Leveraging Deep Learning Decision-Support System in Specialized Oncology Center: A Multi-Reader Retrospective Study on Detection of Pulmonary Lesions in Chest X-ray Images

Chest X-ray (CXR) is considered to be the most widely used modality for detecting and monitoring various thoracic findings, including lung carcinoma and other pulmonary lesions. However, X-ray imaging shows particular limitations when detecting primary and secondary tumors and is prone to reading errors due to limited resolution and disagreement between radiologists. To address these issues, we developed a deep-learning-based automatic detection algorithm (DLAD) to automatically detect and localize suspicious lesions on CXRs. Five radiologists were invited to retrospectively evaluate 300 CXR images from a specialized oncology center, and the performance of individual radiologists was subsequently compared with that of DLAD. The proposed DLAD achieved significantly higher sensitivity (0.910 (0.854–0.966)) than that of all assessed radiologists (RAD 10.290 (0.201–0.379), p < 0.001, RAD 20.450 (0.352–0.548), p < 0.001, RAD 30.670 (0.578–0.762), p < 0.001, RAD 40.810 (0.733–0.887), p = 0.025, RAD 50.700 (0.610–0.790), p < 0.001). The DLAD specificity (0.775 (0.717–0.833)) was significantly lower than for all assessed radiologists (RAD 11.000 (0.984–1.000), p < 0.001, RAD 20.970 (0.946–1.000), p < 0.001, RAD 30.980 (0.961–1.000), p < 0.001, RAD 40.975 (0.953–0.997), p < 0.001, RAD 50.995 (0.985–1.000), p < 0.001). The study results demonstrate that the proposed DLAD could be utilized as a decision-support system to reduce radiologists’ false negative rate.

[1]  G. Chassagnon,et al.  Artificial intelligence: A critical review of applications for lung nodule and lung cancer. , 2022, Diagnostic and interventional imaging.

[2]  Eleni Eirini Karolou,et al.  TraCon: A novel dataset for real-time traffic cones detection using deep learning , 2022, NiDS.

[3]  R. Hamoudi,et al.  Identifying malignant nodules on chest X-rays: A validation study of radiologist versus artificial intelligence diagnostic accuracy , 2022, Advances in Biomedical and Health Sciences.

[4]  Florin C. Ghesu,et al.  An Artificial Intelligence–Based Chest X-ray Model on Human Nodule Detection Accuracy From a Multicenter Study , 2021, JAMA network open.

[5]  Artur Przelaskowski,et al.  Checklist for responsible deep learning modeling of medical images based on COVID-19 detection studies , 2020, Pattern Recognition.

[6]  Juan Yu,et al.  Multi-resolution convolutional networks for chest X-ray radiograph based lung nodule detection , 2019, Artif. Intell. Medicine.

[7]  Sang Min Lee,et al.  Short-term Reproducibility of Pulmonary Nodule and Mass Detection in Chest Radiographs: Comparison among Radiologists and Four Different Computer-Aided Detections with Convolutional Neural Net , 2019, Scientific Reports.

[8]  E. Kotter,et al.  Deep Convolutional Neural Network-based Software Improves Radiologist Detection of Malignant Lung Nodules on Chest Radiographs. , 2019, Radiology.

[9]  K. Delman,et al.  Lung Surveillance Strategy for High-Grade Soft Tissue Sarcomas: Chest X-Ray or CT Scan? , 2019, Journal of the American College of Surgeons.

[10]  Haibo Hu,et al.  Automated Lung Nodule Detection and Classification Using Deep Learning Combined with Multiple Strategies , 2019, Sensors.

[11]  Luke Oakden-Rayner,et al.  Exploring large scale public medical image datasets , 2019, Academic radiology.

[12]  Shouliang Qi,et al.  Detection and Classification of Pulmonary Nodules Using Convolutional Neural Networks: A Survey , 2019, IEEE Access.

