Prediction of Chemotherapy Response of Osteosarcoma Using Baseline 18F-FDG Textural Features Machine Learning Approaches with PCA

Evaluation of response to neoadjuvant chemotherapy in osteosarcoma using dynamic contrast-enhanced MRI: development and external validation of a model

OBJECTIVE

To identify which dynamic contrast-enhanced (DCE-)MRI features best predict histological response to neoadjuvant chemotherapy in patients with an osteosarcoma.

METHODS

Patients with osteosarcoma who underwent DCE-MRI before and after neoadjuvant chemotherapy prior to resection were retrospectively included at two different centers. Data from the center with the larger cohort (training cohort) was used to identify which method for region-of-interest selection (whole slab or focal area method) and which change in DCE-MRI features (time to enhancement, wash-in rate, maximum relative enhancement and area under the curve) gave the most accurate prediction of histological response. Models were created using logistic regression and cross-validated. The most accurate model was then externally validated using data from the other center (test cohort).

RESULTS

Fifty-five (27 poor response) and 30 (19 poor response) patients were included in training and test cohorts, respectively. Intraclass correlation coefficient of relative DCE-MRI features ranged 0.81-0.97 with the whole slab and 0.57-0.85 with the focal area segmentation method. Poor histological response was best predicted with the whole slab segmentation method using a single feature threshold, relative wash-in rate <2.3. Mean accuracy was 0.85 (95%CI: 0.75-0.95), and area under the receiver operating characteristic curve (AUC-index) was 0.93 (95%CI: 0.86-1.00). In external validation, accuracy and AUC-index were 0.80 and 0.80.

CONCLUSION

In this study, a relative wash-in rate of <2.3 determined with the whole slab segmentation method predicted histological response to neoadjuvant chemotherapy in osteosarcoma. Consistent performance was observed in an external test cohort.

Research status and progress of radiomics in bone and soft tissue tumors: A review

Bone and soft tissue tumors are diverse, accompanying by complex histological components and significantly divergent biological behaviors. It is a challenge to address the demand for qualitative imaging as traditional imaging is restricted to the detection of anatomical structures and aberrant signals. With the improvement of digitalization in hospitals and medical centers, the introduction of electronic medical records and easier access to large amounts of information coupled with the improved computational power, traditional medicine has evolved into the combination of human brain, minimal data, and artificial intelligence. Scholars are committed to mining deeper levels of imaging data, and radiomics is worthy of promotion. Radiomics extracts subvisual quantitative features, analyzes them based on medical images, and quantifies tumor heterogeneity by outlining the region of interest and modeling. Two observers separately examined PubMed, Web of Science and CNKI to find existing studies, case reports, and clinical guidelines about research status and progress of radiomics in bone and soft tissue tumors from January 2010 to February 2023. When evaluating the literature, factors such as patient age, medical history, and severity of the condition will be considered. This narrative review summarizes the application and progress of radiomics in bone and soft tissue tumors.

Deep Learning Approaches to Osteosarcoma Diagnosis and Classification: A Comparative Methodological Approach

BACKGROUND

Osteosarcoma is the most common primary malignancy of the bone, being most prevalent in childhood and adolescence. Despite recent progress in diagnostic methods, histopathology remains the gold standard for disease staging and therapy decisions. Machine learning and deep learning methods have shown potential for evaluating and classifying histopathological cross-sections.

METHODS

This study used publicly available images of osteosarcoma cross-sections to analyze and compare the performance of state-of-the-art deep neural networks for histopathological evaluation of osteosarcomas.

RESULTS

The classification performance did not necessarily improve when using larger networks on our dataset. In fact, the smallest network combined with the smallest image input size achieved the best overall performance. When trained using 5-fold cross-validation, the MobileNetV2 network achieved 91% overall accuracy.

CONCLUSIONS

The present study highlights the importance of careful selection of network and input image size. Our results indicate that a larger number of parameters is not always better, and the best results can be achieved on smaller and more efficient networks. The identification of an optimal network and training configuration could greatly improve the accuracy of osteosarcoma diagnoses and ultimately lead to better disease outcomes for patients.

