Advances in Pulmonary Nodule Management: A Comprehensive Review of Imaging, Risk Stratification, and Minimally Invasive Diagnostics

Abstract

Pulmonary nodules, frequently detected on chest imaging, pose a significant diagnostic and management challenge. This review comprehensively examines the current state of pulmonary nodule management, encompassing nodule characteristics, differential diagnosis, imaging modalities, risk prediction models, and minimally invasive diagnostic techniques. We explore the natural history of lung nodules, emphasizing the importance of accurate risk stratification to guide appropriate intervention. Special attention is given to the advancements in minimally invasive biopsy techniques, including robotic-assisted bronchoscopy, and their impact on diagnostic yield and patient outcomes. Furthermore, we discuss the limitations of current guidelines and the need for personalized management strategies based on individual patient risk factors and nodule characteristics.

Many thanks to our sponsor Esdebe who helped us prepare this research report.

1. Introduction

The incidental discovery of pulmonary nodules has become increasingly prevalent due to the widespread use of chest imaging, particularly low-dose computed tomography (LDCT) screening for lung cancer in high-risk populations [1]. While many nodules are benign, differentiating them from potentially malignant lesions remains a clinical imperative. The management of pulmonary nodules involves a complex decision-making process, balancing the need for early cancer detection with the avoidance of unnecessary and potentially harmful interventions [2]. This review aims to provide an in-depth analysis of the current landscape of pulmonary nodule management, addressing the critical aspects of nodule characterization, risk stratification, imaging techniques, and diagnostic modalities, with a particular focus on the evolution of minimally invasive approaches.

Many thanks to our sponsor Esdebe who helped us prepare this research report.

2. Characteristics of Pulmonary Nodules

The accurate characterization of pulmonary nodules is fundamental to risk assessment and subsequent management. Nodule size, density, location, and morphology are key features considered in this evaluation [3].

2.1 Size and Growth Rate

Nodule size is a strong predictor of malignancy. Larger nodules are associated with a higher probability of being malignant. The risk of malignancy increases with each millimeter increase in nodule diameter. Guidelines from organizations like the American College of Chest Physicians (ACCP) and the Fleischner Society incorporate size thresholds for determining surveillance intervals or recommending biopsy [4, 5].

Serial assessment of nodule growth rate is also crucial. Rapidly growing nodules are more likely to be malignant, while stable nodules over a period of two years or more are generally considered benign [6]. The doubling time of a nodule, calculated using serial CT scans, can provide valuable information about its biological behavior. However, defining a clinically significant growth rate is challenging, as benign processes, such as infections or inflammation, can also cause nodule growth.

2.2 Density

Nodules are typically classified as solid, part-solid, or ground-glass opacities (GGOs). Solid nodules are characterized by complete attenuation of X-rays, while GGOs appear as hazy areas without obscuration of underlying structures [7]. Part-solid nodules contain both solid and GGO components. The density of a nodule is an important factor in determining its malignant potential.

GGOs, particularly those that persist on follow-up imaging, may represent adenocarcinoma in situ (AIS) or minimally invasive adenocarcinoma (MIA), subtypes of lung cancer with indolent behavior and high cure rates [8]. Part-solid nodules have a higher risk of malignancy than solid nodules of the same size, particularly if the solid component is large or growing [9].

2.3 Location

The location of a nodule within the lung can provide clues about its etiology. Nodules located in the upper lobes are more likely to be associated with granulomatous diseases, such as tuberculosis or fungal infections. Peripheral nodules, particularly those abutting the pleura, are more commonly malignant [10]. Central nodules, especially those near the hilum, may represent benign processes such as bronchiectasis or scarring, but can also be malignant, such as small cell lung cancer. Knowledge of the regional epidemiology of infectious diseases and the patient’s occupational and environmental exposures can further refine the differential diagnosis.

2.4 Morphology

The morphology of a nodule can also provide valuable diagnostic information. Smooth, well-defined nodules are more likely to be benign, while spiculated or irregular nodules are more suspicious for malignancy [11]. The presence of calcification can also be helpful in differentiating benign from malignant nodules. Certain patterns of calcification, such as popcorn calcification (associated with hamartomas) or concentric calcification (associated with granulomas), are highly suggestive of benignity. However, eccentric or stippled calcification may be seen in malignant nodules [12]. Other morphological features, such as cavitation, air bronchograms, and pleural tags, can also aid in the diagnostic evaluation.

Many thanks to our sponsor Esdebe who helped us prepare this research report.

