Lung nodules: size still matters

Radiomic-based prediction of lesion-specific systemic treatment response in metastatic disease

Despite sharing the same histologic classification, individual tumors in multi metastatic patients may present with different characteristics and varying sensitivities to anticancer therapies. In this study, we investigate the utility of radiomic biomarkers for prediction of lesion-specific treatment resistance in multi metastatic leiomyosarcoma patients. Using a dataset of n=202 lung metastases (LM) from n=80 patients with 1648 pre-treatment computed tomography (CT) radiomics features and LM progression determined from follow-up CT, we developed a radiomic model to predict the progression of each lesion. Repeat experiments assessed the relative predictive performance across LM volume groups. Lesion-specific radiomic models indicate up to a 4.5-fold increase in predictive capacity compared with a no-skill classifier, with an area under the precision-recall curve of 0.70 for the most precise model (FDR = 0.05). Precision varied by administered drug and LM volume. The effect of LM volume was controlled by removing radiomic features at a volume-correlation coefficient threshold of 0.20. Predicting lesion-specific responses using radiomic features represents a novel strategy by which to assess treatment response that acknowledges biological diversity within metastatic subclones, which could facilitate management strategies involving selective ablation of resistant clones in the setting of systemic therapy.

Value of circulating tumor cell assisting low-dose computed tomography in screening pulmonary nodules based on existing liquid biopsy techniques: a systematic review with meta-analysis and trial sequential analysis

OBJECTIVE

This study aims to assess the diagnostic utility of circulating tumor cells (CTCs) in conjunction with low-dose computed tomography (LDCT) for differentiating between benign and malignant pulmonary nodules and to substantiate the foundation for their integration into clinical practice.

METHODS

A systematic literature review was performed independently by two researchers utilizing databases including PubMed, Web of Science, The Cochrane Library, Embase, and Medline, to collate studies up to September 15, 2023, that investigated the application of CTCs in diagnosing pulmonary nodules. A meta-analysis was executed employing Stata 15.0 and Revman 5.4 to calculate the pooled sensitivity, specificity, positive and negative likelihood ratios (PLR and NLR), diagnostic odds ratio (DOR), and the area under the receiver operating characteristic curve (AUC). Additionally, trial sequential analysis was conducted using dedicated TSA software.

RESULTS

The selection criteria identified 16 studies, encompassing a total of 3409 patients. The meta-analysis revealed that CTCs achieved a pooled sensitivity of 0.84 (95% CI 0.80 to 0.87), specificity of 0.80 (95% CI 0.73 to 0.86), PLR of 4.23 (95% CI 3.12 to 5.72), NLR of 0.20 (95% CI 0.16 to 0.25), DOR of 20.92 (95% CI 13.52 to 32.36), and AUC of 0.89 (95% CI 0.86 to 0.93).

CONCLUSIONS

Circulating tumor cells demonstrate substantial diagnostic accuracy in distinguishing benign from malignant pulmonary nodules. The incorporation of CTCs into the diagnostic protocol can significantly augment the diagnostic efficacy of LDCT in screening for malignant lung diseases.

Prognosis of unresected versus resected early-stage pulmonary carcinoid tumors ≤3 cm in size: A population-based study

PURPOSE

The observation-based prognosis, rather than resection, for small carcinoid tumors is still unclear. This lack of clarity has important implications for counseling elderly patients or patients for whom surgical resection poses a high risk. This study compared the outcomes of observation and surgical resection in patients with pulmonary carcinoid (PC) tumors ≤3 cm in size without metastasis.

METHODS

Data of patients with PC tumors with ≤3 cm in diameter and without lymph node and distant metastases were retrieved from Surveillance, Epidemiology, and End Results (SEER) registry. To reduce the inherent bias of retrospective studies, propensity score matching analysis was performed. Overall survival (OS) and lung carcinoid-specific survival (LCSS) were analyzed using Kaplan-Meier plots. Multivariate analysis was used to determine predictors of LCSS in different size subgroups.

