Lung cancer screening with CT: Mayo Clinic experience

Increased Utilization of Low-Dose CT for Lung Cancer Screening at an Arkansas Community Oncology Clinic

BACKGROUND

Low-dose CT (LDCT) is underused in Arkansas for lung cancer screening, a rural state with a high incidence of lung cancer. The objective was to determine whether offering free LDCT increased the number of high-risk individuals screened in a rural catchment area.

METHODS

There were 5,402 patients enrolled in screening at Highlands Oncology, a community oncology clinic in Northwest Arkansas, from 2013 to 2020. Screenings were separated into time periods: period 1 (10 months for-fee), period 2 (10 months free with targeted advertisements and primary care outreach), and period 3 (62 months free with only primary care outreach). In all, 5,035 high-risk participants were eligible for analysis based on National Comprehensive Cancer Network Clinical Practice Guidelines in Oncology. Enrollment rates, incidence densities (IDs), Cox proportional hazard models, and Kaplan-Meier curves were performed to investigate differences between enrollment periods and high-risk groups.

RESULTS

Patient volume increased drastically once screenings were offered free of charge (period 1 = 4.6 versus period 2 = 66.0 and period 3 = 69.8 average patients per month). Incidence density per 1,000 person-years increased through each period (IDPeriod 1 = 17.2; IDPeriod 2 = 20.8; IDPeriod 3 = 25.5 cases). Cox models revealed significant differences in lung cancer risk between high-risk groups (P = .012) but not enrollment periods (P = .19). Kaplan-Meier lung cancer-free probabilities differed significantly between high-risk groups (log-rank P = .00068) but not enrollment periods (log-rank P = .18).

CONCLUSIONS

This study suggests that eligible patients are more receptive to free LDCT screening, despite most insurances not having a required copay for eligible patients.

Evaluation of Pulmonary Nodules by Radiologists vs. Radiomics in Stand-Alone and Complementary CT and MRI

Increased attention has been given to MRI in radiation-free screening for malignant nodules in recent years. Our objective was to compare the performance of human readers and radiomic feature analysis based on stand-alone and complementary CT and MRI imaging in classifying pulmonary nodules. This single-center study comprises patients with CT findings of pulmonary nodules who underwent additional lung MRI and whose nodules were classified as benign/malignant by resection. For radiomic features analysis, 2D segmentation was performed for each lung nodule on axial CT, T2-weighted (T2w), and diffusion (DWI) images. The 105 extracted features were reduced by iterative backward selection. The performance of radiomics and human readers was compared by calculating accuracy with Clopper-Pearson confidence intervals. Fifty patients (mean age 63 +/- 10 years) with 66 pulmonary nodules (40 malignant) were evaluated. ACC values for radiomic features analysis vs. radiologists based on CT alone (0.68; 95%CI: 0.56, 0.79 vs. 0.59; 95%CI: 0.46, 0.71), T2w alone (0.65; 95%CI: 0.52, 0.77 vs. 0.68; 95%CI: 0.54, 0.78), DWI alone (0.61; 95%CI:0.48, 0.72 vs. 0.73; 95%CI: 0.60, 0.83), combined T2w/DWI (0.73; 95%CI: 0.60, 0.83 vs. 0.70; 95%CI: 0.57, 0.80), and combined CT/T2w/DWI (0.83; 95%CI: 0.72, 0.91 vs. 0.64; 95%CI: 0.51, 0.75) were calculated. This study is the first to show that by combining quantitative image information from CT, T2w, and DWI datasets, pulmonary nodule assessment through radiomics analysis is superior to using one modality alone, even exceeding human readers' performance.

Potential added value of an AI software with prediction of malignancy for the management of incidental lung nodules

Purpose

To determine the impact of an artificial intelligence software predicting malignancy in the management of incidentally discovered lung nodules.

Materials and methods

In this retrospective study, all lung nodules ≥ 6 mm and ≤ 30 mm incidentally discovered on emergency CT scans performed between June 1, 2017 and December 31, 2017 were assessed. Artificial intelligence software using deep learning algorithms was applied to determine their likelihood of malignancy: most likely benign (AI score < 50%), undetermined (AI score 50-75%) or probably malignant (AI score > 75%). Predictions were compared to two-year follow-up and Brock's model.

Results

Ninety incidental pulmonary nodules in 83 patients were retrospectively included. 36 nodules were benign, 13 were malignant and 41 remained indeterminate at 2 years follow-up.AI analysis was possible for 81/90 nodules. The 34 benign nodules had an AI score between 0.02% and 96.73% (mean = 48.05 ± 37.32), while the 11 malignant nodules had an AI score between 82.89% and 100% (mean = 93.9 ± 2.3). The diagnostic performance of the AI software for positive diagnosis of malignant nodules using a 75% malignancy threshold was: sensitivity = 100% [95% CI 72%-100%]; specificity = 55.8% [38-73]; PPV = 42.3% [23-63]; NPV = 100% [82-100]. With its apparent high NPV, the addition of an AI score to the initial CT could have avoided a guidelines-recommended follow-up in 50% of the benign pulmonary nodules (6/12 nodules).

Conclusion

Artificial intelligence software using deep learning algorithms presents a strong NPV (100%, with a 95% CI 82-100), suggesting potential use for reducing the need for follow-up of nodules categorized as benign.

Tips and Tricks in Thoracic Radiology for Beginners: A Findings-Based Approach

This review has the purpose of illustrating schematically and comprehensively the key concepts for the beginner who approaches chest radiology for the first time. The approach to thoracic imaging may be challenging for the beginner due to the wide spectrum of diseases, their overlap, and the complexity of radiological findings. The first step consists of the proper assessment of the basic imaging findings. This review is divided into three main districts (mediastinum, pleura, focal and diffuse diseases of the lung parenchyma): the main findings will be discussed in a clinical scenario. Radiological tips and tricks, and relative clinical background, will be provided to orient the beginner toward the differential diagnoses of the main thoracic diseases.

Improving the efficiency of identifying malignant pulmonary nodules before surgery via a combination of artificial intelligence CT image recognition and serum autoantibodies

OBJECTIVE

To construct a new pulmonary nodule diagnostic model with high diagnostic efficiency, non-invasive and simple to measure.

METHODS

This study included 424 patients with radioactive pulmonary nodules who underwent preoperative 7-autoantibody (7-AAB) panel testing, CT-based AI diagnosis, and pathological diagnosis by surgical resection. The patients were randomly divided into a training set (n = 212) and a validation set (n = 212). The nomogram was developed through forward stepwise logistic regression based on the predictive factors identified by univariate and multivariate analyses in the training set and was verified internally in the verification set.

RESULTS

A diagnostic nomogram was constructed based on the statistically significant variables of age as well as CT-based AI diagnostic, 7-AAB panel, and CEA test results. In the validation set, the sensitivity, specificity, positive predictive value, and AUC were 82.29%, 90.48%, 97.24%, and 0.899 (95%[CI], 0.851-0.936), respectively. The nomogram showed significantly higher sensitivity than the 7-AAB panel test result (82.29% vs. 35.88%, p < 0.001) and CEA (82.29% vs. 18.82%, p < 0.001); it also had a significantly higher specificity than AI diagnosis (90.48% vs. 69.04%, p = 0.022). For lesions with a diameter of ≤ 2 cm, the specificity of the Nomogram was higher than that of the AI diagnostic system (90.00% vs. 67.50%, p = 0.022).

CONCLUSIONS

Based on the combination of a 7-AAB panel, an AI diagnostic system, and other clinical features, our Nomogram demonstrated good diagnostic performance in distinguishing lung nodules, especially those with ≤ 2 cm diameters.

KEY POINTS

• A novel diagnostic model of lung nodules was constructed by combining high-specific tumor markers with a high-sensitivity artificial intelligence diagnostic system. • The diagnostic model has good diagnostic performance in distinguishing malignant and benign pulmonary nodules, especially for nodules smaller than 2 cm. • The diagnostic model can assist the clinical decision-making of pulmonary nodules, with the advantages of high diagnostic efficiency, noninvasive, and simple measurement.