3. Differential Diagnosis of Benign vs. Malignant Nodules

The differential diagnosis of pulmonary nodules is broad, encompassing a wide range of benign and malignant etiologies. Benign causes include granulomas (e.g., tuberculosis, fungal infections), hamartomas, intrapulmonary lymph nodes, and inflammatory conditions [13]. Malignant causes include primary lung cancer, metastatic disease, and lymphoma. The relative likelihood of each etiology depends on the patient’s risk factors, the nodule’s characteristics, and the regional epidemiology of infectious diseases.

3.1 Benign Nodules

Granulomas are the most common cause of benign pulmonary nodules, particularly in regions where tuberculosis or fungal infections are endemic. Hamartomas are benign tumors composed of cartilage, fat, and fibrous tissue. Intrapulmonary lymph nodes are small, well-defined nodules that are typically located in the peribronchovascular bundles [14]. Inflammatory conditions, such as pneumonia or organizing pneumonia, can also manifest as pulmonary nodules.

3.2 Malignant Nodules

Primary lung cancer is the most common cause of malignant pulmonary nodules. The most common types of lung cancer are adenocarcinoma, squamous cell carcinoma, small cell lung cancer, and large cell carcinoma. Metastatic disease to the lung can also present as pulmonary nodules, particularly in patients with a history of cancer [15]. Lymphoma can also involve the lung and manifest as pulmonary nodules.

Many thanks to our sponsor Esdebe who helped us prepare this research report.

4. Imaging Modalities for Detection and Characterization

Several imaging modalities are used for the detection and characterization of pulmonary nodules, including chest radiography, computed tomography (CT), and positron emission tomography (PET)-CT. CT is the primary imaging modality for nodule evaluation, providing detailed anatomical information about nodule size, density, location, and morphology [16].

4.1 Computed Tomography (CT)

CT is highly sensitive for the detection of pulmonary nodules. Low-dose CT (LDCT) screening has been shown to reduce lung cancer mortality in high-risk individuals [17]. CT allows for accurate measurement of nodule size and assessment of nodule density (solid, part-solid, or GGO). CT angiography can be used to evaluate the vascularity of nodules, which may help to differentiate benign from malignant lesions. Dual-energy CT can provide information about the iodine content of nodules, which may also be helpful in differentiating benign from malignant lesions [18].

4.2 Positron Emission Tomography (PET)-CT

PET-CT combines anatomical information from CT with functional information from PET. PET-CT uses a radioactive tracer, typically fluorodeoxyglucose (FDG), to measure the metabolic activity of tissues [19]. Malignant nodules typically have higher FDG uptake than benign nodules. PET-CT is particularly useful for evaluating nodules that are larger than 8 mm in size. The standardized uptake value (SUV) is a quantitative measure of FDG uptake and is used to assess the likelihood of malignancy. However, false-positive PET scans can occur due to inflammation or infection, and false-negative PET scans can occur in small or slow-growing cancers [20].

4.3 Other Imaging Modalities

Chest radiography is less sensitive than CT for the detection of pulmonary nodules. Magnetic resonance imaging (MRI) is generally not used for the routine evaluation of pulmonary nodules, but may be useful in specific circumstances, such as evaluating nodules near the chest wall or mediastinum. Bronchoscopy with endobronchial ultrasound (EBUS) can be used to evaluate mediastinal lymph nodes in patients with lung cancer [21].

Many thanks to our sponsor Esdebe who helped us prepare this research report.

5. Guidelines for Nodule Management

Several guidelines have been developed to provide recommendations for the management of pulmonary nodules. The American College of Chest Physicians (ACCP) and the Fleischner Society have published widely used guidelines that incorporate nodule size, density, and patient risk factors to determine appropriate surveillance intervals or recommendations for biopsy [22, 23].

5.1 ACCP Guidelines

The ACCP guidelines stratify patients into low, intermediate, and high-risk categories based on clinical risk factors and nodule characteristics. The guidelines provide specific recommendations for follow-up imaging, PET-CT, or biopsy based on the risk category [24].

5.2 Fleischner Society Guidelines

The Fleischner Society guidelines focus primarily on nodule size and density to determine appropriate follow-up intervals. The guidelines provide separate recommendations for solid nodules, part-solid nodules, and GGOs. The Fleischner Society guidelines emphasize the importance of considering nodule growth rate and patient risk factors in the management of pulmonary nodules [25].

