Deep learning radiomic nomogram to predict recurrence in soft tissue sarcoma: a multi-institutional study

Predicting progression-free survival in sarcoma using MRI-based automatic segmentation models and radiomics nomograms: a preliminary multicenter study

OBJECTIVES

Some sarcomas are highly malignant, associated with high recurrence despite treatment. This multicenter study aimed to develop and validate a radiomics signature to estimate sarcoma progression-free survival (PFS).

MATERIALS AND METHODS

The study retrospectively enrolled 202 consecutive patients with pathologically diagnosed sarcoma, who had pre-treatment axial fat-suppressed T2-weighted images (FS-T2WI), and included them in the ROI-Net model for training. Among them, 120 patients were included in the radiomics analysis, all of whom had pre-treatment axial T1-weighted and transverse FS-T2WI images, and were randomly divided into a development group (n = 96) and a validation group (n = 24). In the development cohort, Least Absolute Shrinkage and Selection Operator (LASSO) Cox regression was used to develop the radiomics features for PFS prediction. By combining significant clinical features with radiomics features, a nomogram was constructed using Cox regression.

RESULTS

The proposed ROI-Net framework achieved a Dice coefficient of 0.820 (0.791-0.848). The radiomics signature based on 21 features could distinguish high-risk patients with poor PFS. Univariate Cox analysis revealed that peritumoral edema, metastases, and the radiomics score were associated with poor PFS and were included in the construction of the nomogram. The Radiomics-T1WI-Clinical model exhibited the best performance, with AUC values of 0.947, 0.907, and 0.924 at 300 days, 600 days, and 900 days, respectively.

CONCLUSION

The proposed ROI-Net framework demonstrated high consistency between its segmentation results and expert annotations. The radiomics features and the combined nomogram have the potential to aid in predicting PFS for patients with sarcoma.

An MRI-based deep transfer learning radiomics nomogram for predicting meningioma grade

The aim of this study was to establish a nomogram based on clinical, radiomics, and deep transfer learning (DTL) features to predict meningioma grade. Three hundred forty meningiomas from one hospital composed the training set, and 102 meningiomas from another hospital composed the test set. The enhanced T1 WI images were used for analysis. The clinical, radiomics and DTL features were selected to construct the model. Radiomics and DTL scores were calculated. The deep transfer learning radiomics (DTLR) nomogram was developed on the basis of selected clinical features, radiomics scores and DTL scores. Receiver operating characteristic (ROC) curves and decision curve analysis (DCA) curves were drawn. The clinical features of sex, shape, indistinct margin and peritumoral edema were selected and used to construct the clinical model. Thirty-two radiomics features and 28 DTL features were selected for model construction. The clinical model had an AUC of 0.788. (95% CI: 0.6996-0.8756), with an accuracy of 0.745, a sensitivity of 0.941, and a specificity of 0.549 in the test set. The DTLR nomogram had the highest AUC of 0.866 (95% CI: 0.7984-0.9340), with an accuracy of 0.804, a sensitivity of 0.745, and a specificity of 0.863 in the test set. Compared with the other models, the DTLR nomogram had the greatest net benefit according to the DCA. There was a significant difference between the DTLR nomogram and the clinical model, no significant difference between the rest models in DeLong test.The DTLR nomogram has superior predictive value in DCA and could be a valuable method in clinical decision-making. Given the results of DeLong test, only the radiomics model is sufficient and there is no need to add DTL features. As a new attempt, the DTLR nomogram needs to be improved in the future study.

Radiomics model for predicting distant metastasis in soft tissue sarcoma of the extremities and trunk treated with surgery

PURPOSE

The aim of this study was to develop a radiomics model based on magnetic resonance imaging (MRI) for predicting metastasis in soft tissue sarcomas (STSs) treated with surgery.

METHODS/PATIENTS

MRI and clinical data of 73 patients with STSs of the extremities and trunk were obtained from TCIA database and Jiangsu Cancer Hospital as the training set, data of other 40 patients were retrospectively collected at our institution as the external validation set. Radiomics features were extracted from both intratumoral and peritumoral regions of fat-suppressed T2-weighted images (FS-T2WIs) of patients, and 3D ResNet10 was used to extract deep learning features. Recursive feature elimination (RFE) and least absolute shrinkage and selection operator (LASSO) algorithms were used for the selection of features. Based on 4 different sets of features, 5 machine learning algorithms were used to construct intratumor, peritumor, combined intratumor and peritumor radiomics models and deep learning radiomics (DLR) model. The area under the ROC curve (AUC) and Decision curve analysis (DCA) were used to evaluate the ability of models to predict metastasis.

RESULTS AND CONCLUSIONS

Based on 20 selected features from the deep-learning and radiomics features set, the DLR model was able to predict metastasis in the validation dataset, with an AUC of 0.9770. The DCA and Hosmer-Lemeshow test revealed that the DLR model had good clinical benefit and consistency. By getting richer information from MRI, The DLR model is a noninvasive, low-cost method for predicting the risk of metastasis in STSs, and can help develop appropriate treatment programs.

Radiomics and Deep Learning Model for Benign and Malignant Soft Tissue Tumors Differentiation of Extremities and Trunk

RATIONALE AND OBJECTIVES

To develop radiomics and deep learning models for differentiating malignant and benign soft tissue tumors (STTs) preoperatively based on fat saturation T2-weighted imaging (FS-T2WI) of patients.

MATERIALS AND METHODS

Data of 115 patients with STTs of extremities and trunk were collected from our hospital as the training set, and data of other 70 patients were collected from another center as the external validation set. Outlined Regions of interest included the intratumor and the peritumor region extending outward by 5 mm, then the corresponding radiomics features were extracted respectively. Deep learning was performed using pretrained 3D ResNet algorithms, and deep learning features were extracted from the entire FS-T2WI of patients. Recursive feature elimination and least absolute shrinkage and selection operator were used to select the radiomics and deep learning features with predictive value. Five machine learning algorithms were applied to build radiomics models, the area under the ROC curve (AUC) in the validation set were used to evaluate the diagnostic performance, and decision curve analysis (DCA) was used to evaluate the clinical benefit of models.

RESULTS

Based on 20 selected deep learning and radiomics features, the deep learning radiomics (DLR) model had the best predictive performance in the validation set, with an AUC of 0.9410. DCA and calibration curves showed that the DLR model had better clinical net benefit and goodness of fit.

CONCLUSION

By extracting more features from FS-T2WI, the DLR model is a noninvasive, low-cost, and highly accurate preoperative differential diagnosis of benign and malignant STTs.

Interpretable multimodal deep learning model for predicting post-surgical international society of urological pathology grade in primary prostate cancer

PURPOSE

To address heterogeneity in prostate cancer (PCa) pathological grading, we developed an interpretable multimodal fusion model integrating 18F prostate-specific membrane antigen (18F-PSMA)-targeted positron emission tomography/computed tomography (18F-PSMA-PET/CT) imaging features with clinical variables for predicting post-surgical ISUP grade (psISUP ≥ 4 vs. < 4).

METHODS

This retrospective study analyzed 222 patients with PCa (2020-2024) undergoing 18F-PSMA PET/CT. We constructed a deep transfer learning framework incorporating radiomic features from PET/CT and clinical parameters. Model performance was validated against three established methods and preoperative biopsy Gleason scores. Additionally, SHapley Additive exPlanations (SHAP) values elucidated feature contributions, and a radiomic nomogram was developed for clinical translation.