CTDN (Convolutional Temporal Based Deep- Neural Network): An Improvised Stacked Hybrid Computational Approach for Anticancer Drug Response Prediction

The characterization of drug - metabolizing enzymes is a significant problem for customized therapy. It is important to choose the right drugs for cancer victims, and the ability to forecast how those drugs will react is usually based on the available information, genetic sequence, and structural properties. To the finest of our knowledge, this is the first study to evaluate optimization algorithms for selection of features and pharmacogenetics categorization using classification methods based on a successful evolutionary algorithm using datasets from the Cancer Cell Line Encyclopaedia (CCLE) and Genomics of Drug Sensitivity in Cancer (GDSC). The study proposes the uses of Firefly and Grey Wolf Optimization techniques for feature extraction, while comparing the traditional Machine Learning (ML), ensemble ML and Stacking Algorithm with the proposed Convolutional Temporal Deep Neural Network or CTDN. With the potential to increase efficiency from the suggested intelligible classifier model for a suggestive chemotherapeutic drugs response prediction, our study is important in particular for selecting an acceptable feature selection method. The comparison analysis demonstrates that the proposed model not only surpasses the prior state-of-the-art methods, but also uses Grey Wolf and Fire Fly Optimization to lessen multicollinearity and overfitting.

An updated systematic review of radiomics in osteosarcoma: utilizing CLAIM to adapt the increasing trend of deep learning application in radiomics

Objective

To update the systematic review of radiomics in osteosarcoma.

Methods

PubMed, Embase, Web of Science, China National Knowledge Infrastructure, and Wanfang Data were searched to identify articles on osteosarcoma radiomics until May 15, 2022. The studies were assessed by Radiomics Quality Score (RQS), Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis (TRIPOD) statement, Checklist for Artificial Intelligence in Medical Imaging (CLAIM), and modified Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2) tool. The evidence supporting radiomics application for osteosarcoma was rated according to meta-analysis results.

Results

Twenty-nine articles were included. The average of the ideal percentage of RQS, the TRIPOD adherence rate and the CLAIM adherence rate were 29.2%, 59.2%, and 63.7%, respectively. RQS identified a radiomics-specific issue of phantom study. TRIPOD addressed deficiency in blindness of assessment. CLAIM and TRIPOD both pointed out shortness in missing data handling and sample size or power calculation. CLAIM identified extra disadvantages in data de-identification and failure analysis. External validation and open science were emphasized by all the above three tools. The risk of bias and applicability concerns were mainly related to the index test. The meta-analysis of radiomics predicting neoadjuvant chemotherapy response by MRI presented a diagnostic odds ratio (95% confidence interval) of 28.83 (10.27–80.95) on testing datasets and was rated as weak evidence.

Conclusions

The quality of osteosarcoma radiomics studies is insufficient. More investigation is needed before using radiomics to optimize osteosarcoma treatment. CLAIM is recommended to guide the design and reporting of radiomics research.

Radiomics in Oncological PET Imaging: A Systematic Review-Part 2, Infradiaphragmatic Cancers, Blood Malignancies, Melanoma and Musculoskeletal Cancers

The objective of this review was to summarize published radiomics studies dealing with infradiaphragmatic cancers, blood malignancies, melanoma, and musculoskeletal cancers, and assess their quality. PubMed database was searched from January 1990 to February 2022 for articles performing radiomics on PET imaging of at least 1 specified tumor type. Exclusion criteria includd: non-oncological studies; supradiaphragmatic tumors; reviews, comments, cases reports; phantom or animal studies; technical articles without a clinically oriented question; studies including <30 patients in the training cohort. The review database contained PMID, first author, year of publication, cancer type, number of patients, study design, independent validation cohort and objective. This database was completed twice by the same person; discrepant results were resolved by a third reading of the articles. A total of 162 studies met inclusion criteria; 61 (37.7%) studies included >100 patients, 13 (8.0%) were prospective and 61 (37.7%) used an independent validation set. The most represented cancers were esophagus, lymphoma, and cervical cancer (n = 24, n = 24 and n = 19 articles, respectively). Most studies focused on 18F-FDG, and prognostic and response to treatment objectives. Although radiomics and artificial intelligence are technically challenging, new contributions and guidelines help improving research quality over the years and pave the way toward personalized medicine.

Automated prediction of the neoadjuvant chemotherapy response in osteosarcoma with deep learning and an MRI-based radiomics nomogram

OBJECTIVES

To implement a pipeline to automatically segment the ROI and to use a nomogram integrating the MRI-based radiomics score and clinical variables to predict responses to neoadjuvant chemotherapy (NAC) in osteosarcoma patients.

