Automated Classification of Osteosarcoma and Benign Tumors using RNA-seq and Plain X-ray

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.

Clinical-radiomics models based on plain X-rays for prediction of lung metastasis in patients with osteosarcoma

OBJECTIVES

Osteosarcoma (OS) is the most common primary malignant bone tumor in adolescents. Lung metastasis (LM) occurs in more than half of patients at different stages of the disease course, which is one of the important factors affecting the long-term survival of OS. To develop and validate machine learning radiomics model based on radiographic and clinical features that could predict LM in OS within 3 years.

METHODS

486 patients (LM = 200, non-LM = 286) with histologically proven OS were retrospectively analyzed and divided into a training set (n = 389) and a validation set (n = 97). Radiographic features and risk factors (sex, age, tumor location, etc.) associated with LM of patients were evaluated. We built eight clinical-radiomics models (k-nearest neighbor [KNN], logistic regression [LR], support vector machine [SVM], random forest [RF], Decision Tree [DT], Gradient Boosting Decision Tree [GBDT], AdaBoost, and extreme gradient boosting [XGBoost]) and compared their performance. The area under the receiver operating characteristic curve (AUC) and accuracy (ACC) were used to evaluate different models.

RESULTS

The radscore, ALP, and tumor size had significant differences between the LM and non-LM groups (t_radscore = -5.829, χ²_ALP = 97.137, t_size = -3.437, P < 0.01). Multivariable LR analyses showed that ALP was an important indicator for predicting LM of OS (odds ratio [OR] = 7.272, P < 0.001). Among the eight models, the SVM-based clinical-radiomics model had the best performance in the validation set (AUC = 0.807, ACC = 0.784).

CONCLUSION

The clinical-radiomics model had good performance in predicting LM in OS, which would be helpful in clinical decision-making.

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.

A deep learning-machine learning fusion approach for the classification of benign, malignant, and intermediate bone tumors

OBJECTIVES

To build and validate deep learning and machine learning fusion models to classify benign, malignant, and intermediate bone tumors based on patient clinical characteristics and conventional radiographs of the lesion.

METHODS

In this retrospective study, data were collected with pathologically confirmed diagnoses of bone tumors between 2012 and 2019. Deep learning and machine learning fusion models were built to classify tumors as benign, malignant, or intermediate using conventional radiographs of the lesion and potentially relevant clinical data. Five radiologists compared diagnostic performance with and without the model. Diagnostic performance was evaluated using the area under the curve (AUC).

RESULTS

A total of 643 patients' (median age, 21 years; interquartile range, 12-38 years; 244 women) 982 radiographs were included. In the test set, the binary category classification task, the radiological model of classification for benign/not benign, malignant/nonmalignant, and intermediate/not intermediate had AUCs of 0.846, 0.827, and 0.820, respectively; the fusion models had an AUC of 0.898, 0.894, and 0.865, respectively. In the three-category classification task, the radiological model achieved a macro average AUC of 0.813, and the fusion model had a macro average AUC of 0.872. In the observation test, the mean macro average AUC of all radiologists was 0.819. With the three-category classification fusion model support, the macro AUC improved by 0.026.

CONCLUSION

We built, validated, and tested deep learning and machine learning models that classified bone tumors at a level comparable with that of senior radiologists. Model assistance may somewhat help radiologists' differential diagnoses of bone tumors.

KEY POINTS

• The deep learning model can be used to classify benign, malignant, and intermediate bone tumors. • The machine learning model fusing information from radiographs and clinical characteristics can improve the classification capacity for bone tumors. • The diagnostic performance of the fusion model is comparable with that of senior radiologists and is potentially useful as a complement to radiologists in a bone tumor differential diagnosis.