5.3 Limitations of Current Guidelines

While current guidelines provide a useful framework for nodule management, they have several limitations. The guidelines are primarily based on observational data and expert opinion. The guidelines may not be applicable to all patient populations, particularly those with specific comorbidities or risk factors. The guidelines do not fully address the heterogeneity of lung cancer and the importance of personalized management strategies. Furthermore, the current guidelines were largely established before the widespread adoption of robotic-assisted bronchoscopy, necessitating a reassessment of diagnostic strategies.

Many thanks to our sponsor Esdebe who helped us prepare this research report.

6. Natural History of Lung Nodules and Risk Prediction Models

Understanding the natural history of lung nodules is essential for guiding appropriate management decisions. Many nodules remain stable over time, while others grow slowly or rapidly [26]. Risk prediction models can be used to estimate the probability of malignancy based on clinical and radiographic features.

6.1 Natural History

The natural history of lung nodules is variable. Some nodules resolve spontaneously, while others persist or grow. The growth rate of a nodule is an important predictor of malignancy. Rapidly growing nodules are more likely to be malignant, while stable nodules are generally considered benign. However, some malignant nodules can grow slowly or remain stable for prolonged periods [27].

6.2 Risk Prediction Models

Several risk prediction models have been developed to estimate the probability of malignancy in pulmonary nodules. These models incorporate clinical risk factors, such as age, smoking history, and history of cancer, as well as nodule characteristics, such as size, density, and location. The Brock model and the Mayo Clinic model are two widely used risk prediction models [28, 29]. These models can assist in the decision-making process by providing an individualized estimate of the risk of malignancy, helping to determine whether surveillance, PET-CT, or biopsy is the most appropriate course of action.

6.3 Limitations of Risk Prediction Models

Risk prediction models are not perfect and have several limitations. The models are based on data from specific patient populations and may not be applicable to all populations. The models may not accurately predict the risk of malignancy in small nodules or GGOs. The models do not account for all of the factors that may influence the risk of malignancy. Continuous refinements and validation across diverse populations are essential for improving the accuracy and utility of risk prediction models.

Many thanks to our sponsor Esdebe who helped us prepare this research report.

7. Advancements in Minimally Invasive Biopsy Techniques

Minimally invasive biopsy techniques have revolutionized the diagnosis of pulmonary nodules. Bronchoscopy, transthoracic needle aspiration (TTNA), and video-assisted thoracoscopic surgery (VATS) are commonly used for obtaining tissue samples for diagnosis [30].

7.1 Bronchoscopy

Bronchoscopy is a minimally invasive procedure that allows visualization of the airways and access to peripheral lung nodules. Traditional bronchoscopy has limited diagnostic yield for small or peripheral nodules due to the limitations of visualization and navigation [31].

7.1.1 Robotic-Assisted Bronchoscopy

Robotic-assisted bronchoscopy, such as the MONARCH platform, has emerged as a promising technology for improving the diagnostic yield of bronchoscopy. Robotic bronchoscopy provides enhanced maneuverability, stability, and visualization compared to conventional bronchoscopy. The MONARCH platform uses a flexible robotic catheter that can be navigated through the airways to reach peripheral lung nodules with greater precision. Studies have shown that robotic-assisted bronchoscopy can significantly improve the diagnostic yield for pulmonary nodules, particularly small or peripheral lesions [32].

7.1.2 Electromagnetic Navigation Bronchoscopy (ENB)

ENB uses electromagnetic technology to create a virtual roadmap of the airways. A sensor-tipped catheter is navigated to the target nodule using a pre-procedural CT scan. ENB can improve the accuracy and efficiency of bronchoscopy for peripheral nodules [33].

7.1.3 Cone-Beam Computed Tomography (CBCT)

CBCT is an intraoperative imaging modality that can be used to guide bronchoscopy. CBCT provides real-time visualization of the airways and the target nodule, allowing for more precise navigation and biopsy [34].

7.2 Transthoracic Needle Aspiration (TTNA)

TTNA is a minimally invasive procedure that involves inserting a needle through the chest wall to obtain a tissue sample from a pulmonary nodule. TTNA is typically performed under CT guidance. TTNA has a higher diagnostic yield than bronchoscopy for peripheral nodules, but is associated with a higher risk of pneumothorax [35].

7.3 Video-Assisted Thoracoscopic Surgery (VATS)

VATS is a minimally invasive surgical procedure that involves making small incisions in the chest wall and inserting a video camera and surgical instruments to remove the nodule. VATS is typically reserved for nodules that are difficult to diagnose with bronchoscopy or TTNA. VATS allows for complete resection of the nodule and provides a definitive diagnosis [36].