Deep Learning-Based Digitally Reconstructed Tomography of the Chest in the Evaluation of Solitary Pulmonary Nodules: A Feasibility Study

RATIONALE AND OBJECTIVES

Computed tomography (CT) is preferred for evaluating solitary pulmonary nodules (SPNs) but access or availability may be lacking, in addition, overlapping anatomy can hinder detection of SPNs on chest radiographs. We developed and evaluated the clinical feasibility of a deep learning algorithm to generate digitally reconstructed tomography (DRT) images of the chest from digitally reconstructed frontal and lateral radiographs (DRRs) and use them to detect SPNs.

METHODS

This single-institution retrospective study included 637 patients with noncontrast helical CT of the chest (mean age 68 years, median age 69 years, standard deviation 11.7 years; 355 women) between 11/2012 and 12/2020, with SPNs measuring 10-30 mm. A deep learning model was trained on 562 patients, validated on 60 patients, and tested on the remaining 15 patients. Diagnostic performance (SPN detection) from planar radiography (DRRs and CT scanograms, PR) alone or with DRT was evaluated by two radiologists in an independent blinded fashion. The quality of the DRT SPN image in terms of nodule size and location, morphology, and opacity was also evaluated, and compared to the ground-truth CT images RESULTS: Diagnostic performance was higher from DRT plus PR than from PR alone (area under the receiver operating characteristic curve 0.95-0.98 versus 0.80-0.85; p < 0.05). DRT plus PR enabled diagnosis of SPNs in 11 more patients than PR alone. Interobserver agreement was 0.82 for DRT plus PR and 0.89 for PR alone; and interobserver agreement for size and location, morphology, and opacity of the DRT SPN was 0.94, 0.68, and 0.38, respectively.

CONCLUSION

For SPN detection, DRT plus PR showed better diagnostic performance than PR alone. Deep learning can be used to generate DRT images and improve detection of SPNs.

Diagnostic utility of trefoil factor families for the early detection of lung cancer and their correlation with tissue expression

Trefoil factors (TFFs) are upregulated in numerous types of cancer, including those of the breast, the colon, the lung and the pancreas, suggesting their potential utility as biomarkers for screening. In the present study, the clinical relevance of serum or urinary TFFs as biomarkers were comprehensively evaluated and the correlation with TFF expression levels in lung cancer tissue was examined. Serum and urine were collected from 199 patients with lung cancer and 198 healthy individuals. Concentrations of serum and urinary TFF1, TFF2 and TFF3 were measured using ELISA and the potential of TFF levels to discriminate between cancer and non-cancer samples was evaluated. In 100 of the cancer cases, expression of TFF1-3 was analyzed using immunohistochemical staining of paraffin sections. Furthermore, the relationship between TFF levels and clinicopathological factors among these cancer cases was analyzed using immunohistochemistry of tissue specimens, quantified and statistically analyzed. While serum levels of all TFFs measured using ELISA were significantly higher in patients with lung cancer compared with those in healthy individuals, urinary TFFs were lower. Areas under the curve (AUC) of the receiver operating characteristic curves for serum/urinary TFF1, TFF2 and TFF3 were 0.709/0.594, 0.722/0.501 and 0.663/0.665, respectively. Furthermore, the combination of serum TFF1, TFF2, TFF3 and urinary TFF1 and TFF3 demonstrated the highest AUC (0.826). In the clinicopathological analysis, serum TFF1 was higher in the early pathological T-stage (pTis/1/2) compared with the later stage (pT3/4) and TFF2 was higher in the pN0/1 than the pN2 group. With regards to the histological types, urinary TFF1 was higher in squamous cell carcinoma than adenocarcinoma (AC), but TFF2 tended to be higher in AC. Using immunohistochemical analysis, although TFF1 and TFF3 expression showed positive correlation with serum concentrations, TFF2 was inversely correlated. In conclusion, serum and urinary TFF levels are promising predictive biomarkers, and their measurements provide a useful in vivo and non-invasive diagnostic screening tool. In particular, TFF1 and TFF3 could be surrogate markers of clinicopathological profiles of human lung cancer.

Low muscle mass in lung cancer is associated with an inflammatory and immunosuppressive tumor microenvironment

BACKGROUND

Computed tomographies (CT) are useful for identifying muscle loss in non-small lung cancer (NSCLC) cachectic patients. However, we lack consensus on the best cutoff point for pectoralis muscle loss. We aimed to characterize NSCLC patients based on muscularity, clinical data, and the transcriptional profile from the tumor microenvironment to build a cachexia classification model.

METHODS

We used machine learning to generate a muscle loss prediction model, and the tumor's cellular and transcriptional profile was characterized in patients with low muscularity. First, we measured the pectoralis muscle area (PMA) of 211 treatment-naive NSCLC patients using CT available in The Cancer Imaging Archive. The cutoffs were established using machine learning algorithms (CART and Cutoff Finder) on PMA, clinical, and survival data. We evaluated the prediction model in a validation set (36 NSCLC). Tumor RNA-Seq (GSE103584) was used to profile the transcriptome and cellular composition based on digital cytometry.

RESULTS

CART demonstrated that a lower PMA was associated with a high risk of death (HR = 1.99). Cutoff Finder selected PMA cutoffs separating low-muscularity (LM) patients based on the risk of death (P-value = 0.003; discovery set). The cutoff presented 84% of success in classifying low muscle mass. The high risk of LM patients was also found in the validation set. Tumor RNA-Seq revealed 90 upregulated secretory genes in LM that potentially interact with muscle cell receptors. The LM upregulated genes enriched inflammatory biological processes. Digital cytometry revealed that LM patients presented high proportions of cytotoxic and exhausted CD8+ T cells.

CONCLUSIONS

Our prediction model identified cutoffs that distinguished patients with lower PMA and survival with an inflammatory and immunosuppressive TME enriched with inflammatory factors and CD8+ T cells.

Reporting and management of incidental lung findings on computed tomography: beyond lung nodules

Non-nodular incidental lung findings can broadly be categorised as airway- or airspace-related abnormalities and diffuse parenchymal abnormalities. Airway-related abnormalities include bronchial dilatation and thickening, foci of low attenuation, emphysema, and congenital variants. Diffuse parenchymal abnormalities relate to the spectrum of diffuse parenchymal lung diseases cover a spectrum from interstitial lung abnormalities (ILAs) and pulmonary cysts to established diffuse parenchymal lung abnormalities such as the idiopathic interstitial pneumonias and cystic lung diseases. In this review, we discuss the main manifestations of these incidental findings, paying attention to their prevalence and importance, descriptors to use when reporting, the limits of what can be considered "normal", and conclude each section with some pragmatic reporting recommendations. We also highlight technical and patient factors which can lead to spurious abnormalities.

AI-Aided Disease Prediction in Visualized Medicine

Artificial intelligence (AI) is playing a vitally important role in promoting the revolution of future technology. Healthcare is one of the promising applications in AI, which covers medical imaging, diagnosis, robotics, disease prediction, pharmacy, health management, and hospital management. Numbers of achievements that made in these fields overturn every aspect in traditional healthcare system. Therefore, to understand the state-of-art AI in healthcare, as well as the chances and obstacles in its development, the applications of AI in disease detection and outlook and the future trends of AI-aided disease prediction were discussed in this chapter.

Navigating patient journey in early diagnosis of lung cancer in India

Lung cancer (LC) is one of the leading causes of cancer deaths worldwide. In India, the incidence of LC is increasing rapidly, and a majority of the patients are diagnosed at advanced stages of the disease when treatment is less likely to be effective. Recent therapeutic developments have significantly improved survival outcomes in patients with LC. Prompt specialist referral remains critical for early diagnosis for improved patient survival. In the Indian scenario, distinguishing LC from benign and endemic medical conditions such as tuberculosis can pose a challenge. Hence, awareness regarding the red flags-signs and symptoms that warrant further investigations and referral-is vital. This review is an effort toward encouraging general physicians to maintain a high index of clinical suspicion for those at risk of developing LC and assisting them in refering patients with concerning symptoms to specialists or multidisciplinary teams as early as possible.