METHODS

A total of 144 osteosarcoma patients treated with NAC were separated into training (n = 101) and test (n = 43) groups. After normalisation, ROIs for the preoperative MRI were segmented by a deep learning segmentation model trained with nnU-Net by using two independent manual segmentations as labels. Radiomics features were extracted using automatically segmented ROIs. Feature selection was performed in the training dataset by five-fold cross-validation. The clinical, radiomics, and clinical-radiomics models were built using multiple machine learning methods with the same training dataset and validated with the same test dataset. The segmentation model was evaluated by the Dice coefficient. AUC and decision curve analysis (DCA) were employed to illustrate the model performance and clinical utility.

RESULTS

36/144 (25.0%) patients were pathological good responders (pGRs) to NAC, while 108/144 (75.0%) were non-pGRs. The segmentation model achieved a Dice coefficient of 0.869 on the test dataset. The clinical and radiomics models reached AUCs of 0.636 with a 95% confidence interval (CI) of 0.427-0.860 and 0.759 (95% CI, 0.589-0.937), respectively, in the test dataset. The clinical-radiomics nomogram demonstrated good discrimination, with an AUC of 0.793 (95% CI, 0.610-0.975), and accuracy of 79.1%. The DCA suggested the clinical utility of the nomogram.

CONCLUSION

The automatic nomogram could be applied to aid radiologists in identifying pGRs to NAC.

KEY POINTS

• The nnU-Net trained by manual labels enables the use of an automatic segmentation tool for ROI delineation of osteosarcoma. • A pipeline using automatic lesion segmentation and followed by a radiomics classifier could aid the evaluation of NAC response of osteosarcoma. • A predictive nomogram composed of clinical variables and MRI-based radiomics score provides support for individualised treatment planning.

H-SegNet: hybrid segmentation network for lung segmentation in chest radiographs using mask region-based convolutional neural network and adaptive closed polyline searching method

Chest x-ray (CXR) is one of the most commonly used imaging techniques for the detection and diagnosis of pulmonary diseases. One critical component in many computer-aided systems, for either detection or diagnosis in digital CXR, is the accurate segmentation of the lung. Due to low-intensity contrast around lung boundary and large inter-subject variance, it has been challenging to segment lung from structural CXR images accurately. In this work, we propose an automatic Hybrid Segmentation Network (H-SegNet) for lung segmentation on CXR. The proposed H-SegNet consists of two key steps: (1) an image preprocessing step based on a deep learning model to automatically extract coarse lung contours; (2) a refinement step to fine-tune the coarse segmentation results based on an improved principal curve-based method coupled with an improved machine learning method. Experimental results on several public datasets show that the proposed method achieves superior segmentation results in lung CXRs, compared with several state-of-the-art methods.

Development and Preliminary Validation of the Physical Education-Study Process Questionnaire : Insights for Physical Education University Students

Background

The Revised Two Factor Study Process Questionnaire (R-SPQ-2F) is used to examine students' study approaches in higher education. The questionnaire is designed to measure two factors: deep and surface approaches. In order to measure these approaches for students in physical education and sport, a new measurement instrument should take into consideration the practical context of this field of education that makes it specific to other fields.

Objective

The present study aims (a) to develop and empirical test of a new instrument for measuring the study process in physical education and sports students, and (b) to test psychometric properties of the tool.

Methods

Two exploratory and confirmatory samples of physical education students enrolled in a bachelor's degree program in physical education at the High Institute of Physical Education and Sports of Kef-Tunisia, aged 19-26 years, were recruited online among female students (n = 414) and male students (n = 393). The participants filled in Google Form survey including Physical Education-Study Process Questionnaire (PE-SPQ) and the Arabic version of the Revised Study Process Questionnaire-2 Factors (R-SPQ-2F).

Results

Exploratory factor analysis showed a suitable four factors solution, which is approved by confirmatory factor analysis indices [χ2 = 466.47, TLI = 0.94, CFI = 0.95; RMSEA = 0.56 IC 90% (0.050-0.062)]. Internal consistency of the PE-SPQ simultaneously checked by McDonald's ω, Cronbach's α and Gutmann's λ6 showed good reliability of the PE-SPQ. Convergent validity examined by Average variance extracted (AVE) was good. The comparison between the AVE root mean square and Pearson correlation coefficients of each factor with his indicators reveals the discriminant validity of the PE-SPQ. Furthermore, Pearson's correlation between the PE-SPQ factors and the R-SPQ-2F establishes the concurrent validity of the new scale.