RESULTS

The fusion model achieved superior discrimination in psISUP grading (test set area under the curve (AUC) = 0.850, 95% confidence interval [CI] 0.769-0.932; validation set AUC = 0.833, 95% CI 0.657-1.000), significantly outperforming preoperative Gleason scores. SHAP analysis identified PSMA uptake heterogeneity and PSA density as key predictive features. The nomogram demonstrated clinical interpretability through visualised risk stratification.

CONCLUSION

Our deep learning-based multimodal fusion model enables accurate preoperative prediction of aggressive PCa pathology (ISUP ≥ 4), potentially optimising surgical planning and personalised therapeutic strategies. The interpretable framework enhances clinical trustworthiness in artificial intelligence-assisted decision-making.

Diagnostic utility of MRI-based convolutional neural networks in soft tissue sarcomas: a mini-review

Purpose

This review assesses the diagnostic performance of MRI-based convolutional neural networks for identifying and grading soft tissue sarcomas, evaluating therapy responses, and assessing the risk for metastases and recurrences.

Methods

Electronic databases, specifically PubMed/MEDLINE and Google Scholar, were diligently scoured for studies that delved into the intersection of convolutional neural networks, soft tissue sarcomas, and MRI. Three topics were included: 1) differentiating and grading soft tissue sarcomas, 2) assessing therapy response, and 3) predicting metastases and recurrences.

Results

This review included 12 articles. Seven articles investigated the differentiation and grading of soft tissue sarcomas. Sensitivity for that issue ranged from 0.85 to 0.95, specificity from 0,33 to 1, and the area under the curve (AUC) from 0.74 to 0.96. Three articles investigated therapy responses, and two discussed metastasis and recurrence prediction. Only one article out of the five articles above presented accurate diagnostic values. That article examined the prediction of lung metastases and demonstrated a sensitivity of 0.47, a specificity of 0.97, and an AUC of 0.83.

Conclusion

AI applications using CNNs demonstrated robust capabilities in differentiating and grading soft tissue sarcomas using MRI. However, studies on therapy response and prediction of metastases and recurrences are still lacking.

Identification of histopathological classification and establishment of prognostic indicators of gastric adenocarcinoma based on deep learning algorithm

The aim of this study is to establish a deep learning (DL) model to predict the pathological type of gastric adenocarcinoma cancer based on whole-slide images(WSIs). We downloaded 356 histopathological images of gastric adenocarcinoma (STAD) patients from The Cancer Genome Atlas database and randomly divided them into the training set, validation set and test set (8:1:1). Additionally, 80 H&E-stained WSIs of STAD were collected for external validation. The CLAM tool was used to cut the WSIs and further construct the model by DL algorithm, achieving an accuracy of over 90% in identifying and predicting histopathological subtypes. External validation results demonstrated the model had a certain generalization ability. Moreover, DL features were extracted from the model to further investigate the differences in immune infiltration and patient prognosis between the two subtypes. The DL model can accurately predict the pathological classification of STAD patients, and provide certain reference value for clinical diagnosis. The nomogram combining DL-signature, gene-signature and clinical features can be used as a prognostic classifier for clinical decision-making and treatment.

Role of radiomics as a predictor of disease recurrence in ovarian cancer: a systematic review

Ovarian cancer is associated with high cancer-related mortality rate attributed to late-stage diagnosis, limited treatment options, and frequent disease recurrence. As a result, careful patient selection is important especially in setting of radical surgery. Radiomics is an emerging field in medical imaging, which may help provide vital prognostic evaluation and help patient selection for radical treatment strategies. This systematic review aims to assess the role of radiomics as a predictor of disease recurrence in ovarian cancer. A systematic search was conducted in Medline, EMBASE, and Web of Science databases. Studies meeting inclusion criteria investigating the use of radiomics to predict post-operative recurrence in ovarian cancer were included in our qualitative analysis. Study quality was assessed using the QUADAS-2 and Radiomics Quality Score tools. Six retrospective studies met the inclusion criteria, involving a total of 952 participants. Radiomic-based signatures demonstrated consistent performance in predicting disease recurrence, as evidenced by satisfactory area under the receiver operating characteristic curve values (AUC range 0.77-0.89). Radiomic-based signatures appear to good prognosticators of disease recurrence in ovarian cancer as estimated by AUC. The reviewed studies consistently reported the potential of radiomic features to enhance risk stratification and personalise treatment decisions in this complex cohort of patients. Further research is warranted to address limitations related to feature reliability, workflow heterogeneity, and the need for prospective validation studies.

Radiomics in Diagnosis, Grading, and Treatment Response Assessment of Soft Tissue Sarcomas: A Systematic Review and Meta-analysis

RATIONALE AND OBJECTIVES

To evaluate radiomics in soft tissue sarcomas (STSs) for diagnostic accuracy, grading, and treatment response assessment, with a focus on clinical relevance.

METHODS

In this diagnostic accuracy study, radiomics was applied using multiple MRI sequences and AI classifiers, with histopathological diagnosis as the reference standard. Statistical analysis involved meta-analysis, random-effects model, and Deeks' funnel plot asymmetry test.

RESULTS

Among 579 unique titles and abstracts, 24 articles were included in the systematic review, with 21 used for meta-analysis. Radiomics demonstrated a pooled sensitivity of 84% (95% CI: 80-87) and specificity of 63% (95% CI: 56-70), AUC of 0.93 for diagnosis, sensitivity of 84% (95% CI: 82-87) and specificity of 73% (95% CI: 68-77), AUC of 0.91 for grading, and sensitivity of 83% (95% CI: 67-94) and specificity of 67% (95% CI: 59-74), AUC of 0.87 for treatment response assessment.

CONCLUSION

Radiomics exhibits potential for accurate diagnosis, grading, and treatment response assessment in STSs, emphasizing the need for standardization and prospective trials.

CLINICAL RELEVANCE STATEMENT

Radiomics offers precise tools for STS diagnosis, grading, and treatment response assessment, with implications for optimizing patient care and treatment strategies in this complex malignancy.

A multimodal neural network with gradient blending improves predictions of survival and metastasis in sarcoma

The objective of this study is to develop a multimodal neural network (MMNN) model that analyzes clinical variables and MRI images of a soft tissue sarcoma (STS) patient, to predict overall survival and risk of distant metastases. We compare the performance of this MMNN to models based on clinical variables alone, radiomics models, and an unimodal neural network. We include patients aged 18 or older with biopsy-proven STS who underwent primary resection between January 1st, 2005, and December 31st, 2020 with complete outcome data and a pre-treatment MRI with both a T1 post-contrast sequence and a T2 fat-sat sequence available. A total of 9380 MRI slices containing sarcomas from 287 patients are available. Our MMNN accepts the entire 3D sarcoma volume from T1 and T2 MRIs and clinical variables. Gradient blending allows the clinical and image sub-networks to optimally converge without overfitting. Heat maps were generated to visualize the salient image features. Our MMNN outperformed all other models in predicting overall survival and the risk of distant metastases. The C-Index of our MMNN for overall survival is 0.77 and the C-Index for risk of distant metastases is 0.70. The provided heat maps demonstrate areas of sarcomas deemed most salient for predictions. Our multimodal neural network with gradient blending improves predictions of overall survival and risk of distant metastases in patients with soft tissue sarcoma. Future work enabling accurate subtype-specific predictions will likely utilize similar end-to-end multimodal neural network architecture and require prospective curation of high-quality data, the inclusion of genomic data, and the involvement of multiple centers through federated learning.

