Radiomics Nomograms Based on Non-enhanced MRI and Clinical Risk Factors for the Differentiation of Chondrosarcoma from Enchondroma

Multiparametric MRI-Based Radiomics Signature with Machine Learning for Preoperative Prediction of Prognosis Stratification in Pediatric Medulloblastoma

RATIONALE AND OBJECTIVES

Despite advances in risk-stratified treatment strategies for children with medulloblastoma (MB), the prognosis for MB with short-term recurrence is extremely poor, and there is still a lack of evaluation of short-term recurrence risk or short-term survival. This study aimed to construct and validate a radiomics model for predicting the outcome of MB based on preoperative multiparametric magnetic resonance images (MRIs) and to provide an objective for clinical decision-making.

MATERIALS AND METHODS

The clinical and imaging data of 64 patients with MB admitted to Shenzhen Children's Hospital from December 2012 to December 2021 and confirmed by pathology were retrospectively collected. According to the 18-month progression-free survival, the cases were classified into a good prognosis group and a poor prognosis group, and all cases were divided into training group (70%) and validation group (30%) randomly. Radiomics features were extracted from MRI of each child. The consistency test, t-test, and the least absolute shrinkage and selection operator were used for feature selection. The support vector machine (SVM) and receiver operator characteristic were used to evaluate the distinguishing ability of the selected features to the prognostic groups. RAD score was calculated based on the selected features. The clinical characteristics and RAD score were included in the multivariate logistic regression, and prediction models were constructed by screening out independent influences. The radiomics nomogram was constructed, and its clinical significance was evaluated.

RESULTS

A total of 1930 radiomic features were extracted from the images of each patient, and 11 features were included in the construction of radiomics score after selected. The area under the curve (AUC) values of the SVM model in the training and validation groups were 0.946 and 0.797, respectively. The radiomics nomogram was constructed based on the training cohort, and the AUC values in the training group and the validation group were 0.926 and 0.835, respectively. The results of clinical decision curve analysis showed that a good net benefit could be obtained from the nomogram.

CONCLUSION

The radiomics nomogram established based on MRI can be used as a noninvasive predictive tool to evaluate the prognosis of children with MB, which is expected to help neurosurgeons better conduct preoperative planning and patient follow-up management.

Bone Reporting and Data System (Bone-RADS) and Other Proposed Practice Guidelines for Reporting Bone Tumors

BACKGROUND

 The purpose of this article is to review the different bone tumor radiology reporting systems [Bone Reporting and Data System (Bone-RADS), Osseous Tumor Reporting and Data System (OT-RADS), Solitary Bone Tumor Imaging Reporting and Data System (BTI-RADS), and Radiological Evaluation Score for Bone Tumors (REST)] and summarize their advantages and disadvantages.

METHODS

 A selective search of PubMed was performed for literature regarding the definition and discussion of bone tumor reporting systems. No time frame was selected, but the search was particularly focused on current literature on musculoskeletal radiology lexicon.

RESULTS

 To date, four major reporting systems has been proposed to standardize and systematize the reporting of imaging studies of bone tumors: Bone-RADS, OT-RADS, BTI-RADS, and REST. Both Bone-RADS and OT-RADS aid in the characterization and management of bone lesions on CT and MRI. OT-RADS and REST can be applied to MRI and radiography, respectively.

CONCLUSION

 Radiologists play a central role in the detection and characterization of asymptomatic (or incidentally detected) and symptomatic bone tumors. There are several existing bone tumor reporting systems with various advantages and disadvantages including emphasis on lesion characterization as well as management of incidentally detected bone lesions.

KEY POINTS

  · Four bone tumor reporting systems have been proposed thus far.. · Bone-RADS guides management of incidental bone lesions on CT and MRI.. · OT-RADS guides management of bone lesions on MRI with high accuracy.. · BTI-RADS classifies bone tumors on CT and MRI..

An Integrated Clinical and Computerized Tomography-Based Radiomic Feature Model to Separate Benign from Malignant Pleural Effusion

INTRODUCTION

Distinguishing between malignant pleural effusion (MPE) and benign pleural effusion (BPE) poses a challenge in clinical practice. We aimed to construct and validate a combined model integrating radiomic features and clinical factors using computerized tomography (CT) images to differentiate between MPE and BPE.

METHODS

A retrospective inclusion of 315 patients with pleural effusion (PE) was conducted in this study (training cohort: n = 220; test cohort: n = 95). Radiomic features were extracted from CT images, and the dimensionality reduction and selection processes were carried out to obtain the optimal radiomic features. Logistic regression (LR), support vector machine (SVM), and random forest were employed to construct radiomic models. LR analyses were utilized to identify independent clinical risk factors to develop a clinical model. The combined model was created by integrating the optimal radiomic features with the independent clinical predictive factors. The discriminative ability of each model was assessed by receiver operating characteristic curves, calibration curves, and decision curve analysis (DCA).