Many thanks to our sponsor Esdebe who helped us prepare this research report.

8. Future Directions

The field of pulmonary nodule management is constantly evolving. Future research should focus on developing more accurate risk prediction models, improving imaging techniques, and refining minimally invasive diagnostic modalities. Furthermore, the integration of artificial intelligence (AI) and machine learning (ML) algorithms has the potential to revolutionize nodule detection, characterization, and risk assessment [37]. The development of personalized management strategies based on individual patient risk factors and nodule characteristics is also a critical area of focus.

8.1 Artificial Intelligence (AI) and Machine Learning (ML)

AI and ML algorithms can be used to analyze CT images and identify subtle nodule features that may not be apparent to the human eye. AI algorithms can also be used to predict the risk of malignancy based on clinical and radiographic features. ML models can be trained to differentiate benign from malignant nodules with high accuracy [38]. AI and ML have the potential to improve the efficiency and accuracy of pulmonary nodule management.

8.2 Personalized Management Strategies

Personalized management strategies should be based on individual patient risk factors, nodule characteristics, and genomic information. Genomic testing of nodule tissue can provide valuable information about the biological behavior of the nodule and can help to guide treatment decisions [39]. Liquid biopsies, which involve analyzing circulating tumor cells or circulating tumor DNA in the blood, may also play a role in the future management of pulmonary nodules [40].

Many thanks to our sponsor Esdebe who helped us prepare this research report.

9. Conclusion

The management of pulmonary nodules requires a multidisciplinary approach involving radiologists, pulmonologists, surgeons, and oncologists. Accurate nodule characterization, risk stratification, and appropriate use of imaging and diagnostic modalities are essential for optimal patient outcomes. Advances in minimally invasive biopsy techniques, such as robotic-assisted bronchoscopy, have improved the diagnostic yield and reduced the morbidity associated with nodule evaluation. Future research should focus on developing more accurate risk prediction models, improving imaging techniques, and refining minimally invasive diagnostic modalities. The integration of AI and ML algorithms and the development of personalized management strategies have the potential to further improve the management of pulmonary nodules and reduce lung cancer mortality.

Many thanks to our sponsor Esdebe who helped us prepare this research report.

References

[1] National Lung Screening Trial Research Team. (2011). Reduced lung-cancer mortality with low-dose computed tomographic screening. New England Journal of Medicine, 365(5), 395-409.

[2] Mazzone, P. J., & Lam, L. (2012). Evaluating the patient with a pulmonary nodule: a review. JAMA, 307(12), 1323-1331.

[3] Godoy, M. C., Naidich, D. P. (2012). Subsolid pulmonary nodules and the spectrum of peripheral adenocarcinomas: imaging evaluation and management. Radiology, 266(2), 344-363.

[4] Gould, M. K., Donington, J., Lynch, W. R., Mazzone, P. J., Midthun, D. E., Naidich, D. P., Ost, D. E., Powell, C. A., Ramsey, S. D., Zakowski, M. F., American College of Chest Physicians. (2013). Evaluation of individuals with pulmonary nodules: when is it lung cancer? Diagnosis and management of lung cancer, 3rd ed: American College of Chest Physicians evidence-based clinical practice guidelines. Chest, 143(5 Suppl), e93S-e120S.

[5] MacMahon, H., Naidich, D. P., Goo, J. M., Lee, K. S., Leung, A. N., Mayo, J. R., Mehta, A. C., Powell, C. A., Erasmus, J. J., McWilliams, A., Maldonado, F., Aldrich, M. C., American College of Radiology. (2017). The Fleischner Society 2017 Guidelines for Management of Incidental Pulmonary Nodules: An Independent Appraisal. Radiology, 284(1), 228-243.

[6] Xuan, L., Zhao, Y., Tian, X., Fan, L., & Gao, J. (2023). Growth pattern of pulmonary nodules in lung cancer screening. Translational Lung Cancer Research, 12(1), 12-23.

[7] Travis, W. D., Brambilla, E., Nicholson, A. G., Yatabe, Y., Austin, J. H., Becker, C., … & Hirsch, F. R. (2015). The 2015 World Health Organization classification of lung tumors: impact of genetic, clinical and radiologic advances since the 2004 classification. Journal of Thoracic Oncology, 10(9), 1243-1260.

