Incorporating Radiomics into Machine Learning Models to Predict Outcomes of Neuroblastoma

The relationship between contrast-enhanced computed tomography radiomics features and mitosis karyorrhexis index in neuroblastoma

OBJECTIVE

Mitosis karyorrhexis index (MKI) can reflect the proliferation status of neuroblastoma cells. This study aimed to investigate the contrast-enhanced computed tomography (CECT) radiomics features associated with the MKI status in neuroblastoma.

MATERIALS AND METHODS

246 neuroblastoma patients were retrospectively included and divided into three groups: low-MKI, intermediate-MKI, and high-MKI. They were randomly stratified into a training set and a testing set at a ratio of 8:2. Tumor regions of interest were delineated on arterial-phase CECT images, and radiomics features were extracted. After reducing the dimensionality of the radiomics features, a random forest algorithm was employed to establish a three-class classification model to predict MKI status.

RESULTS

The classification model consisted of 5 radiomics features. The mean area under the curve (AUC) of the classification model was 0.916 (95% confidence interval (CI) 0.913-0.921) in the training set and 0.858 (95% CI 0.841-0.864) in the testing set. Specifically, the classification model achieved AUCs of 0.928 (95% CI 0.927-0.934), 0.915 (95% CI 0.912-0.919), and 0.901 (95% CI 0.900-0.909) for predicting low-MKI, intermediate-MKI, and high-MKI, respectively, in the training set. In the testing set, the classification model achieved AUCs of 0.873 (95% CI 0.859-0.882), 0.860 (95% CI 0.852-0.872), and 0.820 (95% CI 0.813-0.839) for predicting low-MKI, intermediate-MKI, and high-MKI, respectively.

CONCLUSIONS

CECT radiomics features were found to be correlated with MKI status and are helpful for reflecting the proliferation status of neuroblastoma cells.

Applications of Artificial Intelligence for Pediatric Cancer Imaging

Artificial intelligence (AI) is transforming medical imaging of adult patients. However, its utilization in pediatric oncology imaging remains constrained, in part due to the inherent data scarcity associated with childhood cancers. Pediatric cancers are rare, and imaging technologies are evolving rapidly, leading to insufficient data of a particular type to effectively train these algorithms. The small market size of pediatrics compared to adults could also contribute to this challenge, as market size is a driver of commercialization. This article provides an overview of the current state of AI applications for pediatric cancer imaging, including applications for medical image acquisition, processing, reconstruction, segmentation, diagnosis, staging, and treatment response monitoring. While current developments are promising, impediments due to diverse anatomies of growing children and nonstandardized imaging protocols have led to limited clinical translation thus far. Opportunities include leveraging reconstruction algorithms to achieve accelerated low-dose imaging and automating the generation of metric-based staging and treatment monitoring scores. Transfer-learning of adult-based AI models to pediatric cancers, multi-institutional data sharing, and ethical data privacy practices for pediatric patients with rare cancers will be keys to unlocking AI's full potential for clinical translation and improved outcomes for these young patients.

Application of Machine Learning and Deep EfficientNets in Distinguishing Neonatal Adrenal Hematomas From Neuroblastoma in Enhanced Computed Tomography Images

Background

The aim of the study was to employ a combination of radiomic indicators based on computed tomography (CT) imaging and machine learning (ML), along with deep learning (DL), to differentiate between adrenal hematoma and adrenal neuroblastoma in neonates.

Methods

A total of 76 neonates were included in this retrospective study (40 with neuroblastomas and 36 with adrenal hematomas) who underwent CT and divided into a training group (n = 38) and a testing group (n = 38). The regions of interest (ROIs) were segmented by two radiologists to extract radiomics features using Pyradiomics package. ML classifications were done using support vector machine (SVM), AdaBoost, Extra Trees, gradient boosting, multi-layer perceptron (MLP), and random forest (RF). EfficientNets was employed and classified, based on radiometrics. The area under curve (AUC) of the receiver operating characteristic (ROC) was calculated to assess the performance of each model.