Predicting bone metastasis-free survival in non-small cell lung cancer from preoperative CT via deep learning

Accurate prediction of bone metastasis-free survival (BMFS) after complete surgical resection in patients with non-small cell lung cancer (NSCLC) may facilitate appropriate follow-up planning. The aim of this study was to establish and validate a preoperative CT-based deep learning (DL) signature to predict BMFS in NSCLC patients. We performed a retrospective analysis of 1547 NSCLC patients who underwent complete surgical resection, followed by at least 36 months of monitoring at two hospitals. We constructed a DL signature from multiparametric CT images using 3D convolutional neural networks, and we integrated this signature with clinical-imaging factors to establish a deep learning clinical-imaging signature (DLCS). We evaluated performance using Harrell's concordance index (C-index) and the time-dependent receiver operating characteristic. We also assessed the risk of bone metastasis (BM) in NSCLC patients at different clinical stages using DLCS. The DL signature successfully predicted BM, with C-indexes of 0.799 and 0.818 for the validation cohorts. DLCS outperformed the DL signature with corresponding C-indexes of 0.806 and 0.834. Ranges for area under the curve at 1, 2, and 3 years were 0.820-0.865 for internal and 0.860-0.884 for external validation cohorts. Furthermore, DLCS successfully stratified patients with different clinical stages of NSCLC as high- and low-risk groups for BM (p < 0.05). CT-based DL can predict BMFS in NSCLC patients undergoing complete surgical resection, and may assist in the assessment of BM risk for patients at different clinical stages.

Omics-imaging signature-based nomogram to predict the progression-free survival of patients with hepatocellular carcinoma after transcatheter arterial chemoembolization

BACKGROUND

Enhanced magnetic resonance imaging (MRI) is widely used in the diagnosis, treatment and prognosis of hepatocellular carcinoma (HCC), but it can not effectively reflect the heterogeneity within the tumor and evaluate the effect after treatment. Preoperative imaging analysis of voxel changes can effectively reflect the internal heterogeneity of the tumor and evaluate the progression-free survival (PFS).

AIM

To predict the PFS of patients with HCC before operation by building a model with enhanced MRI images.

METHODS

Delineate the regions of interest (ROI) in arterial phase, portal venous phase and delayed phase of enhanced MRI. After extracting the combinatorial features of ROI, the features are fused to obtain deep learning radiomics (DLR)_Sig. DeLong's test was used to evaluate the diagnostic performance of different typological features. K-M analysis was applied to assess PFS in different risk groups, and the discriminative ability of the model was evaluated using the C-index.

RESULTS

Tumor diameter and diolame were independent factors influencing the prognosis of PFS. Delong's test revealed multi-phase combined radiomic features had significantly greater area under the curve values than did those of the individual phases (P < 0.05).In deep transfer learning (DTL) and DLR, significant differences were observed between the multi-phase and individual phases feature sets (P < 0.05). K-M survival analysis revealed a median survival time of high risk group and low risk group was 12.8 and 14.2 months, respectively, and the predicted probabilities of 6 months, 1 year and 2 years were 92%, 60%, 40% and 98%, 90%,73%, respectively. The C-index was 0.764, indicating relatively good consistency between the predicted and observed results. DTL and DLR have higher predictive value for 2-year PFS in nomogram.

CONCLUSION

Based on the multi-temporal characteristics of enhanced MRI and the constructed Nomograph, it provides a new strategy for predicting the PFS of transarterial chemoembolization treatment of HCC.

Deep Learning Radiomics Nomogram Based on Magnetic Resonance Imaging for Differentiating Type I/II Epithelial Ovarian Cancer

RATIONALE AND OBJECTIVES

To develop and validate a T2-weighted magnetic resonance imaging (MRI)-based deep learning radiomics nomogram (DLRN) to differentiate between type I and type II epithelial ovarian cancer (EOC).

MATERIALS AND METHODS

This multicenter study incorporated 437 patients from five centers, divided into training (n = 271), internal validation (n = 68), and external validation (n = 98) sets. The deep learning (DL) model was constructed using the largest orthogonal slices of the tumor area. The extracted radiomics features were employed in building the radiomics model. The clinical model was developed based on clinical characteristics. A DLRN was built by integrating the DL signature, radiomics signature, and independent clinical predictors. Model performances were evaluated through receiver operating characteristic (ROC) analysis, Brier score, calibration curve, and decision curve analysis (DCA). The areas under the ROC curve (AUCs) were compared using the DeLong test. A two-tailed P < 0.05 was considered significantly different.

RESULTS

The DLRN exhibited satisfactory discrimination between type I and type II EOC with the AUC of 0.888 (95% confidence interval [CI] 0.810, 0.966) and 0.866 (95% CI 0.786, 0.946) in the internal and external validation sets, respectively. These AUCs significantly exceeded those of the clinical model (P = 0.013 and 0.043, in the internal and external validation sets, respectively). The DLRN demonstrated optimal classification accuracy and clinical application value, according to Brier scores, calibration curves, and DCA.

CONCLUSION

A T2-weighted MRI-based DLRN showed promising potential in differentiating between type I and type II EOC, which could offer assistance in clinical decision-making.

Multi-institutional validation of a radiomics signature for identification of postoperative progression of soft tissue sarcoma

BACKGROUND

To develop a magnetic resonance imaging (MRI)-based radiomics signature for evaluating the risk of soft tissue sarcoma (STS) disease progression.

METHODS

We retrospectively enrolled 335 patients with STS (training, validation, and The Cancer Imaging Archive sets, n = 168, n = 123, and n = 44, respectively) who underwent surgical resection. Regions of interest were manually delineated using two MRI sequences. Among 12 machine learning-predicted signatures, the best signature was selected, and its prediction score was inputted into Cox regression analysis to build the radiomics signature. A nomogram was created by combining the radiomics signature with a clinical model constructed using MRI and clinical features. Progression-free survival was analyzed in all patients. We assessed performance and clinical utility of the models with reference to the time-dependent receiver operating characteristic curve, area under the curve, concordance index, integrated Brier score, decision curve analysis.

RESULTS

For the combined features subset, the minimum redundancy maximum relevance-least absolute shrinkage and selection operator regression algorithm + decision tree classifier had the best prediction performance. The radiomics signature based on the optimal machine learning-predicted signature, and built using Cox regression analysis, had greater prognostic capability and lower error than the nomogram and clinical model (concordance index, 0.758 and 0.812; area under the curve, 0.724 and 0.757; integrated Brier score, 0.080 and 0.143, in the validation and The Cancer Imaging Archive sets, respectively). The optimal cutoff was - 0.03 and cumulative risk rates were calculated.

DATA CONCLUSION

To assess the risk of STS progression, the radiomics signature may have better prognostic power than a nomogram/clinical model.