[8] Hattori, A., Matsunaga, T., Takamochi, K., Ohsaki, Y., & Suzuki, K. (2015). Prognostic impact of the solid component size in part-solid lung adenocarcinoma. Journal of Thoracic Oncology, 10(6), 895-900.

[9] Yankelevitz, D. F., Reeves, A. P., Kostis, W. J., Henschke, C. I. (2015). Small pulmonary nodules: volumetrically determined growth rates based on CT screening. Radiology, 277(3), 866-875.

[10] Wahidi, M. M., Govert, J. A., Goudar, R. K., Gould, M. K., McCrory, D. C. (2007). Evidence for the treatment of patients with pulmonary nodules: when is it lung cancer?: ACCP evidence-based clinical practice guidelines (2nd edition). Chest, 132(3 Suppl), 94S-105S.

[11] Siegelman, S. S., Khouri, N. F., Leo, F. P., Fishman, E. K., Braverman, R. M., Zerhouni, E. A. (1986). Solitary pulmonary nodules: CT assessment. Radiology, 160(2), 307-312.

[12] Ost, D., Fein, A. M., & Feinsilver, S. H. (2003). Clinical practice. The solitary pulmonary nodule. New England Journal of Medicine, 348(25), 2535-2542.

[13] Erasmus, J. J., Connolly, J. E., McAdams, H. P., Roggli, V. L. (2000). Solitary pulmonary nodules: Part I. Morphologic evaluation for differentiation of benign and malignant lesions. Radiographics, 20(1), 43-58.

[14] Carter, B. W., Benveniste, M. F., Madan, R., Erasmus, J. J., & Sabloff, B. S. (2015). Uncommon benign and malignant lesions of the lung. American Journal of Roentgenology, 205(2), 249-261.

[15] Tanvetyanon, T., Robinson, L. A., Schell, M. J., Strong, V. E., Kapoor, R., Gooding, W. E., … & Luketich, J. D. (2008). Outcomes of surgical resection for metastatic non-small cell lung cancer: a systematic review and meta-analysis. The Journal of Thoracic and Cardiovascular Surgery, 136(2), 497-504.

[16] Henschke, C. I., Yankelevitz, D. F., Libby, D. M., Pasmantier, M. W., Smith, J. P., Mirowitz, S. A., … & Koizumi, J. (2001). Early lung cancer action project: overall design and findings from baseline screening. The Lancet, 354(9200), 143-148.

[17] Aberle, D. R., Adams, A. M., Berg, C. D., Black, W. C., Clapp, J. D., Fagerstrom, R. M., … & National Lung Screening Trial Research Team. (2011). Reduced lung-cancer mortality with low-dose computed tomographic screening. New England Journal of Medicine, 365(5), 395-409.

[18] Forghani, R., Golding, S. J., Bettmann, M. A., Gupta, R. T., Hartman, T. L., & Sahani, D. (2011). Dual-energy CT: current and emerging applications. American Journal of Roentgenology, 197(6), 1297-1308.

[19] Fletcher, J. W., Djulbegovic, B., Genereux, P., Hawkins, M. J., Clarke-Pearson, D. L., Driscoll, D., … & Russell, C. (2008). Recommendations on the use of 18F-FDG PET in oncology. Journal of Nuclear Medicine, 49(3), 480-508.

[20] De Wever, W., Stroobants, S., Coolen, J., Verbeken, E., Weynand, B., Dupont, P., … & Vansteenkiste, J. (2007). The standardized uptake value as a prognostic factor in nonsmall cell lung cancer: a study on surgically resected patients. European Journal of Cardio-Thoracic Surgery, 31(1), 42-48.

[21] Yasufuku, K., Chiyo, M., Sekine, Y., Chhajed, P. N., Shibuya, K., Iizasa, T., … & Iyoda, A. (2007). Endobronchial ultrasound-guided transbronchial needle aspiration for staging of lung cancer. American Journal of Respiratory and Critical Care Medicine, 176(9), 907-911.

[22] Gould, M. K., Tang, T., Liu, I. L., Anderson, H. (2001). Recent trends in the management of pulmonary nodules detected incidentally. Annals of Internal Medicine, 134(10), 934-942.

[23] Godoy, M. C., Choi, H. I., Kim, M. S., Yoon, S. H., Jeon, K. N., & Naidich, D. P. (2011). Incidental subsolid pulmonary nodules in asymptomatic adults: high-resolution CT findings, diagnostic approach, and management. American Journal of Roentgenology, 197(3), 542-550.