Conclusion

The PE-SPQ scale is valid and reliable and can be used to assess study process factors in physical education students.

Clinical Perspectives for 18F-FDG PET Imaging in Pediatric Oncology: Μetabolic Tumor Volume and Radiomics

Pediatric cancer, although rare, requires the most optimized treatment approach to obtain high survival rates and minimize serious long-term side effects in early adulthood. 18F-FDG PET/CT is most helpful and widely used in staging, recurrence detection, and response assessment in pediatric oncology. The well-known 18F-FDG PET metabolic indices of metabolic tumor volume (MTV) and tumor lesion glycolysis (TLG) have already revealed an independent significant prognostic value for survival in oncologic patients, although the corresponding cut-off values remain study-dependent and not validated for use in clinical practice. Advanced tumor “radiomic” analysis sheds new light into these indices. Numerous patterns of texture 18F-FDG uptake features can be extracted from segmented PET tumor images due to new powerful computational systems supporting complex “deep learning” algorithms. This high number of “quantitative” tumor imaging data, although not decrypted in their majority and once standardized for the different imaging systems and segmentation methods, could be used for the development of new “clinical” models for specific cancer types and, more interestingly, for specific age groups. In addition, data from novel techniques of tumor genome analysis could reveal new genes as biomarkers for prognosis and/or targeted therapies in childhood malignancies. Therefore, this ever-growing information of “radiogenomics”, in which the underlying tumor “genetic profile” could be expressed in the tumor-imaging signature of “radiomics”, possibly represents the next model for precision medicine in pediatric cancer management. This paper reviews 18F-FDG PET image segmentation methods as applied to pediatric sarcomas and lymphomas and summarizes reported findings on the values of metabolic and radiomic features in the assessment of these pediatric tumors.

Intelligent Segmentation Medical Assistance System for MRI Images of Osteosarcoma in Developing Countries

Osteosarcoma is the most common primary malignant bone tumor in children and adolescents. It has a high degree of malignancy and a poor prognosis in developing countries. The doctor manually explained that magnetic resonance imaging (MRI) suffers from subjectivity and fatigue limitations. In addition, the structure, shape, and position of osteosarcoma are complicated, and there is a lot of noise in MRI images. Directly inputting the original data set into the automatic segmentation system will bring noise and cause the model's segmentation accuracy to decrease. Therefore, this paper proposes an osteosarcoma MRI image segmentation system based on a deep convolution neural network, which solves the overfitting problem caused by noisy data and improves the generalization performance of the model. Firstly, we use Mean Teacher to optimize the data set. The noise data is put into the second round of training of the model to improve the robustness of the model. Then, we segment the image using a deep separable U-shaped network (SepUNet) and conditional random field (CRF). SepUnet can segment lesion regions of different sizes at multiple scales; CRF further optimizes the boundary. Finally, this article calculates the area of the tumor area, which provides a more intuitive reference for assisting doctors in diagnosis. More than 80000 MRI images of osteosarcoma from three hospitals in China were tested. The results show that the proposed method guarantees the balance of speed, accuracy, and cost under the premise of improving accuracy.

Big Data to Knowledge: Application of Machine Learning to Predictive Modeling of Therapeutic Response in Cancer

Background

In recent years, the availability of high throughput technologies, establishment of large molecular patient data repositories, and advancement in computing power and storage have allowed elucidation of complex mechanisms implicated in therapeutic response in cancer patients. The breadth and depth of such data, alongside experimental noise and missing values, requires a sophisticated human-machine interaction that would allow effective learning from complex data and accurate forecasting of future outcomes, ideally embedded in the core of machine learning design.

Objective

In this review, we will discuss machine learning techniques utilized for modeling of treatment response in cancer, including Random Forests, support vector machines, neural networks, and linear and logistic regression. We will overview their mathematical foundations and discuss their limitations and alternative approaches in light of their application to therapeutic response modeling in cancer.

Conclusion

We hypothesize that the increase in the number of patient profiles and potential temporal monitoring of patient data will define even more complex techniques, such as deep learning and causal analysis, as central players in therapeutic response modeling.