Deep learning radiomics-based prediction model of metachronous distant metastasis following curative resection for retroperitoneal leiomyosarcoma: a bicentric study

BACKGROUND

Combining conventional radiomics models with deep learning features can result in superior performance in predicting the prognosis of patients with tumors; however, this approach has never been evaluated for the prediction of metachronous distant metastasis (MDM) among patients with retroperitoneal leiomyosarcoma (RLS). Thus, the purpose of this study was to develop and validate a preoperative contrast-enhanced computed tomography (CECT)-based deep learning radiomics model for predicting the occurrence of MDM in patients with RLS undergoing complete surgical resection.

METHODS

A total of 179 patients who had undergone surgery for the treatment of histologically confirmed RLS were retrospectively recruited from two tertiary sarcoma centers. Semantic segmentation features derived from a convolutional neural network deep learning model as well as conventional hand-crafted radiomics features were extracted from preoperative three-phase CECT images to quantify the sarcoma phenotypes. A conventional radiomics signature (RS) and a deep learning radiomics signature (DLRS) that incorporated hand-crafted radiomics and deep learning features were developed to predict the risk of MDM. Additionally, a deep learning radiomics nomogram (DLRN) was established to evaluate the incremental prognostic significance of the DLRS in combination with clinico-radiological predictors.

RESULTS

The comparison of the area under the curve (AUC) values in the external validation set, as determined by the DeLong test, demonstrated that the integrated DLRN, DLRS, and RS models all exhibited superior predictive performance compared with that of the clinical model (AUC 0.786 [95% confidence interval 0.649-0.923] vs. 0.822 [0.692-0.952] vs. 0.733 [0.573-0.892] vs. 0.511 [0.359-0.662]; both P < 0.05). The decision curve analyses graphically indicated that utilizing the DLRN for risk stratification provided greater net benefits than those achieved using the DLRS, RS and clinical models. Good alignment with the calibration curve indicated that the DLRN also exhibited good performance.

CONCLUSIONS

The novel CECT-based DLRN developed in this study demonstrated promising performance in the preoperative prediction of the risk of MDM following curative resection in patients with RLS. The DLRN, which outperformed the other three models, could provide valuable information for predicting surgical efficacy and tailoring individualized treatment plans in this patient population.

TRIAL REGISTRATION

Not applicable.

The impact of radiomics in the management of soft tissue sarcoma

INTRODUCTION

Soft tissue sarcomas (STSs) are rare malignancies. Pre-therapeutic tumour grading and assessment are crucial in making treatment decisions. Radiomics is a high-throughput method for analysing imaging data, providing quantitative information beyond expert assessment. This review highlights the role of radiomic texture analysis in STSs evaluation.

MATERIALS AND METHODS

We conducted a systematic review according to the Systematic Reviews and Meta-Analyses (PRISMA) guidelines. A comprehensive search was conducted in PubMed/MEDLINE and Scopus using the search terms: 'radiomics [All Fields] AND ("soft tissue sarcoma" [All Fields] OR "soft tissue sarcomas" [All Fields])'. Only original articles, referring to humans, were included.

RESULTS

A preliminary search conducted on PubMed/MEDLINE and Scopus provided 74 and 93 studies respectively. Based on the previously described criteria, 49 papers were selected, with a publication range from July 2015 to June 2023. The main domains of interest were risk stratification, histological grading prediction, technical feasibility/reproductive aspects, treatment response.

CONCLUSIONS

With an increasing interest over the last years, the use of radiomics appears to have potential for assessing STSs from initial diagnosis to predicting treatment response. However, additional and extensive research is necessary to validate the effectiveness of radiomics parameters and to integrate them into a comprehensive decision support system.

CT and MRI radiomics of bone and soft-tissue sarcomas: an updated systematic review of reproducibility and validation strategies

OBJECTIVE

To systematically review radiomic feature reproducibility and model validation strategies in recent studies dealing with CT and MRI radiomics of bone and soft-tissue sarcomas, thus updating a previous version of this review which included studies published up to 2020.

METHODS

A literature search was conducted on EMBASE and PubMed databases for papers published between January 2021 and March 2023. Data regarding radiomic feature reproducibility and model validation strategies were extracted and analyzed.

RESULTS

Out of 201 identified papers, 55 were included. They dealt with radiomics of bone (n = 23) or soft-tissue (n = 32) tumors. Thirty-two (out of 54 employing manual or semiautomatic segmentation, 59%) studies included a feature reproducibility analysis. Reproducibility was assessed based on intra/interobserver segmentation variability in 30 (55%) and geometrical transformations of the region of interest in 2 (4%) studies. At least one machine learning validation technique was used for model development in 34 (62%) papers, and K-fold cross-validation was employed most frequently. A clinical validation of the model was reported in 38 (69%) papers. It was performed using a separate dataset from the primary institution (internal test) in 22 (40%), an independent dataset from another institution (external test) in 14 (25%) and both in 2 (4%) studies.

CONCLUSIONS

Compared to papers published up to 2020, a clear improvement was noted with almost double publications reporting methodological aspects related to reproducibility and validation. Larger multicenter investigations including external clinical validation and the publication of databases in open-access repositories could further improve methodology and bring radiomics from a research area to the clinical stage.

CRITICAL RELEVANCE STATEMENT

An improvement in feature reproducibility and model validation strategies has been shown in this updated systematic review on radiomics of bone and soft-tissue sarcomas, highlighting efforts to enhance methodology and bring radiomics from a research area to the clinical stage.

KEY POINTS

• 2021-2023 radiomic studies on CT and MRI of musculoskeletal sarcomas were reviewed. • Feature reproducibility was assessed in more than half (59%) of the studies. • Model clinical validation was performed in 69% of the studies. • Internal (44%) and/or external (29%) test datasets were employed for clinical validation.

A CT-based Deep Learning Radiomics Nomogram for the Prediction of EGFR Mutation Status in Head and Neck Squamous Cell Carcinoma

RATIONALE AND OBJECTIVES

Accurately assessing epidermal growth factor receptor (EGFR) mutation status in head and neck squamous cell carcinoma (HNSCC) patients is crucial for prognosis and treatment selection. This study aimed to construct and validate a contrast-enhanced computed tomography (CECT)-based deep learning radiomics nomogram (DLRN) to predict EGFR mutation status of HNSCC.

MATERIALS AND METHODS

A total of 300 HNSCC patients who underwent CECT scans were enrolled in this study. Participants from two hospitals were separated into a training set (n = 200, 56 EGFR-negative and 144 EGFR-positive) from one hospital and an external test set from the other hospital (n = 100, 37 EGFR-negative and 63 EGFR-positive). The least absolute shrinkage and selection operator method was used to select the key features from CECT-based manually extracted radiomics (MER) features and features automatically extracted using a deep learning model (DL, extracted using a GoogLeNet model). The selected independent clinical factors, MER features, and DL features were then combined to construct a DLRN. The DLRN's performance was evaluated using receiver operating characteristics curves.

RESULTS

Five MER and six DL features were finally chosen. The DLRN, which includes "gender" and "necrotic areas," along with the selected features, predicted EGFR mutation status of HNSCC (EGFR-negative vs. positive) well in both the training (area under the curve [AUC], 0.901) and test (AUC, 0.875) sets.

CONCLUSION

A DLRN using CECT was built to predict EGFR mutation in HNSCC. The model showed high predictive ability and may aid in treatment selection and patient prognosis.

