Development, Validation, and Comparison of Image-Based, Clinical Feature-Based and Fusion Artificial Intelligence Diagnostic Models in Differentiating Benign and Malignant Pulmonary Ground-Glass Nodules

Objective This study aimed to develop effective artificial intelligence (AI) diagnostic models based on CT images of pulmonary nodules only, on descriptional and quantitative clinical or image features, or on a combination of both to differentiate benign and malignant ground-glass nodules (GGNs) to assist in the determination of surgical intervention. Methods Our study included a total of 867 nodules (benign nodules: 112; malignant nodules: 755) with postoperative pathological diagnoses from two centers. For the diagnostic models to discriminate between benign and malignant GGNs, we adopted three different artificial intelligence (AI) approaches: a) an image-based deep learning approach to build a deep neural network (DNN); b) a clinical feature-based machine learning approach based on the clinical and image features of nodules; c) a fusion diagnostic model integrating the original images and the clinical and image features. The performance of the models was evaluated on an internal test dataset (the “Changzheng Dataset”) and an independent test dataset collected from an external institute (the “Longyan Dataset”). In addition, the performance of automatic diagnostic models was compared with that of manual evaluations by two radiologists on the ‘Longyan dataset’. Results The image-based deep learning model achieved an appealing diagnostic performance, yielding AUC values of 0.75 (95% confidence interval [CI]: 0.62, 0.89) and 0.76 (95% CI: 0.61, 0.90), respectively, on both the Changzheng and Longyan datasets. The clinical feature-based machine learning model performed well on the Changzheng dataset (AUC, 0.80 [95% CI: 0.64, 0.96]), whereas it performed poorly on the Longyan dataset (AUC, 0.62 [95% CI: 0.42, 0.83]). The fusion diagnostic model achieved the best performance on both the Changzheng dataset (AUC, 0.82 [95% CI: 0.71-0.93]) and the Longyan dataset (AUC, 0.83 [95% CI: 0.70-0.96]), and it achieved a better specificity (0.69) than the radiologists (0.33-0.44) on the Longyan dataset. Conclusion The deep learning models, including both the image-based deep learning model and the fusion model, have the ability to assist radiologists in differentiating between benign and malignant nodules for the precise management of patients with GGNs.

[1]  M. Tripathi.  Analysis of Convolutional Neural Network based Image Classification Techniques , 2021, Journal of Innovative Image Processing.

[2]  Zhenxing Jiang,et al.  Establishment and verification of a prediction model based on clinical characteristics and positron emission tomography/computed tomography (PET/CT) parameters for distinguishing malignant from benign ground-glass nodules. , 2021, Quantitative imaging in medicine and surgery.

[3]  R. Vliegenthart,et al.  Human-recognizable CT image features of subsolid lung nodules associated with diagnosis and classification by convolutional neural networks , 2021, European Radiology.

[4]  H. Aerts,et al.  Prognostic Value of Deep Learning-Mediated Treatment Monitoring in Lung Cancer Patients Receiving Immunotherapy , 2021, Frontiers in Oncology.

[5]  Weijun Peng,et al.  Computer-aided diagnosis of ground glass pulmonary nodule by fusing deep learning and radiomics features , 2021, Physics in medicine and biology.

[6]  A. Jemal,et al.  Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries , 2021, CA: a cancer journal for clinicians.

[7]  Fa-jin Lv,et al.  Benign and Malignant Part-Solid Nodules: Differentiation Via Thin-Section Computed Tomography , 2021 .

[8]  Fa-jin Lv,et al.  Benign and malignant pulmonary part-solid nodules: differentiation via thin-section computed tomography. , 2021, Quantitative imaging in medicine and surgery.

[9]  Wei-min Li,et al.  Assessing PD-L1 expression in non-small cell lung cancer and predicting responses to immune checkpoint inhibitors using deep learning on computed tomography images , 2021, Theranostics.

[10]  Manan Shah,et al.  An anatomization on breast cancer detection and diagnosis employing multi-layer perceptron neural network (MLP) and Convolutional neural network (CNN) , 2020 .

[11]  Najmul Hasan,et al.  DenseNet Convolutional Neural Networks Application for Predicting COVID-19 Using CT Image , 2020, SN Computer Science.

[12]  Xiang Wang,et al.  Predicting the invasiveness of lung adenocarcinomas appearing as ground-glass nodule on CT scan using multi-task learning and deep radiomics , 2020, Translational lung cancer research.

[13]  Y. Yen,et al.  Predicting malignancy: subsolid nodules detected on LDCT in a surgical cohort of East Asian patients , 2020, Journal of thoracic disease.

[14]  P. Pang,et al.  Non-invasive evaluation for benign and malignant subcentimeter pulmonary ground-glass nodules (≤1 cm) based on CT texture analysis. , 2020, The British journal of radiology.

[15]  Erratum: Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. , 2020, CA: a cancer journal for clinicians.