RESULTS

In total, 4552 patients with early-stage PCs ≤3 cm in diameter, including 435 (9.56%) who were observed and 4117 (90.44%) treated by surgery, were recruited. Patients with surgery had significantly better OS and LCSS than those who were observed. However, patients with observation had comparable LCSS to those with surgery for PCs with tumor diameters ≤1 cm. Multivariate analysis indicated that surgical resection was an independent prognostic factor for LCSS in 1 cm < tumors ≤2 cm, and 2 cm < tumors ≤3 cm groups, but not for tumors ≤1 cm in diameter.

CONCLUSION

Surgical resection of small PCs is associated with a survival advantage over observation. However, for early PCs ≤1 cm in diameter, observation may be considered in patients with high risk for surgical resection.

Approach to Pulmonary Nodules in Connective Tissue Disease

The assessment of pulmonary nodules is a common and often challenging clinical scenario. This evaluation becomes even more complex in patients with connective tissue diseases (CTDs), as a range of disease-related factors must also be taken into account. These diseases are characterized by immune-mediated chronic inflammation, leading to tissue damage, collagen deposition, and subsequent organ dysfunction. A thorough examination of nodule features in these patients is required, incorporating anatomic and functional information, along with patient demographics, clinical factors, and disease-specific knowledge. This integrated approach is vital for effective risk stratification and precise diagnosis. This review article addresses specific CTD-related factors that should be taken into account when evaluating pulmonary nodules in this patient group.

Radiomics and Clinical Data for the Diagnosis of Incidental Pulmonary Nodules and Lung Cancer Screening: Radiolung Integrative Predictive Model

INTRODUCTION

Early diagnosis of lung cancer (LC) is crucial to improve survival rates. Radiomics models hold promise for enhancing LC diagnosis. This study assesses the impact of integrating a clinical and a radiomic model based on deep learning to predict the malignancy of pulmonary nodules (PN).

METHODOLOGY

Prospective cross-sectional study of 97 PNs from 93 patients. Clinical data included epidemiological risk factors and pulmonary function tests. The region of interest of each chest CT containing the PN was analysed. The radiomic model employed a pre-trained convolutional network to extract visual features. From these features, 500 with a positive standard deviation were chosen as inputs for an optimised neural network. The clinical model was estimated by a logistic regression model using clinical data. The malignancy probability from the clinical model was used as the best estimate of the pre-test probability of disease to update the malignancy probability of the radiomic model using a nomogram for Bayes' theorem.

RESULTS

The radiomic model had a positive predictive value (PPV) of 86%, an accuracy of 79% and an AUC of 0.67. The clinical model identified DLCO, obstruction index and smoking status as the most consistent clinical predictors associated with outcome. Integrating the clinical features into the deep-learning radiomic model achieves a PPV of 94%, an accuracy of 76% and an AUC of 0.80.

CONCLUSIONS

Incorporating clinical data into a deep-learning radiomic model improved PN malignancy assessment, boosting predictive performance. This study supports the potential of combined image-based and clinical features to improve LC diagnosis.

Multidisciplinary virtual management of pulmonary nodules

INTRODUCTION AND OBJECTIVES

Multidisciplinary nodule clinics provide high-quality care and favor adherence to guidelines. Virtual care has shown savings benefits along with patient satisfaction. Our aim is to describe the first year of operation of a multidisciplinary virtual lung nodule clinic, the population evaluated and issued decisions. Secondarily, among discharged patients, we aimed to analyze their follow-up prior to the existence of our consultation, evaluating its adherence to guidelines.

MATERIALS AND METHODS

Observational study including all patients evaluated at the Virtual Lung Nodule Clinic (VLNC) (March 2018- March 2019). Clinical and radiological data were recorded. Recommendations, based on 2017 Fleischner Society guidelines, were categorized into follow-up, discharge or referral to lung cancer consultation. Discharged patients were classified according to adherence to guidelines of their previous management, into adequate, prolonged and non-indicated follow-up.