Radiomics in Musculoskeletal Tumors

Sarcomas are heterogeneous rare tumors predominantly affecting the musculoskeletal (MSK) system. Due to significant variations in their natural history and variable response to conventional treatments, the discovery of novel diagnostic and prognostic biomarkers to guide therapeutic decision-making is an active and ongoing field of research. As new cellular, molecular, and metabolic biomarkers continue to be discovered, quantitative radiologic imaging is becoming increasingly important in sarcoma management. Radiomics offers the potential for discovering novel imaging diagnostic and predictive biomarkers using standard-of-care medical imaging. In this review, we detail the core concepts of radiomics and the application of radiomics to date in MSK sarcoma research. Also described are specific challenges related to radiomic studies, as well as viewpoints on clinical adoption and future perspectives in the field.

Deep Learning Radiomics Model of Dynamic Contrast-Enhanced MRI for Evaluating Vessels Encapsulating Tumor Clusters and Prognosis in Hepatocellular Carcinoma

BACKGROUND

Vessels encapsulating tumor cluster (VETC) is a critical prognostic factor and therapeutic predictor of hepatocellular carcinoma (HCC). However, noninvasive evaluation of VETC remains challenging.

PURPOSE

To develop and validate a deep learning radiomic (DLR) model of dynamic contrast-enhanced MRI (DCE-MRI) for the preoperative discrimination of VETC and prognosis of HCC.

STUDY TYPE

Retrospective.

POPULATION

A total of 221 patients with histologically confirmed HCC and stratified this cohort into training set (n = 154) and time-independent validation set (n = 67).

FIELD STRENGTH/SEQUENCE

A 1.5 T and 3.0 T; DCE imaging with T1-weighted three-dimensional fast spoiled gradient echo.

ASSESSMENT

Histological specimens were used to evaluate VETC status. VETC+ cases had a visible pattern (≥5% tumor area), while cases without any pattern were VETC-. The regions of intratumor and peritumor were segmented manually in the arterial, portal-venous and delayed phase (AP, PP, and DP, respectively) of DCE-MRI and reproducibility of segmentation was evaluated. Deep neural network and machine learning (ML) classifiers (logistic regression, decision tree, random forest, SVM, KNN, and Bayes) were used to develop nine DLR, 54 ML and clinical-radiological (CR) models based on AP, PP, and DP of DCE-MRI for evaluating VETC status and association with recurrence.

STATISTICAL TESTS

The Fleiss kappa, intraclass correlation coefficient, receiver operating characteristic curve, area under the curve (AUC), Delong test and Kaplan-Meier survival analysis. P value <0.05 was considered as statistical significance.

RESULTS

Pathological VETC+ were confirmed in 68 patients (training set: 46, validation set: 22). In the validation set, DLR model based on peritumor PP (peri-PP) phase had the best performance (AUC: 0.844) in comparison to CR (AUC: 0.591) and ML (AUC: 0.672) models. Significant differences in recurrence rates between peri-PP DLR model-predicted VETC+ and VETC- status were found.

DATA CONCLUSIONS

The DLR model provides a noninvasive method to discriminate VETC status and prognosis of HCC patients preoperatively.

EVIDENCE LEVEL

4.

TECHNICAL EFFICACY

Stage 2.

Impact of Surveillance Imaging in Detecting Local and Metastatic Lung Recurrences Among Patients with Sarcomas of the Extremities: A Systematic Review and Meta-analysis

BACKGROUND

The surveillance guidelines following treatment completion for patients with high-grade sarcomas of the extremities are based largely upon expert opinions and consensus. In the current meta-analysis, we aim to study the utility of surveillance imaging to diagnose local and metastatic pulmonary relapses among patients with extremity soft tissue sarcomas and primary bone sarcomas.

PATIENTS AND METHODS

A meta-analysis was performed to assess the sensitivity, specificity and diagnostic odds ratio (DOR) of surveillance imaging to diagnose local and metastatic pulmonary relapse among patients with sarcoma of the extremities. In addition, impact of surveillance imaging on overall survival was assessed. Heterogeneity among eligible studies was evaluated by I2 statistics. Sensitivity analysis was assessed using influence plots and Baujat plots.

RESULTS

Ten studies including 2160 patients with sarcoma were found eligible. For diagnoses of local recurrence based on surveillance imaging (nine studies, 1917 patients), the estimated sensitivity, specificity, and DOR were 13.6%, 99.5%, and 78.15, respectively. Only 16.7% of local recurrences were diagnosed based on imaging. For diagnoses of metastatic pulmonary recurrence (eight studies; 1868 patients), estimated sensitivity, specificity, and DOR were 76.1%, 99.3%, and 1059.9, respectively. A sensitivity analysis showed significant heterogeneity among included studies. None of the included studies showed an overall-survival benefit with the use of surveillance imaging.

CONCLUSION

The current meta-analysis challenges the notion of routine use of imaging to detect local relapse, while favoring chest imaging, using either chest radiography or computed tomography scan, for surveillance. Further studies are required to study the ideal surveillance strategy including timing and imaging modality.

Radiomics and artificial intelligence for soft-tissue sarcomas: Current status and perspectives

This article proposes a summary of the current status of the research regarding the use of radiomics and artificial intelligence to improve the radiological assessment of patients with soft tissue sarcomas (STS), a heterogeneous group of rare and ubiquitous mesenchymal malignancies. After a first part explaining the principle of radiomics approaches, from raw image post-processing to extraction of radiomics features mined with unsupervised and supervised machine-learning algorithms, and the current research involving deep learning algorithms in STS, especially convolutional neural networks, this review details their main research developments since the formalisation of 'radiomics' in oncologic imaging in 2010. This review focuses on CT and MRI and does not involve ultrasonography. Radiomics and deep radiomics have been successfully applied to develop predictive models to discriminate between benign soft-tissue tumors and STS, to predict the histologic grade (i.e., the most important prognostic marker of STS), the response to neoadjuvant chemotherapy and/or radiotherapy, and the patients' survivals and probability for presenting distant metastases. The main findings, limitations and expectations are discussed for each of these outcomes. Overall, after a first decade of publications emphasizing the potential of radiomics through retrospective proof-of-concept studies, almost all positive but with heterogeneous and often non-replicable methods, radiomics is now at a turning point in order to provide robust demonstrations of its clinical impact through open-science, independent databases, and application of good and standardized practices in radiomics such as those provided by the Image Biomarker Standardization Initiative, without forgetting innovative research paths involving other '-omics' data to better understand the relationships between imaging of STS, gene-expression profiles and tumor microenvironment.

Evaluation of Lymphatic Vessel Invasion Determined by D2-40 Using Preoperative MRI-Based Radiomics for Invasive Breast Cancer

RATIONALE AND OBJECTIVES

Preoperative prediction of LVI status can facilitate personalized therapeutic planning. This study aims to investigate the efficacy of preoperative MRI-based radiomics for predicting lymphatic vessel invasion (LVI) determined by D2-40 in patients with invasive breast cancer.

MATERIALS AND METHODS

A total of 203 patients with pathologically confirmed invasive breast cancer, who underwent preoperative breast MRI, were retrospectively enrolled and randomly assigned to the following cohorts: training cohort (n=141) and test cohort (n=62). Then, univariate and multivariate logistic regression were performed to select independent risk factors and build a clinical model. Afterwards, least absolute shrinkage and selection operator (LASSO) logistic regression was performed to select predictive features extracted from the early and delay enhancement dynamic contrast-enhanced (DCE)-MRI images, and a radiomics signature was established. Subsequently, a nomogram model was constructed by incorporating the radiomics score and risk factors. Receiver operating characteristic curves were performed to determine the performance of various models. The efficacy of the various models was evaluated using calibration and decision curves.