Results

Among all features, the least absolute shrinkage and selection operator (LASSO) logistic regression selected nine features. These radiomics features were used to construct radiomics model. In the training cohort, the AUCs of SVM, MLP and Extra Trees models were 0.967, 0.969 and 1.000, respectively. The corresponding AUCs of the test cohort were 0.985, 0.971 and 0.958, respectively. In the classification task, the AUC of the DL framework was 0.987.

Conclusion

ML decision classifiers and DL framework constructed from CT-based radiomics features offered a non-invasive method to differentiate neonatal adrenal hematoma from neuroblastoma and performed better than the clinical experts.

Predicting Neuroblastoma Patient Risk Groups, Outcomes, and Treatment Response Using Machine Learning Methods: A Review

Neuroblastoma, a paediatric malignancy with high rates of cancer-related morbidity and mortality, is of significant interest to the field of paediatric cancers. High-risk NB tumours are usually metastatic and result in survival rates of less than 50%. Machine learning approaches have been applied to various neuroblastoma patient data to retrieve relevant clinical and biological information and develop predictive models. Given this background, this study will catalogue and summarise the literature that has used machine learning and statistical methods to analyse data such as multi-omics, histological sections, and medical images to make clinical predictions. Furthermore, the question will be turned on its head, and the use of machine learning to accurately stratify NB patients by risk groups and to predict outcomes, including survival and treatment response, will be summarised. Overall, this study aims to catalogue and summarise the important work conducted to date on the subject of expression-based predictor models and machine learning in neuroblastoma for risk stratification and patient outcomes including survival, and treatment response which may assist and direct future diagnostic and therapeutic efforts.

A narrative review of radiomics and deep learning advances in neuroblastoma: updates and challenges

Neuroblastoma is an extremely heterogeneous tumor that commonly occurs in children. The diagnosis and treatment of this tumor pose considerable challenges due to its varied clinical presentations and intricate genetic aberrations. Presently, various imaging modalities, including computed tomography, magnetic resonance imaging, and positron emission tomography, are utilized to assess neuroblastoma. Nevertheless, these conventional imaging modalities have limitations in providing quantitative information for accurate diagnosis and prognosis. Radiomics, an emerging technique, can extract intricate medical imaging information that is imperceptible to the human eye and transform it into quantitative data. In conjunction with deep learning algorithms, radiomics holds great promise in complementing existing imaging modalities. The aim of this review is to showcase the potential of radiomics and deep learning advancements to enhance the diagnostic capabilities of current imaging modalities for neuroblastoma.

Computed tomography-based radiomics to assess risk stratification in pediatric malignant peripheral neuroblastic tumors

This study aimed to develop and validate an analysis system based on preoperative computed tomography (CT) to predict the risk stratification in pediatric malignant peripheral neuroblastic tumors (PNTs). A total of 405 patients with malignant PNTs (184 girls and 221 boys; mean age, 33.8 ± 29.1 months) were retrospectively evaluated between January 2010 and June 2018. Radiomic features were extracted from manually segmented tumors on preoperative CT images. Spearman's rank correlation coefficient and the least absolute shrinkage and selection operator (LASSO) were used to eliminate redundancy and select features. A risk model was built to stratify low-, intermediate-, and high-risk groups. An image-defined risk factor (IDRFs) model was developed to classify 266 patients with malignant PNTs and one or more IDRFs into high-risk and non-high-risk groups. The performance of the predictive models was evaluated with respect to accuracy (Acc) and receiver operating characteristic (ROC) curve, including the area under the ROC curve (AUC). The risk model demonstrated good discrimination capability, with an area under the curve (AUC) of 0.903 to distinguish high-risk from non-high-risk groups, and 0.747 to classify intermediate- and low-risk groups. In the IDRF-based risk model with the number of IDRFs, the AUC was 0.876 for classifying the high-risk and non-high-risk groups. Radiomic analysis based on preoperative CT images has the potential to stratify the risk of pediatric malignant PNTs. It had outstanding efficiency in distinguishing patients in the high-risk group, and this predictive model of risk stratification could assist in selecting optimal aggressive treatment options.

