A Community-based Pulmonary Nodule Clinic: Improving Lung Cancer Stage at Diagnosis

Diagnostic value of artificial intelligence based on computed tomography (CT) density in benign and malignant pulmonary nodules: a retrospective investigation

Objective

To evaluate the diagnostic value of artificial intelligence (AI) in the detection and management of benign and malignant pulmonary nodules (PNs) using computed tomography (CT) density.

Methods

A retrospective analysis was conducted on the clinical data of 130 individuals diagnosed with PNs based on pathological confirmation. The utilization of AI and physicians has been employed in the diagnostic process of distinguishing benign and malignant PNs. The CT images depicting PNs were integrated into AI-based software. The gold standard for evaluating the accuracy of AI diagnosis software and physician interpretation was the pathological diagnosis.

Results

Out of 226 PNs screened from 130 patients diagnosed by AI and physician reading based on CT, 147 were confirmed by pathology. AI had a sensitivity of 94.69% and radiologists had a sensitivity of 85.40% in identifying PNs. The chi-square analysis indicated that the screening capacity of AI was superior to that of physician reading, with statistical significance (p < 0.05). 195 of the 214 PNs suggested by AI were confirmed pathologically as malignant, and 19 were identified as benign; among the 29 PNs suggested by AI as low risk, 13 were confirmed pathologically as malignant, and 16 were identified as benign. From the physician reading, 193 PNs were identified as malignant, 183 were confirmed malignant by pathology, and 10 appeared benign. Physician reading also identified 30 low-risk PNs, 19 of which were pathologically malignant and 11 benign. The physician readings and AI had kappa values of 0.432 and 0.547, respectively. The physician reading and AI area under curves (AUCs) were 0.814 and 0.798, respectively. Both of the diagnostic techniques had worthy diagnostic value, as indicated by their AUCs of >0.7.

Conclusion

It is anticipated that the use of AI-based CT diagnosis in the detection of PNs would increase the precision in early detection of lung carcinoma, as well as yield more precise evidence for clinical management.

Establishing a Pulmonary Nodule Clinic Service for Early Diagnosis of Lung Cancer - Review of International Options and Considerations for Greece

Early diagnosis of lung cancer in pulmonary nodules identified by computed tomography (CT) may be critical in reducing the epidemiological burden of the disease, particularly in countries where such a burden is considerably high and risk factors for lung cancer very prevalent. The establishment and operation of pulmonary nodule clinics (PNCs), ie, multidisciplinary services that watch and evaluate nodules found through deliberate screening efforts or as incidental findings, is increasingly becoming a key tool to implement such early-intervention, cancer-risk management policies elsewhere in the world. This review aims to research and present in a structured manner findings from published sources on options and considerations for setting up a PNC in a country such as Greece. These refer to the type of services a PNC would provide to optimize diagnosis of suspect pulmonary nodules, its structure and organization, including processes, human resources and technology infrastructure, its target audience, ie, who would be eligible to use its services, and the expected outcomes of its operation, in terms of a set of key performance indicators. Our review also revealed critical key success factors that should be considered when designing the introduction of a PNC in a health care setting, including optimal referral pathways, aligned clinical decision making and patient preferences and participation/empowerment. Our findings may inform health care systems with a high lung cancer burden and no available PNC service on options and considerations before introducing such a service in their respective settings.

Lung Cancer Screening in Greece: A Modelling Study to Estimate the Impact on Lung Cancer Life Years

(1) Background: Lung cancer causes a substantial epidemiological burden in Greece. Yet, no formal national lung cancer screening program has been introduced to date. This study modeled the impact on lung cancer life years (LCLY) of a hypothetical scenario of comprehensive screening for lung cancer with low-dose computed tomography (LDCT) of the high-risk population in Greece, as defined by the US Preventive Services Taskforce, would be screened and linked to care (SLTC) for lung cancer versus the current scenario of background (opportunistic) screening only; (2) Methods: A stochastic model was built to monitor a hypothetical cohort of 100,000 high-risk men and women as they transitioned between health states (without cancer, with cancer, alive, dead) over 5 years. Transition probabilities were based on clinical expert opinion. Cancer cases, cancer-related deaths, and LCLYs lost were modeled in current and hypothetical scenarios. The difference in outcomes between the two scenarios was calculated. 150 iterations of simulation scenarios were conducted for 100,000 persons; (3) Results: Increasing SLTC to a hypothetical 100% of eligible high-risk people in Greece leads to a statistically significant reduction in deaths and in total years lost due to lung cancer, when compared with the current SLTC paradigm. Over 5 years, the model predicted a difference of 339 deaths and 944 lost years between the hypothetical and current scenario. More specifically, the hypothetical scenario led to fewer deaths (−24.56%, p < 0.001) and fewer life years lost (−31.01%, p < 0.001). It also led to a shift to lower-stage cancers at the time of diagnosis; (4) Conclusions: Our study suggests that applying a 100% screening strategy amongst high-risk adults aged 50−80, would result in additional averted deaths and LCLYs gained over 5 years in Greece.

Convolutional Neural Network ensembles for accurate lung nodule malignancy prediction 2 years in the future

Convolutional Neural Networks (CNNs) have been utilized for to distinguish between benign lung nodules and those that will become malignant. The objective of this study was to use an ensemble of CNNs to predict which baseline nodules would be diagnosed as lung cancer in a second follow up screening after more than one year. Low-dose helical computed tomography images and data were utilized from the National Lung Screening Trial (NLST). The malignant nodules and nodule positive controls were divided into training and test cohorts. T0 nodules were used to predict lung cancer incidence at T1 or T2. To increase the sample size, image augmentation was performed using rotations, flipping, and elastic deformation. Three CNN architectures were designed for malignancy prediction, and each architecture was trained using seven different seeds to create the initial weights. This enabled variability in the CNN models which were combined to generate a robust, more accurate ensemble model. Augmenting images using only rotation and flipping and training with images from T0 yielded the best accuracy to predict lung cancer incidence at T2 from a separate test cohort (Accuracy = 90.29%; AUC = 0.96) based on an ensemble 21 models. Images augmented by rotation and flipping enabled effective learning by increasing the relatively small sample size. Ensemble learning with deep neural networks is a compelling approach that accurately predicted lung cancer incidence at the second screening after the baseline screen mostly 2 years later.