RESULTS

Out of the total 1,834 radiomic features extracted, 15 optimal radiomic features explicitly related to MPE were picked to develop the radiomic model. Among the radiomic models, the SVM model demonstrated the highest predictive performance [area under the curve (AUC), training cohort: 0.876, test cohort: 0.774]. Six clinically independent predictive factors, including age, effusion laterality, procalcitonin, carcinoembryonic antigen, carbohydrate antigen 125 (CA125), and neuron-specific enolase (NSE), were selected for constructing the clinical model. The combined model (AUC: 0.932, 0.870) exhibited superior discriminative performance in the training and test cohorts compared to the clinical model (AUC: 0.850, 0.820) and the radiomic model (AUC: 0.876, 0.774). The calibration curves and DCA further confirmed the practicality of the combined model.

CONCLUSION

This study presented the development and validation of a combined model for distinguishing MPE and BPE. The combined model was a powerful tool for assisting in the clinical diagnosis of PE patients.

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.

Research Progress of Artificial Intelligence in the Grading and Classification of Meningiomas

A meningioma is a common primary central nervous system tumor. The histological features of meningiomas vary significantly depending on the grade and subtype, leading to differences in treatment and prognosis. Therefore, early diagnosis, grading, and typing of meningiomas are crucial for developing comprehensive and individualized diagnosis and treatment plans. The advancement of artificial intelligence (AI) in medical imaging, particularly radiomics and deep learning (DL), has contributed to the increasing research on meningioma grading and classification. These techniques are fast and accurate, involve fully automated learning, are non-invasive and objective, enable the efficient and non-invasive prediction of meningioma grades and classifications, and provide valuable assistance in clinical treatment and prognosis. This article provides a summary and analysis of the research progress in radiomics and DL for meningioma grading and classification. It also highlights the existing research findings, limitations, and suggestions for future improvement, aiming to facilitate the future application of AI in the diagnosis and treatment of meningioma.

A retrospective external validation study of the Birmingham Atypical Cartilage Tumour Imaging Protocol (BACTIP) for the management of solitary central cartilage tumours of the proximal humerus and around the knee

OBJECTIVES

This study aimed to externally validate the Birmingham Atypical Cartilage Tumour Imaging Protocol (BACTIP) recommendations for differentiation/follow-up of central cartilage tumours (CCTs) of the proximal humerus, distal femur, and proximal tibia and to propose BACTIP adaptations if the results provide new insights.

METHODS

MRIs of 123 patients (45 ± 11 years, 37 men) with an untreated CCT with MRI follow-up (n = 62) or histopathological confirmation (n = 61) were retrospectively/consecutively included and categorised following the BACTIP (2003-2020 / Ghent University Hospital/Belgium). Tumour length and endosteal scalloping differences between enchondroma, atypical cartilaginous tumour (ACT), and high-grade chondrosarcoma (CS II/III/dedifferentiated) were evaluated. ROC-curve analysis for differentiating benign from malignant CCTs and for evaluating the BACTIP was performed.

RESULTS

For lesion length and endosteal scalloping, ROC-AUCs were poor and fair-excellent, respectively, for differentiating different CCT groups (0.59-0.69 versus 0.73-0.91). The diagnostic performance of endosteal scalloping and the BACTIP was higher than that of lesion length. A 1° endosteal scalloping cut-off differentiated enchondroma from ACT + high-grade chondrosarcoma with a sensitivity of 90%, reducing the potential diagnostic delay. However, the specificity was 29%, inducing overmedicalisation (excessive follow-up). ROC-AUC of the BACTIP was poor for differentiating enchondroma from ACT (ROC-AUC = 0.69; 95%CI = 0.51-0.87; p = 0.041) and fair-good for differentiation between other CCT groups (ROC-AUC = 0.72-0.81). BACTIP recommendations were incorrect/unsafe in five ACTs and one CSII, potentially inducing diagnostic delay. Eleven enchondromas received unnecessary referrals/follow-up.

CONCLUSION

Although promising as a useful tool for management/follow-up of CCTs of the proximal humerus, distal femur, and proximal tibia, five ACTs and one chondrosarcoma grade II were discharged, potentially inducing diagnostic delay, which could be reduced by adapting BACTIP cut-off values.

CLINICAL RELEVANCE STATEMENT

Mostly, Birmingham Atypical Cartilage Tumour Imaging Protocol (BACTIP) assesses central cartilage tumours of the proximal humerus and the knee correctly. Both when using the BACTIP and when adapting cut-offs, caution should be taken for the trade-off between underdiagnosis/potential diagnostic delay in chondrosarcomas and overmedicalisation in enchondromas.

KEY POINTS

• This retrospective external validation confirms the Birmingham Atypical Cartilage Tumour Imaging Protocol as a useful tool for initial assessment and follow-up recommendation of central cartilage tumours in the proximal humerus and around the knee in the majority of cases. • Using only the Birmingham Atypical Cartilage Tumour Imaging Protocol, both atypical cartilaginous tumours and high-grade chondrosarcomas (grade II, grade III, and dedifferentiated chondrosarcomas) can be misdiagnosed, excluding them from specialist referral and further follow-up, thus creating a potential risk of delayed diagnosis and worse prognosis. • Adapted cut-offs to maximise detection of atypical cartilaginous tumours and high-grade chondrosarcomas, minimise underdiagnosis and reduce potential diagnostic delay in malignant tumours but increase unnecessary referral and follow-up of benign tumours.