Machine learning-based predictive model for the differential diagnosis of ≤ 5 cm gastric stromal tumor and gastric schwannoma based on CT images

The clinical symptoms of ≤ 5 cm gastric stromal tumor (GST) and gastric schwannoma (GS) are similar, but the treatment regimens are different. This study explored the value of computed tomography (CT) combined with machine learning (ML) algorithms to find the best model to discriminate them. A total of 126 patients with GST ≤ 5 cm and 35 patients with GS ≤ 5 during 2013-2022 were included. CT imaging features included qualitative data (tumor location, growth pattern, lobulation, surface ulcer status, necrosis, calcification, and surrounding lymph nodes) and quantitative data [long diameter (LD); short diameter (SD); LD/SD ratio; degree of enhancement (DE); heterogeneous degree (HD)]. Patients were randomly divided into a training set (n = 112) and test set (n = 49) using 7:3 stratified sampling. The univariate and multivariate logistic regression analysis were used to identify independent risk factors. Five ML algorithms were used to build prediction models: Support Vector Machine, k-Nearest Neighbor, Random Forest, Extra Trees, and Extreme Gradient Boosting Machine. The analysis identified that HDv, lobulation, and tumor growth site were independent risk factors (P < 0.05). We should focus on these three imaging features of tumors, which are relatively easy to obtain. The area under the curve for the SVM, KNN, RF, ET, and XGBoost prediction models were, respectively, 0.790, 0.895, 0.978, 0.988, and 0.946 for the training set, and were, respectively, 0.848, 0.892, 0.887, 0.912, and 0.867 for the test set. The CT combined with ML algorithms generated predictive models to improve the differential diagnosis of ≤ 5 cm GST and GS which has important clinical practical value. The Extra Trees algorithm resulted in the optimal model.

Role of MRI radiomics for the prediction of MYCN amplification in neuroblastomas

OBJECTIVES

We evaluate MR radiomics and develop machine learning-based classifiers to predict MYCN amplification in neuroblastomas.

METHODS

A total of 120 patients with neuroblastomas and baseline MR imaging examination available were identified of whom 74 (mean age ± standard deviation [SD] of 6 years and 2 months ± 4 years and 9 months; 43 females and 31 males, 14 MYCN amplified) underwent imaging at our institution. This was therefore used to develop radiomics models. The model was tested in a cohort of children with the same diagnosis but imaged elsewhere (n = 46, mean age ± SD: 5 years 11 months ± 3 years 9 months, 26 females and 14 MYCN amplified). Whole tumour volumes of interest were adopted to extract first-order histogram and second-order radiomics features. Interclass correlation coefficient and maximum relevance and minimum redundancy algorithm were applied for feature selection. Logistic regression, support vector machine, and random forest were employed as the classifiers. Receiver operating characteristic (ROC) analysis was performed to evaluate the diagnostic accuracy of the classifiers on the external test set.

RESULTS

The logistic regression model and the random forest both showed an AUC of 0.75. The support vector machine classifier obtained an AUC of 0.78 on the test set with a sensitivity of 64% and a specificity of 72%.

CONCLUSION

The study provides preliminary retrospective evidence demonstrating the feasibility of MRI radiomics in predicting MYCN amplification in neuroblastomas. Future studies are needed to explore the correlation between other imaging features and genetic markers and to develop multiclass predictive models.

KEY POINTS

• MYCN amplification in neuroblastomas is an important determinant of disease prognosis. • Radiomics analysis of pre-treatment MR examinations can be used to predict MYCN amplification in neuroblastomas. • Radiomics machine learning models showed good generalisability to external test set, demonstrating reproducibility of the computational models.

Radiomics analysis of contrast-enhanced computed tomography in predicting the International Neuroblastoma Pathology Classification in neuroblastoma

PURPOSE

To predict the International Neuroblastoma Pathology Classification (INPC) in neuroblastoma using a computed tomography (CT)-based radiomics approach.