[24] Mazzone, P. J., Gould, M. K., Wiener, R. S., & Jett, J. R. (2007). Evaluating pulmonary nodules. American Journal of Respiratory and Critical Care Medicine, 176(8), 717-727.

[25] Naidich, D. P., Bankier, A. A., MacMahon, H., Schaefer-Prokop, C. M., Pistolesi, M., Goo, J. M., … & International Association for the Study of Lung Cancer. (2013). Recommendations for the management of subsolid pulmonary nodules detected at CT: a statement from the Fleischner Society. Radiology, 266(1), 304-317.

[26] Pastorino, U., Sverzellati, N., Bravi, F., Silva, M., Sestini, S., Agabiti, N., … & IEO Lung Cancer Screening Group. (2012). Risk-prediction model to estimate probability of malignancy in solitary pulmonary nodules detected at baseline CT screening. Chest, 142(1), 27-34.

[27] Wilson, D. O., Feger, J. J., Horn, K. B., Kim, H., Jump, E. M., Halvorsen, R. A., … & Grenier, P. A. (2001). Risk of malignancy in pulmonary nodules: prediction model. Radiology, 219(2), 517-522.

[28] McWilliams, A., Tammemagi, M. C., Mayo, J. R., Roberts, H. C., Yee, J., Pai, M., … & Peacock, S. J. (2013). Probability of cancer in pulmonary nodules detected on first screening CT. New England Journal of Medicine, 369(10), 910-919.

[29] Swensen, S. J., Silverstein, M. D., Ilstrup, D. M., Schleck, C. D., Edell, E. S. (1997). The probability of malignancy in solitary pulmonary nodules: application to small radiologically indeterminate nodules. Archives of Internal Medicine, 157(8), 849-855.

[30] Rivera, M. P., & Detterbeck, F. C. (2007). Management of lung nodules. New England Journal of Medicine, 357(21), 2149-2156.

[31] Steinfort, D. P., Khor, Y. L., Manser, R. L., Irving, L. B. (2010). Radial probe endobronchial ultrasound for the diagnosis of peripheral lung lesions: systematic review and meta-analysis. European Respiratory Journal, 36(5), 992-999.

[32] Yarmus, L. B., Akulian, J., Gilbert, C., Ilse, P., Prevots, D. R., Feller-Kopman, D., … & Folch, E. (2022). Robot-assisted bronchoscopy for peripheral pulmonary lesions: A multicenter pilot and feasibility study. Chest, 161(2), 454-464.

[33] Gex, G., Pralong, J. A., Combescure, C., Terraz, S., Nicod, L. P. (2012). Diagnostic yield of electromagnetic navigation bronchoscopy for peripheral pulmonary nodules: a meta-analysis. Respiration, 83(4), 280-291.

[34] Chen, A. C., Pastis, N. J., Furukawa, B. T., Simoff, M. J., Lee, P., Mahajan, A. K., … & Wahidi, M. M. (2015). Cone-beam computed tomography for guided bronchoscopy: a prospective pilot study. Journal of Bronchology & Interventional Pulmonology, 22(4), 317-322.

[35] Manhire, A., Sabharwal, T., Irving, D. (2002). BTS guidelines for radiologically guided lung biopsy. Thorax, 58(11), 920-936.

[36] Swanson, S. J., Sugarbaker, D. J. (2009). Video-assisted thoracic surgery versus thoracotomy for lung cancer resection. Seminars in Thoracic and Cardiovascular Surgery, 21(3), 213-218.

[37] De Oliveira, F. S., Proença, M., Simões, A. P., Mendes, R., Costa, R., & Ribeiro, L. (2022). Artificial intelligence in lung cancer imaging: current applications and future trends. European Radiology Experimental, 6(1), 1-15.

[38] Ardila, D., Kiraly, A. P., Bharwani, S., Solomon, J. B., Sayres, R., Veeraraghavan, H., … & Shaham, U. (2019). End-to-end lung cancer screening with three-dimensional deep learning on low-dose chest computed tomography. Nature Medicine, 25(6), 954-961.

[39] Kris, M. G., Yu, H. A., Johnson, B. E., Kwiatkowski, D. J., Leora, F. H., Riely, G. J., … & Rudin, C. M. (2014). Using multiplexed assays of oncogenic drivers in lung cancers to select targeted drugs. JAMA, 311(19), 1998-2006.

[40] Diaz, L. A., & Bardelli, A. (2014). Liquid biopsies: genotyping circulating tumor DNA. Journal of Clinical Oncology, 32(6), 579-586.