KSNM60 in Clinical Nuclear Oncology

Since the foundation of the Korean Society of Nuclear Medicine in 1961, clinical nuclear oncology has been a major part of clinical nuclear medicine in Korea. There are several important events for the development of clinical nuclear oncology in Korea. First, a scintillating type gamma camera was adopted in 1969, which enabled to perform modern oncological gamma imaging. Second, Tc-99 m generator was imported to Korea since 1979, which promoted the wide clinical use of gamma camera imaging by using various kinds of Tc-99 m labeled radiopharmaceuticals. Third, a gamma camera with single photon emission tomography (SPECT) capability was first installed in 1980, which has been used for various kinds of tumor SPECT imaging. Fourth, in 1994, clinical positron emission tomography (PET) scanner and cyclotron with a production of F-18 fluorodeoxyglucose were first installed in Korea. Fifth, Korean Board of Nuclear Medicine was established in 1995, which contributed in the education and manpower training of dedicated nuclear medicine physicians in Korea. Finally, an integrated PET/CT scanner was first installed in 2002. Since that, PET/CT imaging has been a major imaging tool in clinical nuclear oncology in Korea. In this review, a brief history of clinical nuclear oncology in Korea is described.

Computational analysis of MRIs predicts osteosarcoma chemoresponsiveness

Aim: This study aimed to improve osteosarcoma chemoresponsiveness prediction by optimization of computational analysis of MRIs. Patients & methods: Our retrospective predictive model involved osteosarcoma patients with MRI scans performed before OsteoSa MAP neoadjuvant cytotoxic chemotherapy. Results: We found that several monofractal and multifractal algorithms were able to classify tumors according to their chemoresponsiveness. The predictive clues were defined as morphological complexity, homogeneity and fractality. The monofractal feature CV for Λ'(G) provided the best predictive association (area under the ROC curve = 0.88; p <0.001), followed by   Y-axis intersection of the regression line  for  box fractal dimension, r²  for  FD_M and tumor circularity. Conclusion: This is the first full-scale study to indicate that computational analysis of pretreatment MRIs could provide imaging biomarkers for the classification of osteosarcoma according to their chemoresponsiveness.

Pelvic Ewing sarcoma: Should all patients receive pre-operative radiotherapy, or should it be delivered selectively?

BACKGROUND

Locally recurrent disease following surgical resection of Ewing sarcoma (ES) confers a poor prognosis. Limited evidence is available evaluating non-selective use of pre-operative radiotherapy (RT) for patients with pelvic ES and its effect on local control and survival.

PATIENTS AND METHODS

49 consecutive patients with pelvic ES were identified retrospectively from a prospectively collated database. Patients either received non-selective pre-operative RT and surgery (n = 27), or selective post-operative RT (n = 22) (surgery alone (n = 11) or surgery and post-operative RT (n = 11)).

RESULTS

Patients who had non-selective pre-operative radiotherapy appeared to have a higher LRFS, 88.0% compared to 66.5% in the selective RT group (p = 0.096, Kaplan Meier; p = 0.028, Chi-squared). Administration of non-selective, pre-operative RT to all patients with pelvic ES elevates the LFRS to that of the good responder group (≥90% tumour necrosis and margins, p = 0.880). There was no difference in metastasis-free survival, 60.0% and 54.5% (p = 0.728) and overall survival (OS), 57.7% and 63.6% (p = 0.893). The majority of pre-operative RT patients had both good necrosis (≥90%) (p = 0.003) and widely excised tumours, 81.5% vs 59.1% (p = 0.080). Tumour volume ≥250 ml was associated with worse LRFS (p = 0.045) and post-operative complications (p = 0.017). There may be improved LRFS (p = 0.057) with pre-operative proton-beam RT compared to surgery and selective post-operative RT.

CONCLUSION

Pre-operative photon or proton-beam RT to all pelvic ES may improve LRFS compared to the selective delivery of post-operative RT. Radiotherapy delivered to all patients results in a greater percentage of highly necrotic tumours at surgical excision, enabling a greater proportion of patients with wide resection margins.