EarlyCDT Lung blood test for risk classification of solid pulmonary nodules: systematic review and economic evaluation

BACKGROUND

EarlyCDT Lung (Oncimmune Holdings plc, Nottingham, UK) is a blood test to assess malignancy risk in people with solid pulmonary nodules. It measures the presence of seven lung cancer-associated autoantibodies. Elevated levels of these autoantibodies may indicate malignant disease. The results of the test might be used to modify the risk of malignancy estimated by existing risk calculators, including the Brock and Herder models.

OBJECTIVES

The objectives were to determine the diagnostic accuracy, clinical effectiveness and cost-effectiveness of EarlyCDT Lung; and to develop a conceptual model and identify evidence requirements for a robust cost-effectiveness analysis.

DATA SOURCES

MEDLINE (including Epub Ahead of Print, In-Process & Other Non-Indexed Citations, Ovid MEDLINE Daily and Ovid MEDLINE), EMBASE, Cochrane Central Register of Controlled Trials, Science Citation Index, EconLit, Cochrane Database of Systematic Reviews, Database of Abstracts of Reviews of Effects, Health Technology Assessment database, NHS Economic Evaluation Database ( NHS EED ) and the international Health Technology Assessment database were searched on 8 March 2021.

REVIEW METHODS

A systematic review was performed of evidence on EarlyCDT Lung, including diagnostic accuracy, clinical effectiveness and cost-effectiveness. Study quality was assessed with the quality assessment of diagnostic accuracy studies-2 tool. Evidence on other components of the pulmonary nodule diagnostic pathway (computerised tomography surveillance, Brock risk, Herder risk, positron emission tomography-computerised tomography and biopsy) was also reviewed. When feasible, bivariate meta-analyses of diagnostic accuracy were performed. Clinical outcomes were synthesised narratively. A simulation study investigated the clinical impact of using EarlyCDT Lung. Additional reviews of cost-effectiveness studies evaluated (1) other diagnostic strategies for lung cancer and (2) screening approaches for lung cancer. A conceptual model was developed.

RESULTS

A total of 47 clinical publications on EarlyCDT Lung were identified, but only five cohorts (695 patients) reported diagnostic accuracy data on patients with pulmonary nodules. All cohorts were small or at high risk of bias. EarlyCDT Lung on its own was found to have poor diagnostic accuracy, with a summary sensitivity of 20.2% (95% confidence interval 10.5% to 35.5%) and specificity of 92.2% (95% confidence interval 86.2% to 95.8%). This sensitivity was substantially lower than that estimated by the manufacturer (41.3%). No evidence on the clinical impact of EarlyCDT Lung was identified. The simulation study suggested that EarlyCDT Lung might potentially have some benefit when considering intermediate risk nodules (10-70% risk) after Herder risk analysis. Two cost-effectiveness studies on EarlyCDT Lung for pulmonary nodules were identified; none was considered suitable to inform the current decision problem. The conceptualisation process identified three core components for a future cost-effectiveness assessment of EarlyCDT Lung: (1) the features of the subpopulations and relevant heterogeneity, (2) the way EarlyCDT Lung test results affect subsequent clinical management decisions and (3) how changes in these decisions can affect outcomes. All reviewed studies linked earlier diagnosis to stage progression and stage shift to final outcomes, but evidence on these components was sparse.

LIMITATIONS

The evidence on EarlyCDT Lung among patients with pulmonary nodules was very limited, preventing meta-analyses and economic analyses.

CONCLUSIONS

The evidence on EarlyCDT Lung among patients with pulmonary nodules is insufficient to draw any firm conclusions as to its diagnostic accuracy or clinical or economic value.

FUTURE WORK

Prospective cohort studies, in which EarlyCDT Lung is used among patients with identified pulmonary nodules, are required to support a future assessment of the clinical and economic value of this test. Studies should investigate the diagnostic accuracy and clinical impact of EarlyCDT Lung in combination with Brock and Herder risk assessments. A well-designed cost-effectiveness study is also required, integrating emerging relevant evidence with the recommendations in this report.

STUDY REGISTRATION

This study is registered as PROSPERO CRD42021242248.

FUNDING

This project was funded by the National Institute for Health and Care Research (NIHR) Health Technology Assessment programme and will be published in full in Health Technology Assessment; Vol. 26, No. 49. See the NIHR Journals Library website for further project information.

Computed tomography findings, associated factors, and management of pulmonary nodules in 54,326 healthy individuals

Introduction

To investigate the pulmonary nodules detected by low-dose computed tomography (LDCT), identified factors affecting the size and number of pulmonary nodules (single or multiple), and the pulmonary nodules diagnosed and management as lung cancer in healthy individuals.

Methods

A retrospective analysis was conducted on 54,326 healthy individuals who received chest LDCT screening. According to the results of screening, the detection rates of pulmonary nodules, grouped according to the size and number of pulmonary nodules (single or multiple), and the patients' gender, age, history of smoking, hypertension, and diabetes were statistically analyzed to determine the correlation between each factor and the characteristics of the nodules. The pulmonary nodules in healthy individuals diagnosed with lung cancer were managed with differently protocols.

Results

The detection rate of pulmonary nodules was 38.8% (21,055/54,326). The baseline demographic characteristics of patients with pulmonary nodules were: 58% male and 42% female patients, 25.7% smoking and 74.3% nonsmoking individuals, 40-60 years old accounted for 49%, 54.8% multiple nodules, and 45.2% single nodules, and ≤5-mm size accounted for 80.4%, 6-10 mm for 18.2%, and 11-30 mm for 1.4%. Multiple pulmonary nodules were more common in hypertensive patients. Diabetes is not an independent risk factor for several pulmonary nodules. Of all patients with lung nodules, 26 were diagnosed with lung cancer, accounting for 0.1% of all patients with pulmonary nodules, 0.6% with nodules ≥5 mm, and 2.2% with nodules ≥8 mm, respectively. Twenty-six patients with lung cancer were treated with surgical resection (57.7%), microwave ablation (MWA, 38.5%), and follow-up (3.8%).

Conclusions

LDCT was suitable for large-scale pulmonary nodules screening in healthy individuals, which was helpful for the early detection of suspicious lesions in the lung. In addition to surgical resection, MWA is an option for early lung cancer treatment.

Comparison of Ultrashort TE Lung MRI and HRCT Lungs for Detection of Pulmonary Nodules in Oncology Patients

Purpose The purpose of this study is to evaluate the detection rate of pulmonary nodules in ultrashort echo time (UTE) lung magnetic resonance imaging (MRI) and to compare it with computed tomography (CT) in oncology patients.

Materials and Methods All individuals undergoing radiotherapy/chemotherapy/regular follow-up or visiting the oncology department and referred to radiology department for nodule detection, during the period of 1 year, were subjected to UTE lung MRI using the sequence Flash 3d_spiralvibe coronal 1.25 mm iso and high-resolution CT lungs and the images were analyzed.

Results Among the total number of nodules detected in both lungs of all patients, nodules detected by CT were 241, and nodules detected by MRI were 212. The nodule detection rate by MRI was 87.96%. The detection rate of nodules for size equal to or more than 5 mm was nearly 100%. For nodules less than 5 mm, and equal to or more than 4 mm, MRI showed a comparable detection rate of 75%, while for nodules less than 4 mm, the detection rate was only 25%.

Conclusion Our study results indicate that lung MRI had a near-complete detection rate for nodules equal to or more than 5 mm in size. Hence, in oncology patients who are undergoing regular follow-up of the lung nodules, lung MRI using UTE can replace low-dose CT, which in turn reduces the radiation dose to the patient.