Intratumoral and peritumoral CT radiomics in predicting prognosis in patients with chondrosarcoma: a multicenter study

OBJECTIVE

To evaluate the efficacy of the CT-based intratumoral, peritumoral, and combined radiomics signatures in predicting progression-free survival (PFS) of patients with chondrosarcoma (CS).

METHODS

In this study, patients diagnosed with CS between January 2009 and January 2022 were retrospectively screened, and 214 patients with CS from two centers were respectively enrolled into the training cohorts (institution 1, n = 113) and test cohorts (institution 2, n = 101). The intratumoral and peritumoral radiomics features were extracted from CT images. The intratumoral, peritumoral, and combined radiomics signatures were constructed respectively, and their radiomics scores (Rad-score) were calculated. The performance of intratumoral, peritumoral, and combined radiomics signatures in PFS prediction in patients with CS was evaluated by C-index, time-dependent area under the receiver operating characteristics curve (time-AUC), and time-dependent C-index (time C-index).

RESULTS

Eleven, 7, and 16 features were used to construct the intratumoral, peritumoral, and combined radiomics signatures, respectively. The combined radiomics signature showed the best prediction ability in the training cohort (C-index, 0.835; 95%; confidence interval [CI], 0.764-0.905) and the test cohort (C-index, 0.800; 95% CI, 0.681-0.920). Time-AUC and time C-index showed that the combined signature outperformed the intratumoral and peritumoral radiomics signatures in the prediction of PFS.

CONCLUSION

The CT-based combined signature incorporating intratumoral and peritumoral radiomics features can predict PFS in patients with CS, which might assist clinicians in selecting individualized surveillance and treatment plans for CS patients.

CRITICAL RELEVANCE STATEMENT

Develop and validate CT-based intratumoral, peritumoral, and combined radiomics signatures to evaluate the efficacy in predicting prognosis of patients with CS.

KEY POINTS

• Reliable prognostic models for preoperative chondrosarcoma are lacking. • Combined radiomics signature incorporating intratumoral and peritumoral features can predict progression-free survival in patients with chondrosarcoma. • Combined radiomics signature may facilitate individualized stratification and management of patients with chondrosarcoma.

Impact of Image Enhancement on the Radiomics Stability of Diffusion-Weighted MRI Images of Cervical Cancer

The diffusion-weighted imaging (DWI) technique is known for its capability to differentiate the diffusion of water molecules between cancerous and non-cancerous cervix tissues, which enhances the accuracy of detection. Despite the potential of DWI-MRI, its accuracy is limited by technical factors influencing in vivo data acquisition, thus impacting the quantification of radiomics features. This study aimed to measure the radiomics stability of manual and semi-automated segmentation on contrast limited adaptive histogram equalization (CLAHE)-enhanced DWI-MRI cervical images. Eighty diffusion-weighted MRI images were obtained from patients diagnosed with cervical cancer, and an active contour model was used to analyze the data. Radiomics analysis was conducted to extract the first statistical order, shape, and textural features with intraclass correlation coefficient (ICC) measurement. The results of the CLAHE segmentation approach showed a marked improvement when compared to the manual and semi-automated segmentation methods, with an ICC value of 0.990 ± 0.005 (p<0.05), compared to 0.864 ± 0.033 (p<0.05) and 0.554 ± 0.185 (p>0.05), respectively. The CLAHE segmentation displayed a higher level of robustness than the manual groups in terms of the features present in both categories. Thus, CLAHE segmentation is owing to its potential to generate radiomics features that are more durable and consistent.

SPECT/CT Radiomics for Differentiating between Enchondroma and Grade I Chondrosarcoma

This study was performed to assess the value of SPECT/CT radiomics parameters in differentiating enchondroma and atypical cartilaginous tumors (ACTs) located in the long bones. Quantitative HDP SPECT/CT data of 49 patients with enchondromas or ACTs in the long bones were retrospectively reviewed. Patients were randomly split into training (n = 32) and test (n = 17) data, and SPECT/CT radiomics parameters were extracted. In training data, LASSO was employed for feature reduction. Selected parameters were compared with classic quantitative parameters for the prediction of diagnosis. Significant parameters from training data were again tested in the test data. A total of 12 (37.5%) and 6 (35.2%) patients were diagnosed as ACTs in training and test data, respectively. LASSO regression selected two radiomics features, zone-length non-uniformity for zone (ZLNUGLZLM) and coarseness for neighborhood grey-level difference (CoarsenessNGLDM). Multivariate analysis revealed higher ZLNUGLZLM as the only significant independent factor for the prediction of ACTs, with sensitivity and specificity of 85.0% and 58.3%, respectively, with a cut-off value of 191.26. In test data, higher ZLNUGLZLM was again associated with the diagnosis of ACTs, with sensitivity and specificity of 83.3% and 90.9%, respectively. HDP SPECT/CT radiomics may provide added value for differentiating between enchondromas and ACTs.

