperative examination showed identical overall accuracy but higher sensitivity (98.1%) and lower specificity (33.3%). For 18F FDG PET-CT, the level of agreement with the histopathological examination was good (Kappa 0.656), while for the intraoperative examination, it was moderate (Kappa 0.409). Regarding the lymph node assessment, 18F FDG PET-CT's accuracy was 82.2%, with a sensitivity of 53.8% and a specificity of 89.8%. The intraoperative examination showed lower accuracy (66.1%) but higher sensitivity (76.9%). The 18F FDG PET-CT Kappa value indicated moderate agreement (0.449), while the intraoperative examination showed poor agreement (0.282). Conclusions: In conclusion, significant effectiveness is shown by 18F FDG PET-CT for preoperative staging of early-stage cervical cancer, offering superior accuracy in detecting primary tumors and micrometastases, particularly in predicting lymph node metastases, thereby enhancing diagnostic accuracy and informing treatment decisions.

MRI radiomics and nutritional-inflammatory biomarkers: a powerful combination for predicting progression-free survival in cervical cancer patients undergoing concurrent chemoradiotherapy

OBJECTIVE

This study aims to develop and validate a predictive model that integrates clinical features, MRI radiomics, and nutritional-inflammatory biomarkers to forecast progression-free survival (PFS) in cervical cancer (CC) patients undergoing concurrent chemoradiotherapy (CCRT). The goal is to identify high-risk patients and guide personalized treatment.

METHODS

We performed a retrospective analysis of 188 patients from two centers, divided into training (132) and validation (56) sets. Clinical data, systemic inflammatory markers, and immune-nutritional indices were collected. Radiomic features from three MRI sequences were extracted and selected for predictive value. We developed and evaluated five models incorporating clinical features, nutritional-inflammatory indicators, and radiomics using C-index. The best-performing model was used to create a nomogram, which was validated through ROC curves, calibration plots, and decision curve analysis (DCA).

RESULTS

Model 5, which integrates clinical features, Systemic Immune-Inflammation Index (SII), Prognostic Nutritional Index (PNI), and MRI radiomics, showed the highest performance. It achieved a C-index of 0.833 (95% CI: 0.792-0.874) in the training set and 0.789 (95% CI: 0.679-0.899) in the validation set. The nomogram derived from Model 5 effectively stratified patients into risk groups, with AUCs of 0.833, 0.941, and 0.973 for 1-year, 3-year, and 5-year PFS in the training set, and 0.812, 0.940, and 0.944 in the validation set.

CONCLUSIONS

The integrated model combining clinical features, nutritional-inflammatory biomarkers, and radiomics offers a robust tool for predicting PFS in CC patients undergoing CCRT. The nomogram provides precise predictions, supporting its application in personalized patient management.

Comparative Analysis of Repeatability in CT Radiomics and Dosiomics Features under Image Perturbation: A Study in Cervical Cancer Patients

This study aims to evaluate the repeatability of radiomics and dosiomics features via image perturbation of patients with cervical cancer. A total of 304 cervical cancer patients with planning CT images and dose maps were retrospectively included. Random translation, rotation, and contour randomization were applied to CT images and dose maps before radiomics feature extraction. The repeatability of radiomics and dosiomics features was assessed using intra-class correlation of coefficient (ICC). Pearson correlation coefficient (r) was adopted to quantify the correlation between the image characteristics and feature repeatability. In general, the repeatability of dosiomics features was lower compared with CT radiomics features, especially after small-sigma Laplacian-of-Gaussian (LoG) and wavelet filtering. More repeatable features (ICC > 0.9) were observed when extracted from the original, Large-sigma LoG filtered, and LLL-/LLH-wavelet filtered images. Positive correlations were found between image entropy and high-repeatable feature number in both CT and dose (r = 0.56, 0.68). Radiomics features showed higher repeatability compared to dosiomics features. These findings highlight the potential of radiomics features for robust quantitative imaging analysis in cervical cancer patients, while suggesting the need for further refinement of dosiomics approaches to enhance their repeatability.

An interpretable clinical ultrasound-radiomics combined model for diagnosis of stage I cervical cancer

Objective

The purpose of this retrospective study was to establish a combined model based on ultrasound (US)-radiomics and clinical factors to predict patients with stage I cervical cancer (CC) before surgery.

Materials and methods

A total of 209 CC patients who had cervical lesions found by transvaginal sonography (TVS) from the First Affiliated Hospital of Anhui Medical University were retrospectively reviewed, patients were divided into the training set (n = 146) and internal validation set (n = 63), and 52 CC patients from Anhui Provincial Maternity and Child Health Hospital and Nanchong Central Hospital were taken as the external validation set. The clinical independent predictors were selected by univariate and multivariate logistic regression analyses. US-radiomics features were extracted from US images. After selecting the most significant features by univariate analysis, Spearman's correlation analysis, and the least absolute shrinkage and selection operator (LASSO) algorithm, six machine learning (ML) algorithms were used to build the radiomics model. Next, the ability of the clinical, US-radiomics, and clinical US-radiomics combined model was compared to diagnose stage I CC. Finally, the Shapley additive explanations (SHAP) method was used to explain the contribution of each feature.