Incidental Cardiac Computed Tomography Findings in Patients Undergoing Atrial Fibrillation Catheter Ablation

Background: Catheter ablation (CA) is an effective technique for the management of atrial fibrillation (AF). Cardiac computed tomography (CCT) is a non-invasive imaging modality that is used as a crucial part of planning before CA procedures which can detect other incidental findings and require further diagnostic investigations.

Objectives: We sought to assess the prevalence and distribution of incidental CCT findings in patients with AF undergoing CA.

Methods: Retrospective analysis over a three-year period (2013-2016) of 218 patients undergoing CCT prior to AF CA. CCT findings were analyzed and incident clinically important findings were reported.

Results: Over the three-year period, 218 patients had undergone CCT. Of these, 28.8% showed clinically significant incidental findings in the chest and upper abdomen. Incidental findings included coronary artery disease (CAD), incomplete cor triatriatum, pericardial effusion, pleural effusion, pulmonary nodules, pulmonary infiltrates, pulmonary mass, thoracic aortic aneurysm, mediastinal nodes, abdominal mass, and liver nodules.

Conclusions: CCT is a cornerstone investigation prior to AF CA and can show multiple incidental findings, thus potentially functioning as a screening method for the detection of other significant conditions. There is still a debate whether further workup is needed or not as most findings will eventually be benign and further investigations could mean financial burden and clinical risks to the patients. Further larger prospective studies are needed with long-term follow-up to determine whether incidental findings on CCT have an impact on the long-term outcomes of patients.

Deep Learning Algorithms for Diagnosis of Lung Cancer: A Systematic Review and Meta-Analysis

We conducted a systematic review and meta-analysis of the diagnostic performance of current deep learning algorithms for the diagnosis of lung cancer. We searched major databases up to June 2022 to include studies that used artificial intelligence to diagnose lung cancer, using the histopathological analysis of true positive cases as a reference. The quality of the included studies was assessed independently by two authors based on the revised Quality Assessment of Diagnostic Accuracy Studies. Six studies were included in the analysis. The pooled sensitivity and specificity were 0.93 (95% CI 0.85−0.98) and 0.68 (95% CI 0.49−0.84), respectively. Despite the significantly high heterogeneity for sensitivity (I2 = 94%, p < 0.01) and specificity (I2 = 99%, p < 0.01), most of it was attributed to the threshold effect. The pooled SROC curve with a bivariate approach yielded an area under the curve (AUC) of 0.90 (95% CI 0.86 to 0.92). The DOR for the studies was 26.7 (95% CI 19.7−36.2) and heterogeneity was 3% (p = 0.40). In this systematic review and meta-analysis, we found that when using the summary point from the SROC, the pooled sensitivity and specificity of DL algorithms for the diagnosis of lung cancer were 93% and 68%, respectively.

High-resolution simulation of B0 field conditions in the human heart from segmented computed tomography images

B₀ inhomogeneity leads to imaging artifacts in cardiac magnetic resonance imaging (MRI), in particular dark band artifacts with steady-state free precession pulse sequences. The limited spatial resolution of MR-derived in vivo B₀ maps and the lack of population data prevent systematic analysis of the problem at hand and the development of optimized B₀ shim strategies. We used readily available clinical computed tomography (CT) images to simulate the B₀ conditions in the human heart at high spatial resolution. Calculated B₀ fields showed consistency with MRI-based B₀ measurements. The B₀ maps for both the simulations and in vivo measurements showed local field inhomogeneities in the vicinity of lung tips with dominant Z3 spherical harmonic terms in the field distribution. The presented simulation approach allows for the derivation of B₀ field conditions at high spatial resolution from CT images and enables the development of subject- and population-specific B₀ shim strategies for the human heart.

Changes in Metabolism as a Diagnostic Tool for Lung Cancer: Systematic Review

Lung cancer is the leading cause of cancer-related mortality worldwide, with five-year survival rates varying from 3–62%. Screening aims at early detection, but half of the patients are diagnosed in advanced stages, limiting therapeutic possibilities. Positron emission tomography-computed tomography (PET-CT) is an essential technique in lung cancer detection and staging, with a sensitivity reaching 96%. However, since elevated 18F-fluorodeoxyglucose (¹⁸F-FDG) uptake is not cancer-specific, PET-CT often fails to discriminate between malignant and non-malignant PET-positive hypermetabolic lesions, with a specificity of only 23%. Furthermore, discrimination between lung cancer types is still impossible without invasive procedures. High mortality and morbidity, low survival rates, and difficulties in early detection, staging, and typing of lung cancer motivate the search for biomarkers to improve the diagnostic process and life expectancy. Metabolomics has emerged as a valuable technique for these pitfalls. Over 150 metabolites have been associated with lung cancer, and several are consistent in their findings of alterations in specific metabolite concentrations. However, there is still more variability than consistency due to the lack of standardized patient cohorts and measurement protocols. This review summarizes the identified metabolic biomarkers for early diagnosis, staging, and typing and reinforces the need for biomarkers to predict disease progression and survival and to support treatment follow-up.

Earlier diagnosis of lung cancer

The purpose of this article is to review options for more rapid diagnosis of lung cancer at an earlier stage, thereby improving survival. These options include screening, allowing general practitioners to refer patients directly to low-dose computed tomography scan instead of a chest X-ray and the abolition of the "visitation filter", i.e. hospital doctors' ability to reject referrals from general practitioners without prior discussion with the referring doctor.

How AI Can Help in the Diagnostic Dilemma of Pulmonary Nodules

Simple Summary

Pulmonary nodules are considered a sign of bronchogenic carcinoma, detecting them early will reduce their progression and can save lives. Lung cancer is the second most common type of cancer in both men and women. This manuscript discusses the current applications of artificial intelligence (AI) in lung segmentation as well as pulmonary nodule segmentation and classification using computed tomography (CT) scans, published in the last two decades, in addition to the limitations and future prospects in the field of AI.

Abstract

Pulmonary nodules are the precursors of bronchogenic carcinoma, its early detection facilitates early treatment which save a lot of lives. Unfortunately, pulmonary nodule detection and classification are liable to subjective variations with high rate of missing small cancerous lesions which opens the way for implementation of artificial intelligence (AI) and computer aided diagnosis (CAD) systems. The field of deep learning and neural networks is expanding every day with new models designed to overcome diagnostic problems and provide more applicable and simply used models. We aim in this review to briefly discuss the current applications of AI in lung segmentation, pulmonary nodule detection and classification.

Associations of systemic inflammation markers with identification of pulmonary nodule and incident lung cancer in Chinese population

Objectives

Neutrophil‐to‐lymphocyte ratio (NLR), platelet‐to‐lymphocyte ratio (PLR), and systemic immune‐inflammation index (SII), easily accessible systemic inflammation response parameters, were reported to associate with poor lung cancer prognosis. However, research on the effects of these markers on the risk of positive nodules (PNs) and lung cancer is limited.

Methods

Participants in this retrospective study were those who had their first computed tomography (CT) screening at Jiangsu Province Hospital's Health Promotion Center between January 1, 2017 and December 31, 2020. We identified PNs (≥6 mm in diameter) from free text of CT reports and lung cancer from medical records. Multivariate logistic analysis was used to assess the association between NLR, PLR, or SII and PNs or lung cancer.

Results

The detected rate of PNs was 9.60% among the 96,476 participants. Age, smoking and body mass index were possible influencing factors for PNs. We observed linear dose‐effect relationship between NLR, PLR, or SII and PNs (p_non‐linear > 0.05). Compared with low quintile, participants with top quintiles of NLR, PLR or SII had an increased risk of PNs, with the adjusted ORs of 1.19 (1.11–1.28), 1.11 (1.04–1.19) or 1.11 (1.03–1.18), respectively. Meanwhile, NLR showed the U‐shaped relationship with lung cancer, with adjusted ORs of 1.40 (1.08–1.81) comparing highest NLR quintile to the third quintile. The high PLR and SII showed significantly associated with lung cancer with adjusted ORs of 1.29 (0.99–1.68) and 1.35 (1.04–1.74) comparing to the lowest quintile.