Clinical and radiomics feature-based outcome analysis in lumbar disc herniation surgery

BACKGROUND

Low back pain is a widely prevalent symptom and the foremost cause of disability on a global scale. Although various degenerative imaging findings observed on magnetic resonance imaging (MRI) have been linked to low back pain and disc herniation, none of them can be considered pathognomonic for this condition, given the high prevalence of abnormal findings in asymptomatic individuals. Nevertheless, there is a lack of knowledge regarding whether radiomics features in MRI images combined with clinical features can be useful for prediction modeling of treatment success. The objective of this study was to explore the potential of radiomics feature analysis combined with clinical features and artificial intelligence-based techniques (machine learning/deep learning) in identifying MRI predictors for the prediction of outcomes after lumbar disc herniation surgery.

METHODS

We included n = 172 patients who underwent discectomy due to disc herniation with preoperative T2-weighted MRI examinations. Extracted clinical features included sex, age, alcohol and nicotine consumption, insurance type, hospital length of stay (LOS), complications, operation time, ASA score, preoperative CRP, surgical technique (microsurgical versus full-endoscopic), and information regarding the experience of the performing surgeon (years of experience with the surgical technique and the number of surgeries performed at the time of surgery). The present study employed a semiautomatic region-growing volumetric segmentation algorithm to segment herniated discs. In addition, 3D-radiomics features, which characterize phenotypic differences based on intensity, shape, and texture, were extracted from the computed magnetic resonance imaging (MRI) images. Selected features identified by feature importance analyses were utilized for both machine learning and deep learning models (n = 17 models).

RESULTS

The mean accuracy over all models for training and testing in the combined feature set was 93.31 ± 4.96 and 88.17 ± 2.58. The mean accuracy for training and testing in the clinical feature set was 91.28 ± 4.56 and 87.69 ± 3.62.

CONCLUSIONS

Our results suggest a minimal but detectable improvement in predictive tasks when radiomics features are included. However, the extent of this advantage should be considered with caution, emphasizing the potential of exploring multimodal data inputs in future predictive modeling.

Artificial intelligence-based radiomics in bone tumors: Technical advances and clinical application

Radiomics is the extraction of predefined mathematic features from medical images for predicting variables of clinical interest. Recent research has demonstrated that radiomics can be processed by artificial intelligence algorithms to reveal complex patterns and trends for diagnosis, and prediction of prognosis and response to treatment modalities in various types of cancer. Artificial intelligence tools can utilize radiological images to solve next-generation issues in clinical decision making. Bone tumors can be classified as primary and secondary (metastatic) tumors. Osteosarcoma, Ewing sarcoma, and chondrosarcoma are the dominating primary tumors of bone. The development of bone tumor model systems and relevant research, and the assessment of novel treatment methods are ongoing to improve clinical outcomes, notably for patients with metastases. Artificial intelligence and radiomics have been utilized in almost full spectrum of clinical care of bone tumors. Radiomics models have achieved excellent performance in the diagnosis and grading of bone tumors. Furthermore, the models enable to predict overall survival, metastases, and recurrence. Radiomics features have exhibited promise in assisting therapeutic planning and evaluation, especially neoadjuvant chemotherapy. This review provides an overview of the evolution and opportunities for artificial intelligence in imaging, with a focus on hand-crafted features and deep learning-based radiomics approaches. We summarize the current application of artificial intelligence-based radiomics both in primary and metastatic bone tumors, and discuss the limitations and future opportunities of artificial intelligence-based radiomics in this field. In the era of personalized medicine, our in-depth understanding of emerging artificial intelligence-based radiomics approaches will bring innovative solutions to bone tumors and achieve clinical application.

The diagnostic value of machine learning for the classification of malignant bone tumor: a systematic evaluation and meta-analysis

Background

Malignant bone tumors are a type of cancer with varying malignancy and prognosis. Accurate diagnosis and classification are crucial for treatment and prognosis assessment. Machine learning has been introduced for early differential diagnosis of malignant bone tumors, but its performance is controversial. This systematic review and meta-analysis aims to explore the diagnostic value of machine learning for malignant bone tumors.

Methods

PubMed, Embase, Cochrane Library, and Web of Science were searched for literature on machine learning in the differential diagnosis of malignant bone tumors up to October 31, 2022. The risk of bias assessment was conducted using QUADAS-2. A bivariate mixed-effects model was used for meta-analysis, with subgroup analyses by machine learning methods and modeling approaches.

Results

The inclusion comprised 31 publications with 382,371 patients, including 141,315 with malignant bone tumors. Meta-analysis results showed machine learning sensitivity and specificity of 0.87 [95% CI: 0.81,0.91] and 0.91 [95% CI: 0.86,0.94] in the training set, and 0.83 [95% CI: 0.74,0.89] and 0.87 [95% CI: 0.79,0.92] in the validation set. Subgroup analysis revealed MRI-based radiomics was the most common approach, with sensitivity and specificity of 0.85 [95% CI: 0.74,0.91] and 0.87 [95% CI: 0.81,0.91] in the training set, and 0.79 [95% CI: 0.70,0.86] and 0.79 [95% CI: 0.70,0.86] in the validation set. Convolutional neural networks were the most common model type, with sensitivity and specificity of 0.86 [95% CI: 0.72,0.94] and 0.92 [95% CI: 0.82,0.97] in the training set, and 0.87 [95% CI: 0.51,0.98] and 0.87 [95% CI: 0.69,0.96] in the validation set.