METHODS

We enrolled 297 patients with neuroblastoma retrospectively and divided them into a training group (n = 208) and a testing group (n = 89). To balance the classes in the training group, a Synthetic Minority Over-sampling Technique was applied. A logistic regression radiomics model based on the radiomics features after dimensionality reduction was then constructed and validated in both the training and testing groups. To evaluate the diagnostic performance of the radiomics model, the receiver operating characteristic curve and calibration curve were utilized. Moreover, the decision curve analysis to assess the net benefits of the radiomics model at different high-risk thresholds was employed.

RESULTS

Seventeen radiomics features were used to construct radiomics model. In the training group, radiomics model achieved an area under the curve (AUC), accuracy, sensitivity, and specificity of 0.851 (95% confidence interval (CI) 0.805-0.897), 0.770, 0.694, and 0.847, respectively. In the testing group, radiomics model achieved an AUC, accuracy, sensitivity, and specificity of 0.816 (95% CI 0.725-0.906), 0.787, 0.793, and 0.778, respectively. The calibration curve indicated that the radiomics model was well fitted in both the training and testing groups (p > 0.05). Decision curve analysis further confirmed that the radiomics model performed well at different high-risk thresholds.

CONCLUSION

Radiomics analysis of contrast-enhanced CT demonstrates favorable diagnostic capabilities in distinguishing the INPC subgroups of neuroblastoma.

CRITICAL RELEVANCE STATEMENT

Radiomics features of contrast-enhanced CT images correlate with the International Neuroblastoma Pathology Classification (INPC) of neuroblastoma.

Development and validation of a CT-based radiomics signature for identifying high-risk neuroblastomas under the revised Children's Oncology Group classification system

BACKGROUND

To develop and validate a radiomics signature based on computed tomography (CT) for identifying high-risk neuroblastomas.

PROCEDURE

This retrospective study included 339 patients with neuroblastomas, who were classified into high-risk and non-high-risk groups according to the revised Children's Oncology Group classification system. These patients were then randomly divided into a training set (n = 237) and a testing set (n = 102). Pretherapy CT images of the arterial phase were segmented by two radiologists. Pyradiomics package and FeAture Explorer software were used to extract and process radiomics features. Radiomics models based on linear discriminant analysis (LDA), logistic regression (LR), and support vector machine (SVM) were constructed, and the area under the curve (AUC), 95% confidence interval (CI), and accuracy were calculated.

RESULTS

The optimal LDA, LR, and SVM models had 11, 12, and 14 radiomics features, respectively. The AUC of the LDA model in the training and testing sets were 0.877 (95% CI: 0.833-0.921) and 0.867 (95% CI: 0.797-0.937), with an accuracy of 0.823 and 0.804, respectively. The AUC of the LR model in the training and testing sets were 0.881 (95% CI: 0.839-0.924) and 0.855 (95% CI: 0.781-0.930), with an accuracy of 0.823 and 0.804, respectively. The AUC of the SVM model in the training and testing sets were 0.879 (95% CI: 0.836-0.923) and 0.862 (95% CI: 0.791-0.934), with an accuracy of 0.827 and 0.804, respectively.

CONCLUSIONS

CT-based radiomics is able to identify high-risk neuroblastomas and may provide additional image biomarkers for the identification of high-risk neuroblastomas.

Whole-tumor radiomics analysis of T2-weighted imaging in differentiating neuroblastoma from ganglioneuroblastoma/ganglioneuroma in children: an exploratory study

PURPOSE

To examine the potential of whole-tumor radiomics analysis of T2-weighted imaging (T2WI) in differentiating neuroblastoma (NB) from ganglioneuroblastoma/ganglioneuroma (GNB/GN) in children.

MATERIALS AND METHODS

This study included 102 children with peripheral neuroblastic tumors, comprising 47 NB patients and 55 GNB/GN patients, which were randomly divided into a training group (n = 72) and a test group (n = 30). Radiomics features were extracted from T2WI images, and feature dimensionality reduction was applied. Linear discriminant analysis was used to construct radiomics models, and one-standard error role combined with leave-one-out cross-validation was used to choose the optimal radiomics model with the least predictive error. Subsequently, the patient age at initial diagnosis and the selected radiomics features were incorporated to construct a combined model. The receiver operator characteristic (ROC) curve, decision curve analysis (DCA) and clinical impact curve (CIC) were applied to evaluate the diagnostic performance and clinical utility of the models.