Conclusions

The high levels of systemic inflammation markers were associated with the risk of positive pulmonary nodules and lung cancer, which suggested systemic immune response may be an important pre‐clinical feature for the early identification of diseases.

Deep Residual Separable Convolutional Neural Network for lung tumor segmentation

Lung cancer is one of the deadliest types of cancers. Computed Tomography (CT) is a widely used technique to detect tumors present inside the lungs. Delineation of such tumors is particularly essential for analysis and treatment purposes. With the advancement in hardware technologies, Machine Learning and Deep Learning methods are outperforming the traditional methods in the field of medical imaging. In order to delineate lung cancer tumors, we have proposed a deep learning-based methodology which includes a maximum intensity projection based pre-processing method, two novel deep learning networks and an ensemble strategy. The two proposed networks named Deep Residual Separable Convolutional Neural Network 1 and 2 (DRS-CNN1 and DRS-CNN2) achieved better performance over the state-of-the-art U-net network and other segmentation networks. For fair comparison, we have evaluated the performances of all networks on Medical Segmentation Decathlon (MSD) and StructSeg 2019 datasets. The DRS-CNN2 achieved a mean Dice Similarity Coefficient (DSC) of 0.649, mean 95 Hausdorff Distance (HD95) of 18.26, mean Sensitivity 0.737 and a mean Precision of 0.765 on independent test sets.

Screening for Lung Cancer: CHEST Guideline and Expert Panel Report

BACKGROUND

Low-dose chest CT screening for lung cancer has become a standard of care in the United States, in large part because of the results of the National Lung Screening Trial (NLST). Additional evidence supporting the net benefit of low-dose chest CT screening for lung cancer, and increased experience in minimizing the potential harms, has accumulated since the prior iteration of these guidelines. Here, we update the evidence base for the benefit, harms, and implementation of low-dose chest CT screening. We use the updated evidence base to provide recommendations where the evidence allows, and statements based on experience and expert consensus where it does not.

METHODS

Approved panelists reviewed previously developed key questions using the Population, Intervention, Comparator, Outcome format to address the benefit and harms of low-dose CT screening, and key areas of program implementation. A systematic literature review was conducted using MEDLINE via PubMed, Embase, and the Cochrane Library on a quarterly basis since the time of the previous guideline publication. Reference lists from relevant retrievals were searched, and additional papers were added. Retrieved references were reviewed for relevance by two panel members. The quality of the evidence was assessed for each critical or important outcome of interest using the Grading of Recommendations, Assessment, Development, and Evaluation approach. Meta-analyses were performed when enough evidence was available. Important clinical questions were addressed based on the evidence developed from the systematic literature review. Graded recommendations and ungraded statements were drafted, voted on, and revised until consensus was reached.

RESULTS

The systematic literature review identified 75 additional studies that informed the response to the 12 key questions that were developed. Additional clinical questions were addressed resulting in seven graded recommendations and nine ungraded consensus statements.

CONCLUSIONS

Evidence suggests that low-dose CT screening for lung cancer can result in a favorable balance of benefit and harms. The selection of screen-eligible individuals, the quality of imaging and image interpretation, the management of screen-detected findings, and the effectiveness of smoking cessation interventions can impact this balance.

Establishing a Cohort and a Biorepository to Identify Biomarkers for Early Detection of Lung Cancer: The Nashville Lung Cancer Screening Trial Cohort

Rationale: A prospective longitudinal cohort of individuals at high risk of developing lung cancer was established to build a biorepository of carefully annotated biological specimens and low-dose computed tomography (LDCT) chest images for derivation and validation of candidate biomarkers for early detection of lung cancer.Objectives: The goal of this study is to characterize individuals with high risk for lung cancer, accumulating valuable biospecimens and LDCT chest scans longitudinally over 5 years.Methods: Participants 55-80 years of age with a 5-year estimated risk of developing lung cancer >1.5% were recruited and enrolled from clinics at the Vanderbilt University Medical Center, Veteran Affairs Medical Center, and Meharry Medical Center. Individual demographic characteristics were assessed via questionnaire at baseline. Participants underwent an LDCT scan, spirometry, sputum cytology, and research bronchoscopy at the time of enrollment. Participants will be followed yearly for 5 years. Positive LDCT scans are followed-up according to standard of care. The clinical, imaging, and biospecimen data are collected prospectively and stored in a biorepository. Participants are offered smoking cessation counseling at each study visit.Results: A total of 480 participants were enrolled at study baseline and consented to sharing their data and biospecimens for research. Participants are followed with yearly clinic visits to collect imaging data and biospecimens. To date, a total of 19 cancers (13 adenocarcinomas, four squamous cell carcinomas, one large cell neuroendocrine, and one small-cell lung cancer) have been identified.Conclusions: We established a unique prospective cohort of individuals at high risk for lung cancer, enrolled at three institutions, for whom full clinical data, well-annotated LDCT scans, and biospecimens are being collected longitudinally. This repository will allow for the derivation and independent validation of clinical, imaging, and molecular biomarkers of risk for diagnosis of lung cancer.Clinical trial registered with ClinicalTrials.gov (NCT01475500).

[Lung Cancer Screening Study in Macao Smoking Individuals]

背景与目的

中国澳门肺癌发病率逐年上升，吸烟人群是肺癌的高发人群，本研究旨在了解中国澳门长期吸烟人群的肺癌发病情况及胸部低剂量计算机断层扫描(low-dose computed tomography, LDCT)肺结节特点。

方法

通过澳门中华医学会会员私家医生推荐及宣传招募中国澳门无症状长期吸烟人士，行胸部LDCT检查，分析肺癌、肺部结节检出率及影像学特点。

结果

符合纳入条件者291例，检出肺癌10例，检出率3.44%(95%CI: 2.78%-4.01%)，其中，肺腺癌5例，鳞癌、小细胞肺癌各2例，腺鳞癌1例。早期肺癌4例，占40%。212例检出肺结节，肺结节总检出率72.9%(95%CI: 67.8%-78.0%); 疑似肺癌结节44例，检出率15.1%(95%CI: 11.0%-19.2%)。单发结节51例，无肺癌检出; 多发结节161例，检出肺癌9例，两组肺癌检出率无统计学差异(P > 0.05)。 < 6 mm实性结节与 < 5 mm非实性结节组168例，未检出肺癌; ≥6 mm实性结节与≥5 mm非实性结节组44例，检出肺癌9例，两组比较有统计学差异(P < 0.05)。

结论

长期吸烟人群中肺癌检出率高，类型以腺癌为主，肺部结节发生率高，当实性结节≥6 mm或非实性结节≥5 mm时，肺癌检出率增高。建议在符合高危因素的男性吸烟人群中推行胸部LDCT筛查肺癌，女性肺癌筛查，应重新界定高危因素。

Screening for Lung Cancer: CHEST Guideline and Expert Panel Report - Executive Summary

BACKGROUND

Low-dose chest CT screening for lung cancer has become a standard of care in the United States, in large part due to the results of the National Lung Screening Trial. Additional evidence supporting the net benefit of low-dose chest CT screening for lung cancer, as well as increased experience in minimizing the potential harms, has accumulated since the prior iteration of these guidelines. Here, we update the evidence base for the benefit, harms, and implementation of low-dose chest CT screening. We use the updated evidence base to provide recommendations where the evidence allows, and statements based on experience and expert consensus where it does not.

METHODS

Approved panelists reviewed previously developed key questions using the PICO (population, intervention, comparator, and outcome) format to address the benefit and harms of low-dose CT screening, as well as key areas of program implementation. A systematic literature review was conducted using MEDLINE via PubMed, Embase, and the Cochrane Library on a quarterly basis since the time of the previous guideline publication. Reference lists from relevant retrievals were searched, and additional papers were added. Retrieved references were reviewed for relevance by two panel members. The quality of the evidence was assessed for each critical or important outcome of interest using the GRADE approach. Meta-analyses were performed where appropriate. Important clinical questions were addressed based on the evidence developed from the systematic literature review. Graded recommendations and un-graded statements were drafted, voted on, and revised until consensus was reached.