RESULTS

A total of 365 patients (58.9% men; median age 64.0 years) were included. Sixty-four percent had smoking history and 23% had chronic obstructive pulmonary disease (COPD). Most nodules were solid (87.4%) and multiple (57.5%). The median diameter was 6.00 mm. 43.8% of patients were discharged following first VLNC evaluation. Among them, 27.5% had received appropriate follow-up, but 66.9% had received poor management. Patients with prolonged follow-up (33.1%) were older (67.0 vs 60.5 years) and had larger nodules (6.00 mm vs 5.00). Non-indicated follow-up patients (33.8%) were more non-smokers (77.8% vs 31.8%) and presented smaller nodules (4.00 vs 5.00 mm).

CONCLUSIONS

During its first year of operation, the VLNC has evaluated a population with a relevant risk profile for lung cancer development, management of which should be cautious and adhere to guidelines. After the first VLNC assessment, approximately one-half of this population was discharged. It was noticeable that previous follow-up of discharged patients was found poorly adherent to guidelines, with a marked tendency to overmanagement.

Path planning algorithm for percutaneous puncture lung mass biopsy procedure based on the multi-objective constraints and fuzzy optimization

Objective. The percutaneous puncture lung mass biopsy procedure, which relies on preoperative CT (Computed Tomography) images, is considered the gold standard for determining the benign or malignant nature of lung masses. However, the traditional lung puncture procedure has several issues, including long operation times, a high probability of complications, and high exposure to CT radiation for the patient, as it relies heavily on the surgeon's clinical experience.Approach.To address these problems, a multi-constrained objective optimization model based on clinical criteria for the percutaneous puncture lung mass biopsy procedure has been proposed. Additionally, based on fuzzy optimization, a multidimensional spatial Pareto front algorithm has been developed for optimal path selection. The algorithm finds optimal paths, which are displayed on 3D images, and provides reference points for clinicians' surgical path planning.Main results.To evaluate the algorithm's performance, 25 data sets collected from the Second People's Hospital of Zigong were used for prospective and retrospective experiments. The results demonstrate that 92% of the optimal paths generated by the algorithm meet the clinicians' surgical needs.Significance.The algorithm proposed in this paper is innovative in the selection of mass target point, the integration of constraints based on clinical standards, and the utilization of multi-objective optimization algorithm. Comparison experiments have validated the better performance of the proposed algorithm. From a clinical standpoint, the algorithm proposed in this paper has a higher clinical feasibility of the proposed pathway than related studies, which reduces the dependency of the physician's expertise and clinical experience on pathway planning during the percutaneous puncture lung mass biopsy procedure.

How will lung cancer screening and lung nodule management change the diagnostic and surgical lung cancer landscape?

INTRODUCTION

Implementation of lung cancer screening, with its subsequent findings, is anticipated to change the current diagnostic and surgical lung cancer landscape. This review aimed to identify and present the most updated expert opinion and discuss relevant evidence regarding the impact of lung cancer screening and lung nodule management on the diagnostic and surgical landscape of lung cancer, as well as summarise points for clinical practice.

METHODS

This article is based on relevant lectures and talks delivered during the European Society of Thoracic Surgeons-European Respiratory Society Collaborative Course on Thoracic Oncology (February 2023). Original lectures and talks and their relevant references were included. An additional literature search was conducted and peer-reviewed studies in English (December 2022 to June 2023) from the PubMed/Medline databases were evaluated with regards to immediate affinity of the published papers to the original talks presented at the course. An updated literature search was conducted (June 2023 to December 2023) to ensure that updated literature is included within this article.

RESULTS

Lung cancer screening suspicious findings are expected to increase the number of diagnostic investigations required therefore impacting on current capacity and resources. Healthcare systems already face a shortage of imaging and diagnostic slots and they are also challenged by the shortage of interventional radiologists. Thoracic surgery will be impacted by the wider lung cancer screening implementation with increased volume and earlier stages of lung cancer. Nonsuspicious findings reported at lung cancer screening will need attention and subsequent referrals where required to ensure participants are appropriately diagnosed and managed and that they are not lost within healthcare systems.

CONCLUSIONS

Implementation of lung cancer screening requires appropriate mapping of existing resources and infrastructure to ensure a tailored restructuring strategy to ensure that healthcare systems can meet the new needs.