RESULTS

Fifteen radiomics features were eventually chosen to construct the optimal radiomics model. The area under the curve (AUC) of the radiomics model in the training group and test group was 0.940 [95% confidence interval (CI) 0.886, 0.995] and 0.799 (95%CI 0.632, 0.966), respectively. The combined model, which incorporated patient age and radiomics features, achieved an AUC of 0.963 (95%CI 0.925, 1.000) in the training group and 0.871 (95%CI 0.744, 0.997) in the test group. DCA and CIC demonstrated that the radiomics model and combined model could provide benefits at various thresholds, with the combined model being superior to the radiomics model.

CONCLUSION

Radiomics features derived from T2WI, in combination with the age of the patient at initial diagnosis, may offer a quantitative method for distinguishing NB from GNB/GN, thus aiding in the pathological differentiation of peripheral neuroblastic tumors in children.

Contrast-enhanced computed tomography radiomics in predicting primary site response to neoadjuvant chemotherapy in high-risk neuroblastoma

PURPOSE

To explore the clinical value of contrast-enhanced computed tomography (CECT) radiomics in predicting primary site response to neoadjuvant chemotherapy in high-risk neuroblastoma.

MATERIALS AND METHODS

Seventy patients were retrospectively included and separated into very good partial response (VGPR) group and non-VGPR group according to the changes in primary tumor volume. The clinical features with statistical difference between the two groups were used to construct the clinical models using a logistic regression (LR) algorithm. The radiomics models based on different radiomics features selected by Kruskal-Wallis (KW) test and recursive feature elimination (RFE) were established using support vector machine (SVM) and LR algorithms. The radiomics score (Radscore) and clinical features were integrated into the combined models. Leave-one-out cross-validation (LOOCV) was used to validate the predictive performance of models in the entire dataset.

RESULTS

The optimal clinical model achieved an area under the curve (AUC) of 0.767 [95% confidence interval (CI): 0.638, 0.896] and an accuracy of 0.771 after LOOCV. The AUCs of the best KW + SVM, KW + LR, RFE + SVM, and RFE + LR radiomics models were 0.816, 0.826, 0.853, and 0.850, respectively, and the corresponding AUCs after LOOCV were 0.780, 0.785, 0.755, and 0.772, respectively. The AUC and accuracy after LOOCV of the optimal combined model was 0.804 (95% CI: 0.694, 0.915) and 0.814, respectively. The Delong test showed a statistical difference in predictive performance between the optimal clinical and combined models after LOOCV (Z = 2.003, P = 0.045). The decision curve analysis showed that the combined model performs better than the clinical model.

CONCLUSION

The CECT radiomics models have a favorable predictive performance in predicting VGPR of high-risk neuroblastoma to neoadjuvant chemotherapy. When integrating radiomics features and clinical features, the predictive performance of the combined models can be further improved.

Artificial intelligence assists precision medicine in cancer treatment

Cancer is a major medical problem worldwide. Due to its high heterogeneity, the use of the same drugs or surgical methods in patients with the same tumor may have different curative effects, leading to the need for more accurate treatment methods for tumors and personalized treatments for patients. The precise treatment of tumors is essential, which renders obtaining an in-depth understanding of the changes that tumors undergo urgent, including changes in their genes, proteins and cancer cell phenotypes, in order to develop targeted treatment strategies for patients. Artificial intelligence (AI) based on big data can extract the hidden patterns, important information, and corresponding knowledge behind the enormous amount of data. For example, the ML and deep learning of subsets of AI can be used to mine the deep-level information in genomics, transcriptomics, proteomics, radiomics, digital pathological images, and other data, which can make clinicians synthetically and comprehensively understand tumors. In addition, AI can find new biomarkers from data to assist tumor screening, detection, diagnosis, treatment and prognosis prediction, so as to providing the best treatment for individual patients and improving their clinical outcomes.