RESULTS

The systematic literature review identified 75 additional studies that informed the response to the 12 key questions that were developed. Additional clinical questions were addressed resulting in 7 graded recommendations and 9 ungraded consensus statements.

CONCLUSIONS

Evidence suggests that low-dose CT screening for lung cancer can result in a favorable balance of benefit and harms. The selection of screen-eligible individuals, the quality of imaging and image interpretation, the management of screen detected findings, and the effectiveness of smoking cessation interventions, can impact this balance.

Lung cancer screening: who pays? Who receives? The European perspectives

Lung cancer is the leading cause of cancer-related death worldwide, and its early detection is critical to achieving a curative treatment and to reducing mortality. Low-dose computed tomography (LDCT) is a highly sensitive technique for detecting noninvasive small lung tumors in high-risk populations. We here analyze the current status of lung cancer screening (LCS) from a European point of view. With economic burden of health care in most European countries resting on the state, it is important to reduce costs of screening and improve its effectiveness. Current cost-effectiveness analyses on LCS have indicated a favorable economic profile. The most recently published analysis reported an incremental cost-effectiveness ratio (ICER) of €3,297 per 1 life-year gained adjusted for the quality of life (QALY) and €2,944 per life-year gained, demonstrating a 90% probability of ICER being below €15,000 and a 98.1% probability of being below €25,000. Different risk models have been used to identify the target population; among these, the PLCOM2012 in particular allows for the selection of the population to be screened with high sensitivity. Risk models should also be employed to define screening intervals, which can reduce the general number of LDCT scans after the baseline round. Future perspectives of screening in a European scenario are related to the will of the policy makers to implement policy on a large scale and to improve the effectiveness of a broad screening of smoking-related disease, including cardiovascular prevention, by measuring coronary calcium score on LDCT. The employment of artificial intelligence (AI) in imaging interpretation, the use of liquid biopsies for the characterization of CT-detected undetermined nodules, and less invasive, personalized surgical treatments, will improve the effectiveness of LCS.

Preoperative computer tomography-guided indocyanine green injection is associated with successful localization of small pulmonary nodules

Background

Localization of small pulmonary nodules (SPNs) is challenging in minimally invasive pulmonary resection, and it is unknown whether computer tomography (CT) guided by indocyanine green (ICG) can provide accurate localization with minimal complications.

Methods

We performed a retrospective study of patients who underwent thoracoscopic resection of pulmonary nodules after CT-guided preoperative localization with ICG from May 2019 to May 2020. Demographics, procedural data, postoperative complications, and pathologic information, were collected, and an analysis of the accuracy and complications after surgery was conducted.

Results

In 471 patients, there was a total of 512 peripheral pulmonary nodules that were ≤2 cm in size. The average time for CT-guided percutaneous ICG injection for localization was 18 minutes, and 98.4% (504/512) of the nodules were successfully localized. The average size of the nodules was 9.1 mm, and the average depth from the pleural surface was 8.9 mm. Overall, 5.9% (28/471) of the patients had asymptomatic pneumothorax after localization, but none needed a tube thoracostomy. All the nodules were resected using video-assisted thoracoscopy technique.

Conclusions

Preoperative CT-guided transthoracic ICG injection is safe and feasible for localization of small lung nodules for minimally invasive pulmonary resection. This technique should be considered for preoperative CT-guided localization of small lung nodules.

Global evolution of research on pulmonary nodules: a bibliometric analysis

Aim: To provide a historical and global picture of research concerning lung nodules, compare the contributions of major countries and explore research trends over the past 10 years. Methods: A bibliometric analysis of publications from Scopus (1970-2020) and Web of Science (2011-2020). Results: Publications about pulmonary nodules showed an enormous growth trend from 1970 to 2020. There is a high level of collaboration among the 20 most productive countries and regions, with the USA located at the center of the collaboration network. The keywords 'deep learning', 'artificial intelligence' and 'machine learning' are current hotspots. Conclusions: Abundant research has focused on pulmonary nodules. Deep learning is emerging as a promising tool for lung cancer diagnosis and management.

Use of a Dual Artificial Intelligence Platform to Detect Unreported Lung Nodules

OBJECTIVE

To investigate the performance of Dual-AI Deep Learning Platform in detecting unreported pulmonary nodules that are 6 mm or greater, comprising computer-vision (CV) algorithm to detect pulmonary nodules, with positive results filtered by natural language processing (NLP) analysis of the dictated report.

METHODS

Retrospective analysis of 5047 chest CT scans and corresponding reports. Cases which were both CV algorithm positive (nodule ≥ 6 mm) and NLP negative (nodule not reported), were outputted for review by 2 chest radiologists.

RESULTS

The CV algorithm detected nodules that are 6 mm or greater in 1830 (36.3%) of 5047 cases. Three hundred fifty-five (19.4%) were unreported by the radiologist, as per NLP algorithm. Expert review determined that 139 (39.2%) of 355 cases were true positives (2.8% of all cases). One hundred thirty (36.7%) of 355 cases were unnecessary alerts-vague language in the report confounded the NLP algorithm. Eighty-six (24.2%) of 355 cases were false positives.

CONCLUSIONS

Dual-AI platform detected actionable unreported nodules in 2.8% of chest CT scans, yet minimized intrusion to radiologist's workflow by avoiding alerts for most already-reported nodules.

Lung Nodule Detection based on Faster R-CNN Framework

BACKGROUND

Lung cancer is a worldwide high-risk disease, and lung nodules are the main manifestation of early lung cancer. Automatic detection of lung nodules reduces the workload of radiologists, the rate of misdiagnosis and missed diagnosis. For this purpose, we propose a Faster R-CNN algorithm for the detection of these lung nodules.

METHOD

Faster R-CNN algorithm can detect lung nodules, and the training set is used to prove the feasibility of this technique. In theory, parameter optimization can improve network structure, as well as detection accuracy.

RESULT

Through experiments, the best parameters are that the basic learning rate is 0.001, step size is 70,000, attenuation coefficient is 0.1, the value of Dropout is 0.5, and the value of Batch Size is 64. Compared with other networks for detecting lung nodules, the optimized and improved algorithm proposed in this paper generally improves detection accuracy by more than 20% when compared with the other traditional algorithms.

CONCLUSION

Our experimental results have proved that the method of detecting lung nodules based on Faster R-CNN algorithm has good accuracy and therefore, presents potential clinical value in lung disease diagnosis. This method can further assist radiologists, and also for researchers in the design and development of the detection system for lung nodules.

Lung cancers and pulmonary nodules detected by computed tomography scan: a population-level analysis of screening cohorts

Background

An increasing number and proportion of younger lung cancer patients have been observed worldwide, raising concerns on the optimal age to begin screening. This study aimed to investigate the association between age and findings in initial CT scans.

Methods

We searched for low-dose CT screening cohorts from electronic databases. Single-arm syntheses weighted by sample size were performed to calculate the detection rates of pulmonary nodules, lung cancers (all stages and stage I), and the proportion of stage I diseases in lung cancers. In addition, we included patients who underwent chest CT in our center as a supplementary cohort. The correlation between the detection rates and age was evaluated by the Pearson Correlation Coefficient.

Results

A total of 37 studies involving 163,442 participants were included. We found the detection rates of pulmonary nodules and lung cancers increased with age. However, the proportion of stage I diseases in lung cancers declined with increased starting age and was significantly higher in the 40-year group than in other groups (40 vs. 45, 50, 55, P<0.001). In addition, the ratio of early-stage lung cancer to the number of nodules declined with age. Similarly, in our center, the detection rates of nodules (R²=0.86, P≤0.001), all lung cancer (R²=0.99, P≤0.001) and stage I diseases (R²=0.87, P=0.001) increased with age, while the proportion of stage I diseases consistently declined with age (R²=0.97, P≤0.001).