Risk of Malignancy in Incidentally Detected Lung Nodules in Patients Aged Younger Than 35 Years

BACKGROUND

The risk of malignancy in pulmonary nodules incidentally detected on computed tomography (CT) in patients who are aged younger than 35 years is unclear.

OBJECTIVE

The aim of this study was to evaluate the incidence of lung cancer in incidental pulmonary nodules in patients who are 15-34 years old.

METHODS

This retrospective study included patients aged 15-34 years who had an incidental pulmonary nodule on chest CT from 2010 to 2018 at our hospital. Patients with prior, current, or suspected malignancy were excluded. A chart review identified patients with diagnosis of malignancy. Incidental pulmonary nodule was deemed benign if stable or resolved on a follow-up CT at least 2 years after initial or if there was a medical visit in our health care network at least 2 years after initial CT without diagnosis of malignancy.Receiver operating characteristic curve analysis was performed with nodule size. Association of categorical variables with lung cancer diagnosis was performed with Fisher exact test, and association of continuous variables was performed with logistic regression.

RESULTS

Five thousand three hundred fifty-five chest CTs performed on patients aged 15-34 years between January 2010 and December 2018. After excluding patients without a reported pulmonary nodule and prior or current malignancy, there were a total of 779 patients. Of these, 690 (89%) had clinical or imaging follow-up after initial imaging. Of these, 545 (70% of total patients) patients had imaging or clinical follow-up greater than 2 years after their initial imaging.A malignant diagnosis was established in 2/779 patients (0.3%; 95% confidence interval, 0.1%-0.9%). Nodule size was strongly associated with malignancy (P = 0.007), with area under the receiver operating characteristic curve of 0.97. There were no malignant nodules that were less than 10 mm in size. Smoking history, number of nodules, and nodule density were not associated with malignancy.

CONCLUSIONS

Risk of malignancy for incidentally detected pulmonary nodules in patients aged 15-34 years is extremely small (0.3%). There were no malignant nodules that were less than 10 mm in size. Routine follow-up of subcentimeter pulmonary nodules should be carefully weighed against the risks.

Is 18F-fluorodeoxyglucose PET recommended for small lung nodules? CT findings of 18F-fluorodeoxyglucose non-avid lung cancer

OBJECTIVES

To determine the image characteristics associated with low 18F-FDG (18F-fluorodeoxyglucose) avidity among 8-15 mm solid lung cancer.

METHODS

Patients satisfying the following criteria were included: underwent surgery between January 2014 and December 2019 for lung cancer, presented 8-15 mm nodule without measurable ground glass component on preoperative CT, and underwent 18F-FDG PET before resection. Image characteristics, including air bronchogram, concave shape, pleural attachment, and background emphysema, were evaluated by two board-certified radiologists. The Mann-Whitney U test was used to compare maximum standardized uptake (SUVmax) values from 18F-FDG PET images.

RESULTS

The analysis included 235 patients. The SUVmax values of lesions with air bronchogram and concave shape were significantly lower than the SUVmax values of lesions without these features (median: 1.55 vs 2.56 and 1.66 vs 2.45, both P < .001), whereas lesions arising from emphysematous lungs had significantly higher SUVmax values than lesions arising from non-emphysematous lungs (2.90 vs 1.69, P < .001). No significant differences were detected between lesions attached and not attached to pleura. The interobserver agreement was almost perfect for air bronchograms and background emphysema (κ = 0.882 and 0.927, respectively), and 89.7% of lesions with air bronchograms and arising from non-emphysematous lungs showed SUVmax values below 2.5.

CONCLUSIONS

Among 8-15 mm solid lung cancer, the presence of air bronchograms and concave shape and the absence of background emphysema were associated with low 18F-FDG accumulation.

ADVANCES IN KNOWLEDGE

18F-FDG PET can be misleading in differentiating certain type of small solid lung cancer.

The differential computed tomography features between small benign and malignant solid solitary pulmonary nodules with different sizes

Background

Computed tomography (CT) has been widely known to be the first choice for the diagnosis of solid solitary pulmonary nodules (SSPNs). However, the smaller the SSPN is, the less the differential CT signs between benign and malignant SSPNs there are, which brings great challenges to their diagnosis. Therefore, this study aimed to investigate the differential CT features between small (≤15 mm) benign and malignant SSPNs with different sizes.