Conclusion

Machine learning is mainly applied in radiomics for diagnosing malignant bone tumors, showing desirable diagnostic performance. Machine learning can be an early adjunctive diagnostic method but requires further research and validation to determine its practical efficiency and clinical application prospects.

Systematic review registration

https://www.crd.york.ac.uk/prospero/, identifier CRD42023387057.

Development of clinical and magnetic resonance imaging-based radiomics nomograms for the differentiation of nodular fasciitis from soft tissue sarcoma

BACKGROUND

Accurate differentiation of nodular fasciitis (NF) from soft tissue sarcoma (STS) before surgery is essential for the subsequent diagnosis and treatment of patients.

PURPOSE

To develop and evaluate radiomics nomograms based on clinical factors and magnetic resonance imaging (MRI) for the preoperative differentiation of NF from STS.

MATERIAL AND METHODS

This retrospective study analyzed the MRI data of 27 patients with pathologically diagnosed NF and 58 patients with STS who were randomly divided into training (n = 62) and validation (n = 23) groups. Univariate and multivariate analyses were performed to identify the clinical factors and semantic features of MRI. Radiomics analysis was applied to fat-suppressed T1-weighted (T1W-FS) images, fat-suppressed T2-weighted (T2W-FS) images, and contrast-enhanced T1-weighted (CE-T1W) images. The radiomics nomograms incorporating the radiomics signatures, clinical factors, and semantic features of MRI were developed. ROC curves and AUCs were carried out to compare the performance of the clinical factors, radiomics signatures, and clinical radiomics nomograms.

RESULTS

Tumor location, size, heterogeneous signal intensity on T2W-FS imaging, heterogeneous signal intensity on CE-T1W imaging, margin definitions on CE-T1W imaging, and septa were independent predictors for differentiating NF from STS (P < 0.05). The performance of the radiomics signatures based on T2W-FS imaging (AUC = 0.961) and CE-T1W imaging (AUC = 0.938) was better than that based on T1W-FS imaging (AUC = 0.833). The radiomics nomograms had AUCs of 0.949, which demonstrated good clinical utility and calibration.

CONCLUSION

The non-invasive clinical radiomics nomograms exhibited good performance in the differentiation of NF from STS, and they have clinical application in the preoperative diagnosis of diseases.

Development of machine learning models aiming at knee osteoarthritis diagnosing: an MRI radiomics analysis

BACKGROUND

To develop and assess the performance of machine learning (ML) models based on magnetic resonance imaging (MRI) radiomics analysis for knee osteoarthritis (KOA) diagnosis.

METHODS

This retrospective study analysed 148 consecutive patients (72 with KOA and 76 without) with available MRI image data, where radiomics features in cartilage portions were extracted and then filtered. Intraclass correlation coefficient (ICC) was calculated to quantify the reproducibility of features, and a threshold of 0.8 was set. The training and validation cohorts consisted of 117 and 31 cases, respectively. Least absolute shrinkage and selection operator (LASSO) regression method was employed for feature selection. The ML classifiers were logistic regression (LR), K-nearest neighbour (KNN) and support vector machine (SVM). In each algorithm, ten models derived from all available planes of three joint compartments and their various combinations were, respectively, constructed for comparative analysis. The performance of classifiers was mainly evaluated and compared by receiver operating characteristic (ROC) analysis.

RESULTS

All models achieved satisfying performances, especially the Final model, where accuracy and area under ROC curve (AUC) of LR classifier were 0.968, 0.983 (0.957-1.000, 95% CI) in the validation cohort, and 0.940, 0.984 (0.969-0.995, 95% CI) in the training cohort, respectively.

CONCLUSION

The MRI radiomics analysis represented promising performance in noninvasive and preoperative KOA diagnosis, especially when considering all available planes of all three compartments of knee joints.

The diagnostic value of magnetic resonance imaging-based texture analysis in differentiating enchondroma and chondrosarcoma

OBJECTIVE

To assess the diagnostic performance of MRI-based texture analysis for differentiating enchondromas and chondrosarcomas, especially on fat-suppressed proton density (FS-PD) images.

MATERIALS AND METHODS

The whole tumor volumes of 23 chondrosarcomas and 24 enchondromas were manually segmented on both FS-PD and T1-weighted images. A total of 861 radiomic features were extracted. SelectKBest was used to select the features. The data were randomly split into training (n = 36) and test (n = 10) for T1-weighted and training (n = 37) and test (n = 10) for FS-PD datasets. Fivefold cross-validation was performed. Fifteen machine learning models were created using the training set. The best models for T1-weighted, FS-PD, and T1-weighted + FS-PD images were selected in terms of accuracy and area under the curve (AUC).