[89Zr]-Pertuzumab PET Imaging Reveals Paclitaxel Treatment Efficacy Is Positively Correlated with HER2 Expression in Human Breast Cancer Xenograft Mouse Models

Paclitaxel (PTX) treatment efficacy varies in breast cancer, yet the underlying mechanism for variable response remains unclear. This study evaluates whether human epidermal growth factor receptor 2 (HER2) expression level utilizing advanced molecular positron emission tomography (PET) imaging is correlated with PTX treatment efficacy in preclinical mouse models of HER2+ breast cancer. HER2 positive (BT474, MDA-MB-361), or HER2 negative (MDA-MB-231) breast cancer cells were subcutaneously injected into athymic nude mice and PTX (15 mg/kg) was administrated. In vivo HER2 expression was quantified through [89Zr]-pertuzumab PET/CT imaging. PTX treatment response was quantified by [18F]-fluorodeoxyglucose ([18F]-FDG) PET/CT imaging. Spearman's correlation, Kendall's tau, Kolmogorov-Smirnov test, and ANOVA were used for statistical analysis. [89Zr]-pertuzumab mean standard uptake values (SUVmean) of BT474 tumors were 4.9 ± 1.5, MDA-MB-361 tumors were 1.4 ± 0.2, and MDA-MB-231 (HER2-) tumors were 1.1 ± 0.4. [18F]-FDG SUVmean changes were negatively correlated with [89Zr]-pertuzumab SUVmean (r = -0.5887, p = 0.0030). The baseline [18F]-FDG SUVmean was negatively correlated with initial [89Zr]-pertuzumab SUVmean (r = -0.6852, p = 0.0002). This study shows PTX treatment efficacy is positively correlated with HER2 expression level in human breast cancer mouse models. Molecular imaging provides a non-invasive approach to quantify biological interactions, which will help in identifying chemotherapy responders and potentially enhance clinical decision-making.

A Systematic Review of PET Textural Analysis and Radiomics in Cancer

Background: Although many works have supported the utility of PET radiomics, several authors have raised concerns over the robustness and replicability of the results. This study aimed to perform a systematic review on the topic of PET radiomics and the used methodologies. Methods: PubMed was searched up to 15 October 2020. Original research articles based on human data specifying at least one tumor type and PET image were included, excluding those that apply only first-order statistics and those including fewer than 20 patients. Each publication, cancer type, objective and several methodological parameters (number of patients and features, validation approach, among other things) were extracted. Results: A total of 290 studies were included. Lung (28%) and head and neck (24%) were the most studied cancers. The most common objective was prognosis/treatment response (46%), followed by diagnosis/staging (21%), tumor characterization (18%) and technical evaluations (15%). The average number of patients included was 114 (median = 71; range 20–1419), and the average number of high-order features calculated per study was 31 (median = 26, range 1–286). Conclusions: PET radiomics is a promising field, but the number of patients in most publications is insufficient, and very few papers perform in-depth validations. The role of standardization initiatives will be crucial in the upcoming years.

Total-Body PET Imaging of Musculoskeletal Disorders

Imaging of musculoskeletal disorders, including arthritis, infection, osteoporosis, sarcopenia, and malignancies, is often limited when using conventional modalities such as radiography, computed tomography (CT), and MR imaging. As a result of recent advances in Positron Emission Tomography (PET) instrumentation, total-body PET/CT offers a longer axial field-of-view, higher geometric sensitivity, and higher spatial resolution compared with standard PET systems. This article discusses the potential applications of total-body PET/CT imaging in the assessment of musculoskeletal disorders.

Machine learning for predictive data analytics in medicine: A review illustrated by cardiovascular and nuclear medicine examples

The evidence-based medicine allows the physician to evaluate the risk-benefit ratio of a treatment through setting and data. Risk-based choices can be done by the doctor using different information. With the emergence of new technologies, a large amount of data is recorded offering interesting perspectives with machine learning for predictive data analytics. Machine learning is an ensemble of methods that process data to model a learning problem. Supervised machine learning algorithms consist in using annotated data to construct the model. This category allows to solve prediction data analytics problems. In this paper, we detail the use of supervised machine learning algorithms for predictive data analytics problems in medicine. In the medical field, data can be split into two categories: medical images and other data. For brevity, our review deals with any kind of medical data excluding images. In this article, we offer a discussion around four supervised machine learning approaches: information-based, similarity-based, probability-based and error-based approaches. Each method is illustrated with detailed cardiovascular and nuclear medicine examples. Our review shows that model ensemble (ME) and support vector machine (SVM) methods are the most popular. SVM, ME and artificial neural networks often lead to better results than those given by other algorithms. In the coming years, more studies, more data, more tools and more methods will, for sure, be proposed.