Results

Long diameter of the cervical lesion (L) and squamous cell carcinoma-associated antigen (SCCa) were independent clinical predictors of stage I CC. The eXtreme Gradient Boosting (Xgboost) model performed the best among the six ML radiomics models, with area under the curve (AUC) values in the training, internal validation, and external validation sets being 0.778, 0.751, and 0.751, respectively. In the final three models, the combined model based on clinical features and rad-score showed good discriminative power, with AUC values in the training, internal validation, and external validation sets being 0.837, 0.828, and 0.839, respectively. The decision curve analysis validated the clinical utility of the combined nomogram. The SHAP algorithm illustrates the contribution of each feature in the combined model.

Conclusion

We established an interpretable combined model to predict stage I CC. This non-invasive prediction method may be used for the preoperative identification of patients with stage I CC.

Diagnostic Accuracy of Transvaginal Ultrasound and Magnetic Resonance Imaging for the Detection of Myometrial Infiltration in Endometrial Cancer: A Systematic Review and Meta-Analysis

In endometrial cancer (EC), deep myometrial invasion (DMI) is a prognostic factor that can be evaluated by various imaging methods; however, the best method of choice is uncertain. We aimed to compare the diagnostic performance of two-dimensional transvaginal ultrasound (TVS) and magnetic resonance imaging (MRI) in the preoperative detection of DMI in patients with EC. Pubmed, Embase and Cochrane Library were systematically searched in May 2023. We included original articles that compared TVS to MRI on the same cohort of patients, with final histopathological confirmation of DMI as reference standard. Several subgroup analyses were performed. Eighteen studies comprising 1548 patients were included. Pooled sensitivity and specificity were 76.6% (95% confidence interval (CI), 70.9-81.4%) and 87.4% (95% CI, 80.6-92%) for TVS. The corresponding values for MRI were 81.1% (95% CI, 74.9-85.9%) and 83.8% (95% CI, 79.2-87.5%). No significant difference was observed (sensitivity: p = 0.116, specificity: p = 0.707). A non-significant difference between TVS and MRI was observed when no-myometrium infiltration vs. myometrium infiltration was considered. However, when only low-grade EC patients were evaluated, the specificity of MRI was significantly better (p = 0.044). Both TVS and MRI demonstrated comparable sensitivity and specificity. Further studies are needed to assess the presence of myometrium infiltration in patients with fertility-sparing wishes.

Constructing a Classification Model for Cervical Cancer Tumor Tissue and Normal Tissue Based on CT Radiomics

This study aimed to develop an automated classification framework for distinguishing between cervical cancer tumor and normal uterine tissue, leveraging CT images for radiomics feature extraction. We retrospectively analyzed CT images from 117 cervical cancer patients. To distinguish between cancerous and healthy tissue, we segmented gross tumor volume and normal uterine tissue as distinct regions of interest (ROIs) using manual segmentation techniques. Key radiomic parameters were extracted from these ROIs. To bolster model's predictive capability, the data was stratified into train data (70%) and validation data (30%). During feature selection phase, we applied Least Absolute Shrinkage and Selection Operator regression algorithm to identify most relevant features. Subsequently, we built classification models using five state-of-the-art machine learning algorithms: Support Vector Machine (SVM), Random Forest (RF), K-Nearest Neighbors (KNN), Extreme Gradient Boosting (XGBoost), and Decision Tree (DT). Ultimately, the performance of each model was evaluated. Through stringent feature selection process, we identified 18 pivotal radiomic features for classification of cervical cancer and normal uterine tissue. When applied to test data, all five models achieved excellent performance, with area under the curve (AUC) values ranging from 0.8866 to 0.9190 (SVM: 0.9144, RF: 0.9078, KNN: 0.9051, DT: 0.8866, XGBoost: 0.9190), all surpassing threshold of 0.8. In terms of test data, all five models had high sensitivity; accuracy of SVM, RF, and XGBoost models was comparable; and specificity of five models was similar. XGBoost model outperformed the others in terms of diagnostic accuracy, achieving an AUC of 0.8737 (95% CI: 0.8198-0.9277) for train data and 0.9190 (95% CI: 0.8525-0.9854) for test data. Our findings underscore the potential of CT radiomics combined with machine learning algorithms for accurately classifying cervical cancer tumors and normal uterine tissue with high recognition capabilities. This approach holds significant promise for clinical diagnostics.