RESULTS

There were 7 men and 16 women in the chondrosarcoma group (mean ± standard deviation age, 45.65 ± 11.24) and 7 men and 17 women in the enchondroma group (mean ± standard deviation age, 46.17 ± 11.79). Naive Bayes was the best model for accuracy and AUC for T1-weighted images (AUC = 0.76, accuracy = 80%, recall = 80%, precision = 80%, F1 score = 80%). The best model for FS-PD images was the K neighbors classifier for accuracy and AUC (AUC = 1.00, accuracy = 80%, recall = 80%, precision = 100%, F1 score = 89%). The best model for T1-weighted + FS-PD images was logistic regression for accuracy and AUC (AUC = 0.84, accuracy = 80%, recall = 60%, precision = 100%, F1 score = 75%).

CONCLUSION

MRI-based machine learning models have promising results in the discrimination of enchondroma and chondrosarcoma based on radiomic features obtained from both FS-PD and T1-weighted images.

The use of radiomics and machine learning for the differentiation of chondrosarcoma from enchondroma

PURPOSE

To construct and compare machine learning models for differentiating chondrosarcoma from enchondroma using radiomic features from T1 and fat suppressed Proton density (PD) magnetic resonance imaging (MRI).

METHODS

Eighty-eight patients (57 with enchondroma, 31 with chondrosarcoma) were retrospectively included. Histogram matching and N4ITK MRI bias correction filters were applied. An experienced musculoskeletal radiologist and a senior resident in radiology performed manual segmentation. Voxel sizes were resampled. Laplacian of Gaussian filter and wavelet-based features were used. One thousand eight hundred eighty-eight features were obtained for each patient, with 944 from T1 and 944 from PD images. Sixty-four unstable features were removed. Seven machine learning models were used for classification.

RESULTS

Classification with all features showed neural network was the best model for both readers' datasets with area under the curve (AUC), classification accuracy (CA), and F1 score of 0.979, 0.984; 0.920, 0.932; and 0.889, 0.903, respectively. Four features, including one common to both readers, were selected using fast correlation based filter. The best performing models with selected features were gradient boosting for Fatih Erdem's dataset and neural network for Gülen Demirpolat's dataset with AUC, CA, and F1 score of 0.990, 0.979; 0.943, 0.955; 0.921, 0.933, respectively. Neural Network was the second-best model for FE's dataset based on AUC (0.984).

CONCLUSION

Using pathology as a gold standard, this study defined and compared seven well-performing models to distinguish enchondromas from chondrosarcomas and provided radiomic feature stability and reproducibility among the readers.

Classification of Chondrosarcoma: From Characteristic to Challenging Imaging Findings

Chondrosarcomas can be classified into various forms according to the presence or absence of a precursor lesion, location, and histological subtype. The new 2020 World Health Organization (WHO) Classification of Tumors of Soft Tissue and Bone classifies chondrogenic bone tumors as benign, intermediate (locally aggressive), or malignant, and separates atypical cartilaginous tumors (ACTs) and chondrosarcoma grade 1 (CS1) as intermediate and malignant tumors. respectively. Furthermore, the classification categorizes chondrosarcomas (including ACT) into eight subtypes: central conventional (grade 1 vs. 2–3), secondary peripheral (grade 1 vs. 2–3), periosteal, dedifferentiated, mesenchymal, and clear cell chondrosarcoma. Most chondrosarcomas are the low-grade, primary central conventional type. The rarer subtypes include clear cell, mesenchymal, and dedifferentiated chondrosarcomas. Comprehensive analysis of the characteristic imaging findings can help differentiate various forms of chondrosarcomas. However, distinguishing low-grade chondrosarcomas from enchondromas or high-grade chondrosarcomas is radiologically and histopathologically challenging, even for experienced radiologists and pathologists.

A systematic review of radiomics in chondrosarcoma: assessment of study quality and clinical value needs handy tools

OBJECTIVE

To evaluate the study quality and clinical value of radiomics studies on chondrosarcoma.

METHODS

PubMed, Embase, Web of Science, China National Knowledge Infrastructure, and Wanfang Data were searched for articles on radiomics for evaluating chondrosarcoma as of January 31, 2022. The study quality was assessed according to Radiomics Quality Score (RQS), Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis (TRIPOD) checklist, Image Biomarker Standardization Initiative (IBSI) guideline, and modified Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2) tool. The level of evidence supporting clinical use of radiomics on chondrosarcoma differential diagnosis was determined based on meta-analyses.

RESULTS

Twelve articles were included. The median RQS was 10.5 (range, -3 to 15), with an adherence rate of 36%. The adherence rate was extremely low in domains of high-level evidence (0%), open science and data (17%), and imaging and segmentation (35%). The adherence rate of the TRIPOD checklist was 61%, and low for section of title and abstract (13%), introduction (42%), and results (56%). The reporting rate of pre-processing steps according to the IBSI guideline was 60%. The risk of bias and concern of application were mainly related to the index test. The meta-analysis on differential diagnosis of enchondromas vs. chondrosarcomas showed a diagnostic odds ratio of 43.90 (95% confidential interval, 25.33-76.10), which was rated as weak evidence.