A systematic review of radiomics in osteosarcoma: utilizing radiomics quality score as a tool promoting clinical translation

OBJECTIVES

To assess the methodological quality and risk of bias in radiomics studies investigating diagnosis, therapy response, and survival of patients with osteosarcoma.

METHODS

In this systematic review, literatures on radiomics in osteosarcoma were included and assessed for methodological quality through the radiomics quality score (RQS). The risk of bias and concern of application was assessed using the Quality Assessment of Diagnostic Accuracy Studies tool. A meta-analysis of studies focusing on predicting osteosarcoma response to neoadjuvant chemotherapy was performed.

RESULTS

Twelve radiomics studies exploring osteosarcoma were identified, and five were included in meta-analysis. The RQS reached an average of 20.4% (6.92 of 36) with good inter-rater agreement (ICC 0.95, 95% CI 0.85-0.99). Four studies validated results with an internal dataset, none of which used external dataset; one study was prospectively designed, and another one shared part of the dataset. The risk of bias and concern of application were mainly related to index test aspect. The meta-analysis showed a diagnostic odds ratio of 43.68 (95%CI 13.5-141.31) for predicting response to neoadjuvant chemotherapy with high heterogeneity and low methodological quality.

CONCLUSIONS

The overall scientific quality of included studies is insufficient; however, radiomics remains a promising technology for predicting treatment response, which might guide therapeutic decision-making and related to prognosis. Improvements in study design, validation, and open science needs to be made to demonstrate the generalizability of findings and to achieve clinical applications. Widespread application of RQS, pre-trained RQS scoring procedure, and modification of RQS in response to clinical needs are necessary.

KEY POINTS

• Limited radiomics studies were established in osteosarcoma with mean RQS of 20.4%, commonly due to unvalidated results, retrospective study design, and absence of open science. • Meta-analysis of radiomics studies predicting osteosarcoma response to neoadjuvant chemotherapy showed high diagnostic odds ratio 43.68, while high heterogeneity and low methodological quality were the main concerns. • A previously trained data extraction instrument allowed reaching moderate inter-rater agreement in RQS applications, while RQS still needs improvement to become a wide adaptive tool in reviews of radiomics studies, in routine self-check before manuscript submitting and in study design.

Development and external validation of an MRI-based radiomics nomogram for pretreatment prediction for early relapse in osteosarcoma: A retrospective multicenter study

PURPOSE

To develop and externally validate an MR-based radiomics nomogram from retrospective multicenter datasets for pretreatment prediction of early relapse (≤ 1 year) in osteosarcoma after surgical resection.

METHODS

This multicenter study retrospectively enrolled 93 patients (training cohort: 62 patients from four hospitals; validation cohort: 31 patients from two hospitals) with clinicopathologically confirmed osteosarcoma who received neoadjuvant chemotherapy and surgical resection at six hospitals between January 2009 and October 2017. Radiomics features were extracted from contrast-enhanced fat-suppressed T1-weighted (CE FS T1-w) images. Least absolute shrinkage and selection operator (LASSO) regression was applied for feature selection and radiomics signature construction. The radiomics nomogram that incorporated the radiomics signature and subjective MRI-assessed candidate predictors was developed to predict early relapse with a multivariate logistic regression model in the training cohort and validated in the external validation cohort. The performance of the nomogram was assessed by its discrimination, calibration, and clinical usefulness.

RESULTS

The radiomics signature comprised six selected features and achieved favorable prediction efficacy. The radiomics nomogram incorporating the radiomics signature and subjective MRI-assessed candidate predictors (joint invasion and perivascular involvement) from the multicenter datasets achieved better discrimination in the training cohort (C-index：0.907, 95 % CI: 0.838-0.977) and external validation cohort (C-index: 0.811, 95 % CI: 0.653-0.970), and good calibration. Decision curve analysis suggested that the combined nomogram was clinically useful.

CONCLUSION

The proposed MRI-based radiomics nomogram could provide a non-invasive tool to predict early relapse of osteosarcoma, which has the potential to improve personalized pretreatment management of osteosarcoma.