RESULTS

Fourteen radiomics features were selected to construct the radiomics model. The size of the lymph node was identified as an independent risk factor of the clinical model. The nomogram model demonstrated the best calibration and discrimination performance in both the training and test cohorts, with an area under the curve of 0.873 (95% confidence interval [CI]: 0.807-0.923) and 0.902 (95% CI: 0.800-0.963), respectively. The decision curve illustrated that the nomogram model added more net benefits, when compared to the radiomics signature and clinical model.

CONCLUSION

The nomogram model based on preoperative DCE-MRI images exhibits satisfactory efficacy for the noninvasive prediction of LVI determined by D2-40 in invasive breast cancer.

Deep learning-based radiomic nomogram to predict risk categorization of thymic epithelial tumors: A multicenter study

PURPOSE

The study was aimed to develop and evaluate a deep learning-based radiomics to predict the histological risk categorization of thymic epithelial tumors (TETs), which can be highly informative for patient treatment planning and prognostic assessment.

METHOD

A total of 681 patients with TETs from three independent hospitals were included and separated into derivation cohort and external test cohort. Handcrafted and deep learning features were extracted from preoperative contrast-enhanced CT images and selected to build three radiomics signatures (radiomics signature [Rad_Sig], deep learning signature [DL_Sig] and deep learning radiomics signature [DLR_Sig]) to predict risk categorization of TETs. A deep learning-based radiomic nomogram (DLRN) was then depicted to visualize the classification evaluation. The performance of predictive models was compared using the receiver operating characteristic and decision curve analysis (DCA).

RESULTS

Among three radiomics signatures, DLR_Sig demonstrated optimum performance with an AUC of 0.883 for the derivation cohort and 0.749 for the external test cohort. Combining DLR_Sig with age and gender, DLRN was depict and exhibited optimum performance among all radiomics models with an AUC of 0.965, accuracy of 0.911, sensitivity of 0.921 and specificity of 0.902 in the derivation cohort, and an AUC of 0.786, accuracy of 0.774, sensitivity of 0.778 and specificity of 0.771 in the external test cohort. The DCA showed that DLRN had greater clinical benefit than other radiomics signatures.

CONCLUSIONS

Our study developed and validated a DLRN to accurately predict the risk categorization of TETs, which has potential to facilitate individualized treatment and improve patient prognosis evaluation.

Machine learning-based radiomics analysis for predicting local recurrence of primary dermatofibrosarcoma protuberans after surgical treatment

BACKGROUND AND PURPOSE

Dermatofibrosarcoma protuberans (DFSP) is characterized by locally invasive growth patterns and high local recurrence rates. Accurately identifying patients with high local recurrence risk may benefit patients during follow-up and has potential value for making treatment decisions. This study aimed to investigate whether machine learning-based radiomics models could accurately predict the local recurrence of primary DFSP after surgical treatment.

MATERIALS AND METHODS

This retrospective study included a total of 146 patients with DFSP who underwent MRI scans between 2010 and 2016 from two different institutions: institution 1 (n = 104) for the training set and institution 2 (n = 42) for the external test set. Three radiomics random survival forest (RSF) models were developed using MRI images. Additionally, the performance of the Ki67 index was compared with the three RSF models in the external validation set.

RESULTS

The average concordance index (C-index) scores of the RSF models based on fat-saturation T2W (FS-T2W) images, fat-saturation T1W with gadolinium contrast (FS-T1W + C) images, and both FS-T2W and FS-T1W + C images from 10-fold cross-validation in the training set were 0.855 (95% CI: 0.629, 1.00), 0.873 (95% CI: 0.711, 1.00), and 0.875 (95% CI: 0.688, 1.00), respectively. In the external validation set, the C-indexes of the three trained RSF models were higher than that of the Ki67 index (0.838, 0.754, and 0.866 vs. 0.601, respectively).

CONCLUSION

Random survival forest models developed using radiomics features derived from MRI images were proven helpful for accurate prediction of local recurrence of primary DFSP after surgical treatment and showed better predicting performance than the Ki67 index.

An MRI-Based Radiomics Nomogram to Assess Recurrence Risk in Sinonasal Malignant Tumors

BACKGROUND

Sinonasal malignant tumors (SNMTs) have a high recurrence risk, which is responsible for the poor prognosis of patients. Assessing recurrence risk in SNMT patients is a current problem.

PURPOSE

To establish an MRI-based radiomics nomogram for assessing relapse risk in patients with SNMT.

STUDY TYPE

Retrospective.

POPULATION

A total of 143 patients with 68.5% females (development/validation set, 98/45 patients).

FIELD STRENGTH/SEQUENCE

A 1.5-T and 3-T, fat-suppressed fast spin echo (FSE) T2-weighted imaging (FS-T2WI), FSE T1-weighted imaging (T1WI), and FSE contrast-enhanced T1WI (T1WI + C).

ASSESSMENT

Three MRI sequences were used to manually delineate the region of interest. Three radiomics signatures (T1WI and FS-T2WI sequences, T1WI + C sequence, and three sequences combined) were built through dimensional reduction of high-dimensional features. The clinical model was built based on clinical and MRI features. The Ki-67-based and tumor-node-metastasis (TNM) model were established for comparison. The radiomics nomogram was built by combining the clinical model and best radiomics signature. The relapse-free survival analysis was used among 143 patients.

STATISTICAL TESTS

The intraclass/interclass correlation coefficients, univariate/multivariate Cox regression analysis, least absolute shrinkage and selection operator Cox regression algorithm, concordance index (C index), area under the curve (AUC), integrated Brier score (IBS), DeLong test, Kaplan-Meier curve, log-rank test, optimal cutoff values. A P value < 0.05 was considered statistically significant.

RESULTS

The T1 + C-based radiomics signature had best prognostic ability than the other two signatures (T1WI and FS-T2WI sequences, and three sequences combined). The radiomics nomogram had better prognostic ability and less error than the clinical model, Ki-67-based model, and TNM model (C index, 0.732; AUC, 0.765; IBS, 0.185 in the validation set). The cutoff values were 0.2 and 0.7 and then the cumulative risk rates were calculated.

DATA CONCLUSION

A radiomics nomogram for assessing relapse risk in patients with SNMT may provide better prognostic ability than the clinical model, Ki-67-based model, and TNM model.

EVIDENCE LEVEL

3.

TECHNICAL EFFICACY

Stage 5.

Preoperative contrast-enhanced CT-based radiomics signature for predicting hypoxia-inducible factor 1α expression in retroperitoneal sarcoma

AIM

To develop and test a contrast-enhanced computed tomography (CECT)-based radiomics signature (RS) to preoperatively predict hypoxia-inducible factor 1α (HIF-1α) expression in retroperitoneal sarcoma (RPS).