CONCLUSIONS

The current scientific and reporting quality of radiomics studies on chondrosarcoma was insufficient. Radiomics has potential in facilitating the optimization of operation decision-making in chondrosarcoma.

KEY POINTS

• Among radiomics studies on chondrosarcoma, although differential diagnostic models showed promising performance, only pieces of weak level of evidence were reached with insufficient study quality. • Since the RQS rating, the TRIPOD checklist, and the IBSI guideline have largely overlapped with each other, it is necessary to establish one widely acceptable methodological and reporting guideline for radiomics research. • The TRIPOD model typing, the phase classification of image mining studies, and the level of evidence category are useful tools to assess the gap between academic research and clinical application, although their modifications for radiomics studies are needed.

Nomogram model based on radiomics signatures and age to assist in the diagnosis of knee osteoarthritis

BACKGROUND

Knee osteoarthritis (KOA) is a common disease in the elderly. An effective method for accurate diagnosis could affect the management and prognosis of patients.

OBJECTIVES

To develop a nomogram model based on X-ray imaging data and age, and to evaluate its effectiveness in the diagnosis of KOA.

METHODS

A total of 4403 knee X-rays from 1174 patients (July 2017 to November 2018) were retrospectively analyzed. Radiomics features were extracted and selected from the X-ray image data to quantify the phenotypic characteristics of the lesion region. Feature selection was performed in three steps to enable the derivation of robust and effective radiomics signatures. Then, logistic regression (LR), support vector machine (SVM) AdaBoost, gradient boosting decision tree (GBDT), and multi-layer perceptron (MLP) was adopted to verify the performance of radiomics signatures. In addition, a nomogram model combining age with radiomics signatures was constructed. At last, receiver operating characteristic (ROC) curve, calibration and decision curves were used to evaluate the discriminative performance.

RESULTS

The LR model has the best classification performance among the four radiomics models in testing cohort (LR AUC vs. SVM AUC: 0.843 vs. 0.818, DeLong test P = 0.0024; LR AUC vs. GBDT AUC: 0.843 vs. 0.821, P = 0.0028; LR AUC vs. MLP AUC: 0.843 vs. 0.822, P = 0.0019). The nomogram model achieved better predictive efficacy than the radiomics model in testing cohort compared to radiomics models although the statistical difference was not significant (Nomogram AUC vs. Radiomics AUC: 0.847 vs. 0.843, P = 0.06). The decision curve analysis revealed that the constructed nomogram had clinical usefulness.

CONCLUSION

The nomogram model combining radiomics signatures with age has good performance for the accurate diagnosis of KOA and may help to improve clinical decision-making.

Development and validation of a MRI-based combined radiomics nomogram for differentiation in chondrosarcoma

Objective

This study aims to develop and validate the performance of an unenhanced magnetic resonance imaging (MRI)-based combined radiomics nomogram for discrimination between low-grade and high-grade in chondrosarcoma.

Methods

A total of 102 patients with 44 in low-grade and 58 in high-grade chondrosarcoma were enrolled and divided into training set (n=72) and validation set (n=30) with a 7:3 ratio in this retrospective study. The demographics and unenhanced MRI imaging characteristics of the patients were evaluated to develop a clinic-radiological factors model. Radiomics features were extracted from T1-weighted (T1WI) images to construct radiomics signature and calculate radiomics score (Rad-score). According to multivariate logistic regression analysis, a combined radiomics nomogram based on MRI was constructed by integrating radiomics signature and independent clinic-radiological features. The performance of the combined radiomics nomogram was evaluated in terms of calibration, discrimination, and clinical usefulness.

Results

Using multivariate logistic regression analysis, only one clinic-radiological feature (marrow edema OR=0.29, 95% CI=0.11-0.76, P=0.012) was found to be independent predictors of differentiation in chondrosarcoma. Combined with the above clinic-radiological predictor and the radiomics signature constructed by LASSO [least absolute shrinkage and selection operator], a combined radiomics nomogram based on MRI was constructed, and its predictive performance was better than that of clinic-radiological factors model and radiomics signature, with the AUC [area under the curve] of the training set and the validation set were 0.78 (95%CI =0.67-0.89) and 0.77 (95%CI =0.59-0.94), respectively. DCA [decision curve analysis] showed that combined radiomics nomogram has potential clinical application value.

Conclusion

The MRI-based combined radiomics nomogram is a noninvasive preoperative prediction tool that combines clinic-radiological feature and radiomics signature and shows good predictive effect in distinguishing low-grade and high-grade bone chondrosarcoma, which may help clinicians to make accurate treatment plans.

Application of a nomogram to radiomics labels in the treatment prediction scheme for lumbar disc herniation

Objective

To investigate and verify the efficiency and effectiveness of a nomogram based on radiomics labels in predicting the treatment of lumbar disc herniation (LDH).