Methods

From May 2018 to November 2021, CT data of 794 patients with small SSPNs (≤15 mm) were retrospectively analyzed. SSPNs were divided into benign and malignant groups, and each group was further classified into three cohorts: cohort I (diameter ≤6 mm), cohort II (6 mm < diameter ≤8 mm), and cohort III (8 mm < diameter ≤15 mm). The differential CT features of benign and malignant SSPNs in three cohorts were identified. Multivariable logistic regression analyses were conducted to identify independent factors of benign SSPNs.

Results

In cohort I, polygonal shape and upper-lobe distribution differed significantly between groups (all P<0.05) and multiparametric analysis showed polygonal shape [adjusted odds ratio (OR): 12.165; 95% confidence interval (CI): 1.512-97.872; P=0.019] was the most effective variation for predicting benign SSPNs, with an area under the receiver operating characteristic curve (AUC) of 0.747 (95% CI: 0.640-0.855; P=0.001). In cohort II, polygonal shape, lobulation, pleural retraction, and air bronchogram differed significantly between groups (all P<0.05), and polygonal shape (OR: 8.870; 95% CI: 1.096-71.772; P=0.041) and the absence of pleural retraction (OR: 0.306; 95% CI: 0.106-0.883; P=0.028) were independent predictors of benign SSPNs, with an AUC of 0.778 (95% CI: 0.694-0.863; P<0.001). In cohort III, 12 CT features showed significant differences between groups (all P<0.05) and polygonal shape (OR: 3.953; 95% CI: 1.508-10.361; P=0.005); calcification (OR: 3.710; 95% CI: 1.305-10.551; P=0.014); halo sign (OR: 6.237; 95% CI: 2.838-13.710; P<0.001); satellite lesions (OR: 6.554; 95% CI: 3.225-13.318; P<0.001); and the absence of lobulation (OR: 0.066; 95% CI: 0.026-0.167; P<0.001), air space (OR: 0.405; 95% CI: 0.215-0.764; P=0.005), pleural retraction (OR: 0.297; 95% CI: 0.179-0.493; P<0.001), bronchial truncation (OR: 0.165; 95% CI: 0.090-0.303; P<0.001), and air bronchogram (OR: 0.363; 95% CI: 0.208-0.633; P<0.001) were independent predictors of benign SSPNs, with an AUC of 0.869 (95% CI: 0.840-0.897; P<0.001).

Conclusions

CT features vary between SSPNs with different sizes. Clarifying the differential CT features based on different diameter ranges may help to minimize ambiguities and discriminate the benign SSPNs from malignant ones.

[4D electromagnetic navigation bronchoscopy for the diagnosis of peripheral pulmonary nodules - An overview and preliminary clinical results]

BACKGROUND

The diagnostic of peripheral pulmonary nodules (PPN) is a particular challenge in interventional bronchology, which is why navigation systems such as electromagnetic navigation (ENB) are increasingly being used. The 4D-ENB represents the most current development of the ENB. It utilizes inspiratory and expiratory CT scans for mapping and thus helps compensate for respiratory movements-induced CT-to-body divergence. The aim of this work was to present the first clinical data and experiences using the 4D-ENB method for diagnosis of PPNs.

METHODS

We retrospectively describe the results of the first nine consecutive patient cases diagnosed at Klinikum Braunschweig using 4D-ENB in a unimodal diagnostic procedure.

RESULTS

Of the first 9 PPNs examined by 4D-ENB, navigation and puncture of the lesion was successful in 8 patients (89%). Diagnostic biopsy was could be carried out in six out of nine patients (67%). There were no significant procedure-related complications.

CONCLUSION

Our preliminary data suggest that 4D-ENB is a promising new alternative for the diagnosis of PPNs. To further improve diagnostic yield, 4D-END, which lacks real-time visualization, should be embedded in a multimodal diagnostic procedure with rEBUS and/or fluoroscopy.