MATERIALS AND METHODS

This study included 129 patients with RPS retrospectively who underwent CECT, including 64 male and 65 female patients (55 [2-84] years). Participants were divided into a training set comprising 85 patients and a test set comprising 44 patients. Clinical data and CECT findings of all patients were collected. RS construction was performed by the minimum redundancy maximum relevance method and least absolute shrinkage and selection operator algorithm. The clinical information was analysed by univariate and multivariate logistic regression analysis. The RS and risk factors were included to build a radiomics nomogram. The predictive efficacy of different models was evaluated by accuracy, area under the receiver operating characteristic curve (AUC), and decision curve analysis.

RESULTS

The RS combined signature was constructed on the basis of multi-phase CECT and had an accuracy of 0.795 and an AUC of 0.719 (95% confidence interval [CI], 0.552-0.886) in the test set, which were higher than that of the radiomics nomogram (accuracy: 0.636; AUC: 0.702 [95% CI, 0.547-0.857]) and the clinical model (accuracy: 0.682; AUC: 0.486 [95% CI, 0.324-0.647]). The decision curve analysis showed that the RS combined signature provided better clinical application than the clinical model and radiomics nomogram.

CONCLUSIONS

The multi-phase CECT-based RS constructed can be used as a powerful tool for predicting HIF-1α expression in patients with RPS.

A CT-Based Deep Learning Radiomics Nomogram to Predict Histological Grades of Head and Neck Squamous Cell Carcinoma

RATIONALE AND OBJECTIVES

Accurate pretreatment assessment of histological differentiation grade of head and neck squamous cell carcinoma (HNSCC) is crucial for prognosis evaluation. This study aimed to construct and validate a contrast-enhanced computed tomography (CECT)-based deep learning radiomics nomogram (DLRN) to predict histological differentiation grades of HNSCC.

MATERIALS AND METHODS

A total of 204 patients with HNSCC who underwent CECT scans were enrolled in this study. The participants recruited from two hospitals were split into a training set (n=124, 74 well/moderately differentiated and 50 poorly differentiated) of patients from one hospital and an external test set of patients from the other hospital (n=80, 49 well/moderately differentiated and 31 poorly differentiated). CECT-based manually-extracted radiomics (MER) features and deep learning (DL) features were extracted and selected. The selected MER features and DL features were then combined to construct a DLRN via multivariate logistic regression. The predictive performance of the DLRN was assessed using ROCs and decision curve analysis (DCA).

RESULTS

Three MER features and seven DL features were finally selected. The DLRN incorporating the selected MER and DL features showed good predictive value for the histological differentiation grades of HNSCC (well/moderately differentiated vs. poorly differentiated) in both the training (AUC, 0.878) and test (AUC, 0.822) sets. DCA demonstrated that the DLRN was clinically useful for predicting histological differentiation grades of HNSCC.

CONCLUSION

A CECT-based DLRN was constructed to predict histological differentiation grades of HNSCC. The DLRN showed good predictive efficacy and might be useful for prognostic evaluation of patients with HNSCC.

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

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

Differentiation Between Lipomas and Atypical Lipomatous Tumors of the Extremities Using Radiomics

BACKGROUND

The differentiation of soft tissue lipomas from atypical lipoma tumors (ALTs) of the extremities is important because of the distinction of the cytogenetic profiles and the treatment decisions.

PURPOSE

To investigate a radiomics method to differentiate between lipomas and ALTs of the extremities.

STUDY TYPE

Retrospective.

POPULATION

Imaging data of 122 patients including 90 cases of lipomas and 32 cases of ALTs.

FIELD STRENGTH/SEQUENCE

Axial T1-weighted imaging and fat suppressed T2-weighted imaging at 3.0T MRI.

ASSESSMENT

Analysis of variance and the least absolute shrinkage and selection operator methods were used for feature selection and the random forest method was used to build three radiomics models based on T1WI, FS T2WI, and their combination (T1&T2WI). Three independent radiologists classified the tumors based on the subjective assessments.

STATISTICAL TESTS

The area under the curve (AUC) of the receiver operating characteristic curve, accuracy, F1-score, specificity, and sensitivity were employed. The differences of the classifiers and discriminating ability of the radiologists and the radiomics model were compared by Delong test. A P value <0.05 was considered significant. Kappa test was used to determine the inter-reader agreements between the radiologists.

RESULT

The AUCs were 0.952 (95% confidence interval [CI]: 0.785-0.998), 0.944 (95% CI: 0.774-0.997), and 0.968 (95% CI: 0.809-1) for T1WI, FS T2WI, and T1&T2WI models in testing sets respectively. Delong test showed there were no significant difference between the different radiomics models (P > 0.05). The AUCs of the radiologists were 0.893 (95% CI: 0.824-0.942), 0.831 (95% CI: 0.752-0.893), and 0.893 (95% CI: 0.824-0.94), respectively. There were significant difference between radiomics model and radiologists' model in the training and entire cohorts (P < 0.05) while there were no significant difference in the testing sets (P > 0.05).

DATA CONCLUSION

Radiomics has the potential to distinguish between lipomas and ALTs of the extremities and their discrimination ability is no weaker than the senor radiologists.

LEVEL OF EVIDENCE

3 TECHNICAL EFFICACY STAGE: 2.

The diagnosis, classification, and treatment of sarcoma in this era of artificial intelligence and immunotherapy

Soft-tissue sarcomas (STS) represent a group of rare and heterogeneous tumors associated with several challenges, including incorrect or late diagnosis, the lack of clinical expertise, and limited therapeutic options. Digital pathology and radiomics represent transformative technologies that appear promising for improving the accuracy of cancer diagnosis, characterization and monitoring. Herein, we review the potential role of the application of digital pathology and radiomics in managing patients with STS. We have particularly described the main results and the limits of the studies using radiomics to refine diagnosis or predict the outcome of patients with soft-tissue sarcomas. We also discussed the current limitation of implementing radiomics in routine settings. Standard management approaches for STS have not improved since the early 1970s. Immunotherapy has revolutionized cancer treatment; nonetheless, immuno-oncology agents have not yet been approved for patients with STS. However, several lines of evidence indicate that immunotherapy may represent an efficient therapeutic strategy for this group of diseases. Thus, we emphasized the remarkable potential of immunotherapy in sarcoma treatment by focusing on recent data regarding the immune landscape of these tumors. We have particularly emphasized the fact that the development of immunotherapy for sarcomas is not an aspect of histology (except for alveolar soft-part sarcoma) but rather that of the tumor microenvironment. Future studies investigating immunotherapy strategies in sarcomas should incorporate at least the presence of tertiary lymphoid structures as a stratification factor in their design, besides including a strong translational program that will allow for a better understanding of the determinants involved in sensitivity and treatment resistance to immune-oncology agents.

Clinical parameters combined with radiomics features of PET/CT can predict recurrence in patients with high-risk pediatric neuroblastoma

Background

This retrospective study aimed to develop and validate a combined model based [¹⁸F]FDG PET/CT radiomics and clinical parameters for predicting recurrence in high-risk pediatric neuroblastoma patients.

Methods

Eighty-four high-risk neuroblastoma patients were retrospectively enrolled and divided into training and test sets according to the ratio of 3:2. [¹⁸F]FDG PET/CT images of the tumor were segmented by 3D Slicer software and the radiomics features were extracted. The effective features were selected by the least absolute shrinkage and selection operator to construct the radiomics score (Rad_score). And the radiomics model (R_model) was constructed based on Rad_score for prediction of recurrence. Then, univariate and multivariate analyses were used to screen out the independent clinical risk parameters and construct the clinical model (C_model). A combined model (RC_model) was developed based on the Rad_score and independent clinical risk parameters and presented as radiomics nomogram. The performance of the above three models was assessed by the area under the receiver operating characteristic curve (AUC) and decision curve analysis (DCA).