Conclusions

Starting lung cancer screening at an earlier age is associated with a higher probability of identifying a curable disease, urging future research to determine the optimal starting age.

Screening rate and influential factors of lung cancer with low-dose computed tomography in Asian population: a systematic review and meta-analysis

BACKGROUND

There are currently no studies synthesizing the screening rate and influential factors of low-dose computed tomography (LDCT)-screened lung cancer in Asian population.

METHODS

A systematic review was conducted, using both English and Chinese language databases on March, 2019. The pooled screening rate and estimated odds ratios (ORs) of influential factors were analyzed using random effects models. Subgroup and meta-regression analyses were also employed to explore the heterogeneity.

RESULTS

The pooled LDCT lung cancer screening rate was 1.12% (95% confidence interval (CI): 0.94%, 1.32%), and increased with age. Adenocarcinoma and stage I lung cancer had higher screening rates. Analysis of influential factors in the general population showed that female and elder age (≥50 years) were significantly influencing LDCT lung cancer screening rate (for female, OR = 1.32, 95% CI: 1.15-1.52; for adults ≥ 50 years, OR = 1.94, 95% CI: 1.52-2.49). Meta-regression analysis indicated that the heterogeneity maybe significantly correlated with the sample size, risk population and source of population.

CONCLUSIONS

Unlike European and American populations, female and adults > 50 years rather than smoking adults were positively associated with screening rate in Asian populations. It is important to further study the benefits of lung cancer screening with LDCT in Asian populations.

SRCIN1 Regulated by circCCDC66/miR-211 Is Upregulated and Promotes Cell Proliferation in Non-Small-Cell Lung Cancer

The incidence and mortality of lung cancer were extremely high. The present study showed that SRCIN1 was an oncogene in non-small-cell lung cancer (NSCLC). Public dataset analysis showed SRCIN1 was significantly overexpressed in NSCLC samples. Also, we found that NSCLC patients with higher SRCIN1 expression had shorter OS time by analyzing TCGA, Kaplan-Meier Plotter, GSE30219, GSE50081, and GSE19188 databases. Overexpression or knockdown of SRCIN1 significantly induced or reduced A549 and H1299 cell proliferation. Furthermore, we found SRCIN1 was directly targeted by miR-211. Overexpression or knockdown of miR-211 suppressed or induced SRCIN1 levels in NSCLC. Moreover, we found that miR-211 affected NSCLC cell proliferation through SRCIN1. Previous studies demonstrated that circRNAs could act as miRNA sponges in cancer cells. In this study, we showed that knockdown of circCCDC66 induced expression of miR-211. Luciferase assay demonstrated that miR-211 suppressed the activity of luciferase reporter-contained circCCDC66 sequences. Moreover, knockdown of circCCDC66 significantly inhibited SRCIN1 levels in both A549 and H1299 cells. These results showed that circCCDC66 acted as a miRNA sponge to affect the miR-211/SRCIN1 axis. Of note, we for the first time revealed that circCCDC66 suppression reduced cell proliferation by about 65% in A549 and by about 40% in H1299 cells. We thought this study could provide novel potential biomarkers for NSCLC.

Estimation of malignancy of pulmonary nodules at CT scans: Effect of computer-aided diagnosis on diagnostic performance of radiologists

OBJECTIVES

To develop a computer-aided diagnosis (CAD) system for distinguishing malignant from benign pulmonary nodules on computed tomography (CT) scans, and to assess whether the diagnostic performance of radiologists with different experiences can be improved with the assistant of CAD.

MATERIALS AND METHODS

A total of 857 malignant nodules from 601 patients and 426 benign nodules from 278 patients were retrospectively collected from four hospitals. In this study, we exploited convolutional neural network in the framework of deep learning to classify whether a nodule was benign or malignant. A total of 745 malignant nodules and 370 benign nodules were used as the training data of our CAD system. The remaining 112 malignant nodules and 56 benign nodules were used as the test data. The participants were two senior chest radiologists, two secondary chest radiologists, and two junior radiology residents. The readers estimated the likelihood of malignancy of pulmonary nodules first without and then with CAD output. Receiver-operating characteristic (ROC) curve was used to evaluate readers' diagnostic performance.

RESULTS

When a threshold level of 58% was used to estimate the likelihood of malignancy, the sensitivity, specificity, and diagnostic accuracy values of our CAD scheme alone were 93.8%, 83.9%, and 90.5%, respectively. For all six readers, the mean area under the ROC curve (A_z ) values without and with CAD system were 0.913 and 0.938, respectively. For each reader, there is a large difference in A_z values that assessed without and with CAD system. With CAD output, the readers made correct changes an average of 15.7 times and incorrect changes an average of 2 times.

CONCLUSION

Our CAD system significantly improved the diagnostic performance of readers regardless of their experience levels for assessment of the likelihood of malignancy of pulmonary nodules.

A Shallow Convolutional Neural Network Predicts Prognosis of Lung Cancer Patients in Multi-Institutional CT-Image Data

Lung cancer is the most common fatal malignancy in adults worldwide, and non-small cell lung cancer (NSCLC) accounts for 85% of lung cancer diagnoses. Computed tomography (CT) is routinely used in clinical practice to determine lung cancer treatment and assess prognosis. Here, we developed LungNet, a shallow convolutional neural network for predicting outcomes of NSCLC patients. We trained and evaluated LungNet on four independent cohorts of NSCLC patients from four medical centers: Stanford Hospital (n = 129), H. Lee Moffitt Cancer Center and Research Institute (n = 185), MAASTRO Clinic (n = 311) and Charité - Universitätsmedizin (n=84). We show that outcomes from LungNet are predictive of overall survival in all four independent survival cohorts as measured by concordance indices of 0.62, 0.62, 0.62 and 0.58 on cohorts 1, 2, 3, and 4, respectively. Further, the survival model can be used, via transfer learning, for classifying benign vs malignant nodules on the Lung Image Database Consortium (n = 1010), with improved performance (AUC=0.85) versus training from scratch (AUC=0.82). LungNet can be used as a noninvasive predictor for prognosis in NSCLC patients and can facilitate interpretation of CT images for lung cancer stratification and prognostication.

Deep Learning in CT Images: Automated Pulmonary Nodule Detection for Subsequent Management Using Convolutional Neural Network

PURPOSE

The purpose of this study is to compare the detection performance of the 3-dimensional convolutional neural network (3D CNN)-based computer-aided detection (CAD) models with radiologists of different levels of experience in detecting pulmonary nodules on thin-section computed tomography (CT).

PATIENTS AND METHODS

We retrospectively reviewed 1109 consecutive patients who underwent follow-up thin-section CT at our institution. The 3D CNN model for nodule detection was re-trained and complemented by expert augmentation. The annotations of a consensus panel consisting of two expert radiologists determined the ground truth. The detection performance of the re-trained CAD model and three other radiologists at different levels of experience were tested using a free-response receiver operating characteristic (FROC) analysis in the test group.

RESULTS

The detection performance of the re-trained CAD model was significantly better than that of the pre-trained network (sensitivity: 93.09% vs 38.44%). The re-trained CAD model had a significantly better detection performance than radiologists (average sensitivity: 93.09% vs 50.22%), without significantly increasing the number of false positives per scan (1.64 vs 0.68). In the training set, 922 nodules less than 3 mm in size in 211 patients at high risk were recommended for follow-up CT according to the Fleischner Society Guidelines. Fifteen of 101 solid nodules were confirmed to be lung cancer.

CONCLUSION

The re-trained 3D CNN-based CAD model, complemented by expert augmentation, was an accurate and efficient tool in identifying incidental pulmonary nodules for subsequent management.

Performance of lung cancer screening with low-dose CT in Gejiu, Yunnan: A population-based, screening cohort study

Background

The performance of lung cancer screening with low‐dose computed tomography (CT) (LDCT) in China is uncertain. This study aimed to evaluate the performance of LDCT lung cancer screening in the Chinese setting.