Explainable machine learning model based on clinical factors for predicting the disappearance of indeterminate pulmonary nodules

BACKGROUND

During lung cancer screening, indeterminate pulmonary nodules (IPNs) are a frequent finding. We aim to predict whether IPNs are resolving or non-resolving to reduce follow-up examinations, using machine learning (ML) models. We incorporated dedicated techniques to enhance prediction explainability.

METHODS

In total, 724 IPNs (size 50-500 mm3, 575 participants) from the Dutch-Belgian Randomized Lung Cancer Screening Trial were used. We implemented six ML models and 14 factors to predict nodule disappearance. Random search was applied to determine the optimal hyperparameters on the training set (579 nodules). ML models were trained using 5-fold cross-validation and tested on the test set (145 nodules). Model predictions were evaluated by utilizing the recall, precision, F1 score, and the area under the receiver operating characteristic curve (AUC). The best-performing model was used for three feature importance techniques: mean decrease in impurity (MDI), permutation feature importance (PFI), and SHAPley Additive exPlanations (SHAP).

RESULTS

The random forest model outperformed the other ML models with an AUC of 0.865. This model achieved a recall of 0.646, a precision of 0.816, and an F1 score of 0.721. The evaluation of feature importance achieved consistent ranking across all three methods for the most crucial factors. The MDI, PFI, and SHAP methods highlighted volume, maximum diameter, and minimum diameter as the top three factors. However, the remaining factors revealed discrepant ranking across methods.

CONCLUSION

ML models effectively predict IPN disappearance using participant demographics and nodule characteristics. Explainable techniques can assist clinicians in developing understandable preliminary assessments.

Fully automatic deep learning-based lung parenchyma segmentation and boundary correction in thoracic CT scans

PURPOSE

The proposed work aims to develop an algorithm to precisely segment the lung parenchyma in thoracic CT scans. To achieve this goal, the proposed technique utilized a combination of deep learning and traditional image processing algorithms. The initial step utilized a trained convolutional neural network (CNN) to generate preliminary lung masks, followed by the proposed post-processing algorithm for lung boundary correction.

METHODS

First, the proposed method trained an improved 2D U-Net CNN model with Inception-ResNet-v2 as its backbone. The model was trained on 32 CT scans from two different sources: one from the VESSEL12 grand challenge and the other from AIIMS Delhi. Further, the model's performance was evaluated on a test dataset of 16 CT scans with juxta-pleural nodules obtained from AIIMS Delhi and the LUNA16 challenge. The model's performance was assessed using evaluation metrics such as average volumetric dice coefficient (DSCavg), average IoU score (IoUavg), and average F1 score (F1avg). Finally, the proposed post-processing algorithm was implemented to eliminate false positives from the model's prediction and to include juxta-pleural nodules in the final lung masks.

RESULTS

The trained model reported a DSCavg of 0.9791 ± 0.008, IoUavg of 0.9624 ± 0.007, and F1avg of 0.9792 ± 0.004 on the test dataset. Applying the post-processing algorithm to the predicted lung masks obtained a DSCavg of 0.9713 ± 0.007, IoUavg of 0.9486 ± 0.007, and F1avg of 0.9701 ± 0.008. The post-processing algorithm successfully included juxta-pleural nodules in the final lung mask.

CONCLUSIONS

Using a CNN model, the proposed method for lung parenchyma segmentation produced precise segmentation results. Furthermore, the post-processing algorithm addressed false positives and negatives in the model's predictions. Overall, the proposed approach demonstrated promising results for lung parenchyma segmentation. The method has the potential to be valuable in the advancement of computer-aided diagnosis (CAD) systems for automatic nodule detection.