[13]  Raymond Y Huang,et al.  Artificial intelligence in cancer imaging: Clinical challenges and applications , 2019, CA: a cancer journal for clinicians.

[14]  Eui Jin Hwang,et al.  Development and Validation of Deep Learning-based Automatic Detection Algorithm for Malignant Pulmonary Nodules on Chest Radiographs. , 2019, Radiology.

[15]  Kai Li,et al.  Object Detection Based on YOLO Network , 2018, 2018 IEEE 4th Information Technology and Mechatronics Engineering Conference (ITOEC).

[16]  Theerawit Wilaiprasitporn,et al.  Automatic Lung Cancer Prediction from Chest X-ray Images Using the Deep Learning Approach , 2018, 2018 11th Biomedical Engineering International Conference (BMEiCON).

[17]  Richard K. G. Do,et al.  Convolutional neural networks: an overview and application in radiology , 2018, Insights into Imaging.

[18]  Ali Farhadi,et al.  You Only Look Once: Unified, Real-Time Object Detection , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[19]  N. Karssemeijer,et al.  Computer-aided detection improves detection of pulmonary nodules in chest radiographs beyond the support by bone-suppressed images. , 2014, Radiology.

[20]  David H. Foos,et al.  Investigation of the Variability in the Assessment of Digital Chest X-ray Image Quality , 2013, Journal of Digital Imaging.

[21]  Jin Mo Goo,et al.  Ground-glass nodules on chest CT as imaging biomarkers in the management of lung adenocarcinoma. , 2011, AJR. American journal of roentgenology.

[22]  N. Müller,et al.  Fleischner Society: glossary of terms for thoracic imaging. , 2008, Radiology.

[23]  W. Herring,et al.  Learning Radiology: Recognizing the Basics , 2007 .

[24]  P. Eng,et al.  Positron emission tomography with CT in the evaluation of non‐small cell lung cancer in populations with a high prevalence of tuberculosis , 2006, Respirology.

[25]  R. H. Albert,et al.  Evaluation of the solitary pulmonary nodule. , 2009, American family physician.

[26]  M. Okada,et al.  Effect of tumor size on prognosis in patients with non-small cell lung cancer: the role of segmentectomy as a type of lesser resection. , 2005, The Journal of thoracic and cardiovascular surgery.

[27]  H. Moss,et al.  The solitary pulmonary nodule. , 2003, European journal of radiology.

[28]  Giampaolo Gavelli,et al.  Prevention and Early Diagnosis of Lung Cancer , 2000 .

[29]  C. Pistitsch,et al.  Usefulness of morphological characteristics for the differentiation of benign from malignant solitary pulmonary lesions using HRCT , 1999, European Radiology.

[30]  J. Coebergh,et al.  Prevalence of co-morbidity in lung cancer patients and its relationship with treatment: a population-based study. , 1998, Lung cancer.

[31]  Feng Li,et al.  Mass screening for lung cancer with mobile spiral computed tomography scanner , 1998, The Lancet.

[32]  A. Vuagnat,et al.  Nosocomial pneumonia in patients with acute respiratory distress syndrome. , 1998, American journal of respiratory and critical care medicine.

[33]  J. Jett,et al.  Primary mediastinal tumors: part II. Tumors of the middle and posterior mediastinum. , 1997, Chest.

[34]  D. Venzon,et al.  Obstructive atelectasis in patients with small cell lung cancer. Incidence and response to treatment. , 1997, Chest.

[35]  P. Molina,et al.  Imaging Evaluation of Obstructive Atelectasis , 1996, Journal of thoracic imaging.

[36]  N L Müller,et al.  Acute pulmonary complications in immunocompromised non-AIDS patients: comparison of diagnostic accuracy of CT and chest radiography. , 1993, Clinical radiology.

[37]  W. E. Miller,et al.  Lung cancer detected during a screening program using four-month chest radiographs. , 1983, Radiology.