Methods

In 2014, a screening cohort of lung cancer with LDCT was established in Gejiu, Yunnan Province, a screening center of the Lung Cancer Screening Program in Rural China (LungSPRC). Participants received a baseline screening and four rounds of annual screening with LDCT in two local hospitals until June 2019. We analyzed the rates of participation, detection, early detection, and the clinical characteristics of lung cancer.

Results

A total of 2006 participants had complete baseline screening results with a compliance rate of 98.4%. Of these, 1411 were high‐risk and 558 were nonhigh‐risk participants. During this period, 40 lung cancer cases were confirmed, of these, 35 were screen‐detected, four were post‐screening and one was an interval case. The positive rate of baseline and annual screening was 9.7% and 9.0%, while the lung cancer detection rate was 0.4% and 0.6%, respectively. The proportion of early lung cancer increased from 37.5% in T0 to 75.0% in T4. Adenocarcinoma was the most common histological subtype. Lung cancer incidence according to the criteria of LungSPRC and National Lung Cancer Screening Trial (NLST) was 513.31 and 877.41 per 100 000 person‐years, respectively.

Conclusions

The program of lung cancer screening with LDCT showed a successful performance in Gejiu, Yunnan. However, further studies are warranted to refine a high‐risk population who will benefit most from LDCT screening and reduce the high false positive results.

Key points

This study reports the results of lung cancer screening with LDCT in Gejiu, Yunnan, a high‐risk area of lung cancer, and it demonstrates that lung cancer screening with LDCT is effective in detecting early‐stage lung cancer. Our program provides an opportunity to explore the performance of LDCT lung cancer screening in the Chinese context.

MEMRI-based imaging pipeline for guiding preclinical studies in mouse models of sporadic medulloblastoma

PURPOSE

Genetically engineered mouse models of sporadic cancers are critical for studying tumor biology and for preclinical testing of therapeutics. We present an MRI-based pipeline designed to produce high resolution, quantitative information about tumor progression and response to novel therapies in mouse models of medulloblastoma (MB).

METHODS

Sporadic MB was modeled in mice by inducing expression of an activated form of the Smoothened gene (aSmo) in a small number of cerebellar granule cell precursors. aSmo mice were imaged and analyzed at defined time-points using a 3D manganese-enhanced MRI-based pipeline optimized for high-throughput.

RESULTS

A semi-automated segmentation protocol was established that estimates tumor volume in a time-frame compatible with a high-throughput pipeline. Both an empirical, volume-based classifier and a linear discriminant analysis-based classifier were tested to distinguish progressing from nonprogressing lesions at early stages of tumorigenesis. Tumor centroids measured at early stages revealed that there is a very specific location of the probable origin of the aSmo MB tumors. The efficacy of the manganese-enhanced MRI pipeline was demonstrated with a small-scale experimental drug trial designed to reduce the number of tumor associated macrophages and microglia.

CONCLUSION

Our results revealed a high level of heterogeneity between tumors within and between aSmo MB models, indicating that meaningful studies of sporadic tumor progression and response to therapy could not be conducted without an imaging-based pipeline approach.

Evolving the pulmonary nodules diagnosis from classical approaches to deep learning-aided decision support: three decades' development course and future prospect

PURPOSE

Lung cancer is the commonest cause of cancer deaths worldwide, and its mortality can be reduced significantly by performing early diagnosis and screening. Since the 1960s, driven by the pressing needs to accurately and effectively interpret the massive volume of chest images generated daily, computer-assisted diagnosis of pulmonary nodule has opened up new opportunities to relax the limitation from physicians' subjectivity, experiences and fatigue. And the fair access to the reliable and affordable computer-assisted diagnosis will fight the inequalities in incidence and mortality between populations. It has been witnessed that significant and remarkable advances have been achieved since the 1980s, and consistent endeavors have been exerted to deal with the grand challenges on how to accurately detect the pulmonary nodules with high sensitivity at low false-positive rate as well as on how to precisely differentiate between benign and malignant nodules. There is a lack of comprehensive examination of the techniques' development which is evolving the pulmonary nodules diagnosis from classical approaches to machine learning-assisted decision support. The main goal of this investigation is to provide a comprehensive state-of-the-art review of the computer-assisted nodules detection and benign-malignant classification techniques developed over three decades, which have evolved from the complicated ad hoc analysis pipeline of conventional approaches to the simplified seamlessly integrated deep learning techniques. This review also identifies challenges and highlights opportunities for future work in learning models, learning algorithms and enhancement schemes for bridging current state to future prospect and satisfying future demand.

CONCLUSION

It is the first literature review of the past 30 years' development in computer-assisted diagnosis of lung nodules. The challenges indentified and the research opportunities highlighted in this survey are significant for bridging current state to future prospect and satisfying future demand. The values of multifaceted driving forces and multidisciplinary researches are acknowledged that will make the computer-assisted diagnosis of pulmonary nodules enter into the main stream of clinical medicine and raise the state-of-the-art clinical applications as well as increase both welfares of physicians and patients. We firmly hold the vision that fair access to the reliable, faithful, and affordable computer-assisted diagnosis for early cancer diagnosis would fight the inequalities in incidence and mortality between populations, and save more lives.

Lung Cancer Screening, Version 3.2018, NCCN Clinical Practice Guidelines in Oncology

Lung cancer is the leading cause of cancer-related mortality in the United States and worldwide. Early detection of lung cancer is an important opportunity for decreasing mortality. Data support using low-dose computed tomography (LDCT) of the chest to screen select patients who are at high risk for lung cancer. Lung screening is covered under the Affordable Care Act for individuals with high-risk factors. The Centers for Medicare & Medicaid Services (CMS) covers annual screening LDCT for appropriate Medicare beneficiaries at high risk for lung cancer if they also receive counseling and participate in shared decision-making before screening. The complete version of the NCCN Guidelines for Lung Cancer Screening provides recommendations for initial and subsequent LDCT screening and provides more detail about LDCT screening. This manuscript focuses on identifying patients at high risk for lung cancer who are candidates for LDCT of the chest and on evaluating initial screening findings.

Solitary Pulmonary Nodule: Morphological Effects on Metabolic Activity Assessment

Objectives

We aimed to evaluate the effects of morphological characteristics of the solitary pulmonary nodules (SPN) on metabolic activity assessment. To the best of our knowledge, this is the first study to compare the volumetric metabolic activity parameters according to the morphologic parameters of the nodules.

Methods

In this retrospective study, ¹⁸F-FDG positron emission tomography and computed tomography scans performed between 2011 and 2018 were evaluated by a nuclear and diagnostic radiologist. One hundred thirteen patients with SPNs with biopsy-proven diagnosis were included. SPNs were classified as solid, partially solid (PS), and ground glass opacity (GGO).

Results

SPN diameter, SUV_max, metabolic tumor volume (MTV), total lesion glycolysis (TLG), and density were significantly higher in the malignant group. SUV_max, MTV, TLG increased in direct proportion to the diameter. There was no a significant difference between GGO, PS, and solid nodules in terms of SUV_max values. MTV and TLG values increased in parallel with the density of the nodules, but this increase was only significant in the malignant group. There was a significant difference between SPNs <2 cm and SPNs ≥2 cm in terms of MTV, while there was no difference in terms of SUV_max. The cut-off value determined by the ROC curve was found to be 4.39 for SUV_max, 7.33 mL for MTV and 31.88 g for TLG. The cut-off values for SUV_max of solid and subsolid nodules were close to each other, but cut-off values for MTV and TLG were higher in solid nodules.

Conclusion

SUV_max, MTV, and TLG are affected by diameter and attenuation. We suggest using different MTV and TLG cut-off values for solid and subsolid nodules, but we suggest using same SUV_max values. MTV can be a more reliable parameter than SUV_max in prediction of malignancy in smaller nodules.