Results

Seven radiomics features were selected for building the R_model. The AUCs of the C_model in training and test sets were 0.744 (95% confidence interval [CI], 0.595–0.874) and 0.750 (95% CI, 0.577–0.904), respectively. The R_model yielded AUCs of 0.813 (95% CI, 0.685–0.916) and 0.869 (95% CI, 0.715–0.985) in the training and test sets, respectively. The RC_model demonstrated the largest AUCs of 0.889 (95% CI, 0.794–0.963) and 0.892 (95% CI, 0.758–0.992) in the training and test sets, respectively. DCA demonstrated that RC_model added more net benefits than either the C_model or the R_model for predicting recurrence in high-risk pediatric neuroblastoma.

Conclusions

The combined model performed well for predicting recurrence in high-risk pediatric neuroblastoma, which can facilitate disease follow-up and management in clinical practice.

Multi-Size Deep Learning Based Preoperative Computed Tomography Signature for Prognosis Prediction of Colorectal Cancer

Background: Preoperative and postoperative evaluation of colorectal cancer (CRC) patients is crucial for subsequent treatment guidance. Our study aims to provide a timely and rapid assessment of the prognosis of CRC patients with deep learning according to non-invasive preoperative computed tomography (CT) and explore the underlying biological explanations.

Methods: A total of 808 CRC patients with preoperative CT (development cohort: n = 426, validation cohort: n = 382) were enrolled in our study. We proposed a novel end-to-end Multi-Size Convolutional Neural Network (MSCNN) to predict the risk of CRC recurrence with CT images (CT signature). The prognostic performance of CT signature was evaluated by Kaplan-Meier curve. An integrated nomogram was constructed to improve the clinical utility of CT signature by combining with other clinicopathologic factors. Further visualization and correlation analysis for CT deep features with paired gene expression profiles were performed to reveal the molecular characteristics of CRC tumors learned by MSCNN in radiographic imaging.

Results: The Kaplan-Meier analysis showed that CT signature was a significant prognostic factor for CRC disease-free survival (DFS) prediction [development cohort: hazard ratio (HR): 50.7, 95% CI: 28.4–90.6, p &lt; 0.001; validation cohort: HR: 2.04, 95% CI: 1.44–2.89, p &lt; 0.001]. Multivariable analysis confirmed the independence prognostic value of CT signature (development cohort: HR: 30.7, 95% CI: 19.8–69.3, p &lt; 0.001; validation cohort: HR: 1.83, 95% CI: 1.19–2.83, p = 0.006). Dimension reduction and visualization of CT deep features demonstrated a high correlation with the prognosis of CRC patients. Functional pathway analysis further indicated that CRC patients with high CT signature presented down-regulation of several immunology pathways. Correlation analysis found that CT deep features were mainly associated with activation of metabolic and proliferative pathways.

Conclusions: Our deep learning based preoperative CT signature can effectively predict prognosis of CRC patients. Integration analysis of multi-omic data revealed that some molecular characteristics of CRC tumor can be captured by deep learning in CT images.

Accurate Prediction of Metachronous Liver Metastasis in Stage I-III Colorectal Cancer Patients Using Deep Learning With Digital Pathological Images

Objectives

Metachronous liver metastasis (LM) significantly impacts the prognosis of stage I-III colorectal cancer (CRC) patients. An effective biomarker to predict LM after surgery is urgently needed. We aimed to develop deep learning-based models to assist in predicting LM in stage I-III CRC patients using digital pathological images.

Methods

Six-hundred eleven patients were retrospectively included in the study and randomly divided into training (428 patients) and validation (183 patients) cohorts according to the 7:3 ratio. Digital HE images from training cohort patients were used to construct the LM risk score based on a 50-layer residual convolutional neural network (ResNet-50). An LM prediction model was established by multivariable Cox analysis and confirmed in the validation cohort. The performance of the integrated nomogram was assessed with respect to its calibration, discrimination, and clinical application value.

Results

Patients were divided into low- and high-LM risk score groups according to the cutoff value and significant differences were observed in the LM of the different risk score groups in the training and validation cohorts (P&lt;0.001). Multivariable analysis revealed that the LM risk score, VELIPI, pT stage and pN stage were independent predictors of LM. Then, the prediction model was developed and presented as a nomogram to predict the 1-, 2-, and 3-year probability of LM. The integrated nomogram achieved satisfactory discrimination, with C-indexes of 0.807 (95% CI: 0.787, 0.827) and 0.812 (95% CI: 0.773, 0.850) and AUCs of 0.840 (95% CI: 0.795, 0.885) and 0.848 (95% CI: 0.766, 0.931) in the training and validation cohorts, respectively. Favorable calibration of the nomogram was confirmed in the training and validation cohorts. Integrated discrimination improvement and net reclassification index indicated that the integrated nomogram was superior to the traditional clinicopathological model. Decision curve analysis confirmed that the nomogram has clinical application value.

Conclusions

The LM risk score based on ResNet-50 and digital HE images was significantly associated with LM. The integrated nomogram could identify stage I-III CRC patients at high risk of LM after primary colectomy, so it may serve as a potential tool to choose the appropriate treatment to improve the prognosis of stage I-III CRC patients.

Construction and Validation of Two Novel Nomograms for Predicting the Overall Survival and Cancer-Specific Survival of NSCLC Patients with Bone Metastasis

Background

Bone metastasis (BM) is the most common site of metastasis in non-small cell lung carcinoma (NSCLC). We aimed to construct and validate 2 novel nomograms predicting the 3-, 6-, and 12-months overall survival (OS) and cancer-specific survival (CSS).

Methods

The clinical data of 7480 patients between 2010 and 2016 were enrolled from the Surveillance, Epidemiology, and End Results database (SEER). The patients were allocated randomly to training and validation cohorts in a 7:3 ratio. Cox proportional hazards regression models were used to identify prognostic risk factors and establish 2 nomograms. The prediction accuracy of nomograms was assessed by C-index, the area under the ROC curve (AUC), and calibration curves.

Results

A total of 244998 NSCLC patients were identified between 2010 and 2016, with 7480 found with BM, accounting for 3.1%. Overall, 7480 patients were enrolled in the OS nomogram construction and were randomized to the training set (n = 5236) and the validation set (n = 2244). Age, sex, race, marital status, histology, grade, primary site, T stage, N stage, liver metastasis, surgery, radiotherapy, and chemotherapy were found to correlate with OS. A total of 7422 samples were included in the CSS nomogram construction, randomly grouped into training set (n = 5195) and the validation set (n = 2227). Age, sex, race, histology, grade, primary site, T stage, N stage, brain metastasis, liver metastasis, surgery, radiotherapy, and chemotherapy were associated with CSS. Two nomograms were conducted to predict the 3-, 6-, and 12-months OS and CSS. The ROC curves and exhibited good performance for predicting OS and CSS.

Conclusion

We established and validated 2 high-performance nomograms to assist clinical doctors in making personalized treatment decisions.