[16]  Mohammed A. Fadhel,et al.  Towards a Better Understanding of Transfer Learning for Medical Imaging: A Case Study , 2020, Applied Sciences.

[17]  Lixin Zheng,et al.  A transfer learning method with deep residual network for pediatric pneumonia diagnosis , 2020, Comput. Methods Programs Biomed..

[18]  Shivajirao M. Jadhav,et al.  Deep convolutional neural network based medical image classification for disease diagnosis , 2019, Journal of Big Data.

[19]  Tae Jung Kim,et al.  Management of incidental pulmonary nodules: current strategies and future perspectives , 2019, Expert review of respiratory medicine.

[20]  Joel J. P. C. Rodrigues,et al.  A novel deep learning based framework for the detection and classification of breast cancer using transfer learning , 2019, Pattern Recognit. Lett..

[21]  Ajit Singh,et al.  Machine Learning With Python , 2019 .

[22]  S. Duffy,et al.  Probability of cancer in lung nodules using sequential volumetric screening up to 12 months: the UKLS trial , 2019, Thorax.

[23]  Guohui Tian,et al.  A novel scene classification model combining ResNet based transfer learning and data augmentation with a filter , 2019, Neurocomputing.

[24]  Yuguo Tang,et al.  Can peritumoral radiomics increase the efficiency of the prediction for lymph node metastasis in clinical stage T1 lung adenocarcinoma on CT? , 2019, European Radiology.

[25]  M. Cham,et al.  The Added Value of Computer-aided Detection of Small Pulmonary Nodules and Missed Lung Cancers , 2018, Journal of thoracic imaging.

[26]  A. Jemal,et al.  Higher Lung Cancer Incidence in Young Women Than Young Men in the United States , 2018, The New England journal of medicine.

[27]  N. Zhong,et al.  Malignant solitary pulmonary nodules: assessment of mass growth rate and doubling time at follow-up CT. , 2018, Journal of thoracic disease.

[28]  Alison Twycross,et al.  What is a case study? , 2017, Evidence Based Journals.

[29]  Wei Li,et al.  Pulmonary Nodule Classification with Deep Convolutional Neural Networks on Computed Tomography Images , 2016, Comput. Math. Methods Medicine.

[30]  A. Jemal,et al.  Cancer statistics in China, 2015 , 2016, CA: a cancer journal for clinicians.

[31]  Ronald M. Summers,et al.  Deep Convolutional Neural Networks for Computer-Aided Detection: CNN Architectures, Dataset Characteristics and Transfer Learning , 2016, IEEE Transactions on Medical Imaging.

[32]  S. Matsumoto,et al.  Dynamic contrast-enhanced CT and MRI for pulmonary nodule assessment. , 2014, AJR. American journal of roentgenology.

[33]  S. Lam,et al.  Probability of cancer in pulmonary nodules detected on first screening CT. , 2013, The New England journal of medicine.

[34]  L. Fan,et al.  Multidetector CT features of pulmonary focal ground-glass opacity: differences between benign and malignant. , 2012, The British journal of radiology.

[35]  Gaël Varoquaux,et al.  Scikit-learn: Machine Learning in Python , 2011, J. Mach. Learn. Res..

[36]  Y. Shim,et al.  Persistent pulmonary nodular ground-glass opacity at thin-section CT: histopathologic comparisons. , 2007, Radiology.

[37]  M. Wahidi,et al.  Evidence for the treatment of patients with pulmonary nodules: when is it lung cancer?: ACCP evidence-based clinical practice guidelines (2nd edition). , 2007, Chest.

[38]  Jin Mo Goo,et al.  Nodular ground-glass opacity at thin-section CT: histologic correlation and evaluation of change at follow-up. , 2007, Radiographics : a review publication of the Radiological Society of North America, Inc.

[39]  J. Habbema,et al.  Risk‐based selection from the general population in a screening trial: Selection criteria, recruitment and power for the Dutch‐Belgian randomised lung cancer multi‐slice CT screening trial (NELSON) , 2007, International journal of cancer.

[40]  T. Naruke,et al.  Evaluation of F-18 fluorodeoxyglucose (FDG) PET scanning for pulmonary nodules less than 3 cm in diameter, with special reference to the CT images. , 2004, Lung cancer.

[41]  O. Miettinen,et al.  CT screening for lung cancer: frequency and significance of part-solid and nonsolid nodules. , 2002, AJR. American journal of roentgenology.

[42]  Kenji Eguchi,et al.  Focal ground-glass opacity detected by low-dose helical CT. , 2002, Chest.

[43]  W. Travis,et al.  [New WHO classification of lung adenocarcinoma and preneoplasia]. , 2016, Annales de pathologie.

[44]  Michael K Gould,et al.  Evidence-Based Clinical Practice Guidelines Nodules : When Is It Lung Cancer ? : ACCP Evaluation of Patients With Pulmonary , 2007 .