Methods

By reviewing medical records that were analysed over the past three years, clinical and imaging data of 200 lumbar disc patients at the Affiliated Hospital of Jiangxi University of Traditional Chinese Medicine were obtained. The collected cases were randomly divided into a training group (n = 140) and a testing group (n = 60) at a ratio of 7:3. Two radiologists with experience in reading orthopaedics images independently segmented the ROIs. The whole intervertebral disc with the most obvious protrusion in the sagittal plane T2WI lumbar MRI as a mask (ROI) is sketched. The LASSO (Least Absolute Shrinkage And Selection Operator) algorithm was used to filter the features after extracting the radiomics features. The multivariate logistic regression model was used to construct a quantitative imaging Rad‑Score for the selected features with nonzero coefficients. The radiomics labels and nomogram were evaluated using the receiver operating characteristic curve (ROC) and the area under the curve (AUC). The calibration curve was used to evaluate the consistency between the nomogram prediction and the actual treatment plan. The DCA decision curve was used to evaluate the clinical applicability of the nomogram.

Result

Following feature extraction, 11 radiomics features were used to construct the radiomics label for predicting the treatment plan of LDH. A nomogram was then constructed. The AUC was 0.93 (95% CI: 0.90–0.97), with a sensitivity of 89%, a specificity of 91%, a positive predictive value of 92.7%, a negative predictive value of 89.4%, and an accuracy of 91%. The calibration curve showed that there was good consistency between the prediction and the actual observation. The DCA decision curve analysis showed that the nomogram of the imaging group has great potential for clinical application when the risk threshold is between 5 and 72%.

Conclusion

A nomogram based on radiomics labels has good predictive value for the treatment of LDH and can be used as a reference for clinical decision-making.

MRI radiomics-based machine learning classification of atypical cartilaginous tumour and grade II chondrosarcoma of long bones

Background

Atypical cartilaginous tumour (ACT) and grade II chondrosarcoma (CS2) of long bones are respectively managed with watchful waiting or curettage and wide resection. Preoperatively, imaging diagnosis can be challenging due to interobserver variability and biopsy suffers from sample errors. The aim of this study is to determine diagnostic performance of MRI radiomics-based machine learning in differentiating ACT from CS2 of long bones.

Methods

One-hundred-fifty-eight patients with surgically treated and histology-proven cartilaginous bone tumours were retrospectively included at two tertiary bone tumour centres. The training cohort consisted of 93 MRI scans from centre 1 (n=74 ACT; n=19 CS2). The external test cohort consisted of 65 MRI scans from centre 2 (n=45 ACT; n=20 CS2). Bidimensional segmentation was performed on T1-weighted MRI. Radiomic features were extracted. After dimensionality reduction and class balancing in centre 1, a machine-learning classifier (Extra Trees Classifier) was tuned on the training cohort using 10-fold cross-validation and tested on the external test cohort. In centre 2, its performance was compared with an experienced musculoskeletal oncology radiologist using McNemar's test.

Findings

After tuning on the training cohort (AUC=0.88), the machine-learning classifier had 92% accuracy (60/65, AUC=0.94) in identifying the lesions in the external test cohort. Its accuracies in correctly classifying ACT and CS2 were 98% (44/45) and 80% (16/20), respectively. The radiologist had 98% accuracy (64/65) with no difference compared to the classifier (p=0.134).

Interpretation

Machine learning showed high accuracy in classifying ACT and CS2 of long bones based on MRI radiomic features.

Funding

ESSR Young Researchers Grant.

Diagnostic Value of CT- and MRI-Based Texture Analysis and Imaging Findings for Grading Cartilaginous Tumors in Long Bones

Objective

To confirm the diagnostic performance of computed tomography (CT)-based texture analysis (CTTA) and magnetic resonance imaging (MRI)-based texture analysis for grading cartilaginous tumors in long bones and to compare these findings to radiological features.

Materials and Methods

Twenty-nine patients with enchondromas, 20 with low-grade chondrosarcomas and 16 with high-grade chondrosarcomas were included retrospectively. Clinical and radiological information and 9 histogram features extracted from CT, T1WI, and T2WI were evaluated. Binary logistic regression analysis was performed to determine predictive factors for grading cartilaginous tumors and to establish diagnostic models. Another 26 patients were included to validate each model. Receiver operating characteristic (ROC) curves were generated, and accuracy rate, sensitivity, specificity and positive/negative predictive values (PPV/NPV) were calculated.

Results

On imaging, endosteal scalloping, cortical destruction and calcification shape were predictive for grading cartilaginous tumors. For texture analysis, variance, mean, perc.01%, perc.10%, perc.99% and kurtosis were extracted after multivariate analysis. To differentiate benign cartilaginous tumors from low-grade chondrosarcomas, the imaging features model reached the highest accuracy rate (83.7%) and AUC (0.841), with a sensitivity of 75% and specificity of 93.1%. The CTTA feature model best distinguished low-grade and high-grade chondrosarcomas, with accuracies of 71.9%, and 80% in the training and validation groups, respectively; T1-TA and T2-TA could not distinguish them well. We found that the imaging feature model best differentiated benign and malignant cartilaginous tumors, with an accuracy rate of 89.2%, followed by the T1-TA feature model (80.4%).

Conclusions

The imaging feature model and CTTA- or MRI-based texture analysis have the potential to differentiate cartilaginous tumors in long bones by grade. MRI-based texture analysis failed to grade chondrosarcomas.