Deep learning techniques in PET/CT imaging: A comprehensive review from sinogram to image space

Positron emission tomography/computed tomography (PET/CT) is increasingly used in oncology, neurology, cardiology, and emerging medical fields. The success stems from the cohesive information that hybrid PET/CT imaging offers, surpassing the capabilities of individual modalities when used in isolation for different malignancies. However, manual image interpretation requires extensive disease-specific knowledge, and it is a time-consuming aspect of physicians' daily routines. Deep learning algorithms, akin to a practitioner during training, extract knowledge from images to facilitate the diagnosis process by detecting symptoms and enhancing images. This acquired knowledge aids in supporting the diagnosis process through symptom detection and image enhancement. The available review papers on PET/CT imaging have a drawback as they either included additional modalities or examined various types of AI applications. However, there has been a lack of comprehensive investigation specifically focused on the highly specific use of AI, and deep learning, on PET/CT images. This review aims to fill that gap by investigating the characteristics of approaches used in papers that employed deep learning for PET/CT imaging. Within the review, we identified 99 studies published between 2017 and 2022 that applied deep learning to PET/CT images. We also identified the best pre-processing algorithms and the most effective deep learning models reported for PET/CT while highlighting the current limitations. Our review underscores the potential of deep learning (DL) in PET/CT imaging, with successful applications in lesion detection, tumor segmentation, and disease classification in both sinogram and image spaces. Common and specific pre-processing techniques are also discussed. DL algorithms excel at extracting meaningful features, and enhancing accuracy and efficiency in diagnosis. However, limitations arise from the scarcity of annotated datasets and challenges in explainability and uncertainty. Recent DL models, such as attention-based models, generative models, multi-modal models, graph convolutional networks, and transformers, are promising for improving PET/CT studies. Additionally, radiomics has garnered attention for tumor classification and predicting patient outcomes. Ongoing research is crucial to explore new applications and improve the accuracy of DL models in this rapidly evolving field.

Radiomics-based decision support tool assists radiologists in small lung nodule classification and improves lung cancer early diagnosis

BACKGROUND

Methods to improve stratification of small (≤15 mm) lung nodules are needed. We aimed to develop a radiomics model to assist lung cancer diagnosis.

METHODS

Patients were retrospectively identified using health records from January 2007 to December 2018. The external test set was obtained from the national LIBRA study and a prospective Lung Cancer Screening programme. Radiomics features were extracted from multi-region CT segmentations using TexLab2.0. LASSO regression generated the 5-feature small nodule radiomics-predictive-vector (SN-RPV). K-means clustering was used to split patients into risk groups according to SN-RPV. Model performance was compared to 6 thoracic radiologists. SN-RPV and radiologist risk groups were combined to generate "Safety-Net" and "Early Diagnosis" decision-support tools.

RESULTS

In total, 810 patients with 990 nodules were included. The AUC for malignancy prediction was 0.85 (95% CI: 0.82-0.87), 0.78 (95% CI: 0.70-0.85) and 0.78 (95% CI: 0.59-0.92) for the training, test and external test datasets, respectively. The test set accuracy was 73% (95% CI: 65-81%) and resulted in 66.67% improvements in potentially missed [8/12] or delayed [6/9] cancers, compared to the radiologist with performance closest to the mean of six readers.

CONCLUSIONS

SN-RPV may provide net-benefit in terms of earlier cancer diagnosis.

MANet: Multi-branch attention auxiliary learning for lung nodule detection and segmentation

BACKGROUND AND OBJECTIVE

Pulmonary nodule detection and segmentation are currently two primary tasks in analyzing chest computed tomography (Chest CT) in order to detect signs of lung cancer, thereby providing early treatment measures to reduce mortality. Even though there are many proposed methods to reduce false positives for obtaining effective detection results, distinguishing between the pulmonary nodule and background region remains challenging because their biological characteristics are similar and varied in size. The purpose of our work is to propose a method for automatic nodule detection and segmentation in Chest CT by enhancing the feature information of pulmonary nodules.

METHODS

We propose a new UNet-based backbone with multi-branch attention auxiliary learning mechanism, which contains three novel modules, namely, Projection module, Fast Cascading Context module, and Boundary Enhancement module, to further enhance the nodule feature representation. Based on that, we build MANet, a lung nodule localization network that simultaneously detects and segments precise nodule positions. Furthermore, our MANet contains a Proposal Refinement step which refines initially generated proposals to effectively reduce false positives and thereby produce the segmentation quality.

RESULTS

Comprehensive experiments on the combination of two benchmarks LUNA16 and LIDC-IDRI show that our proposed model outperforms state-of-the-art methods in the tasks of nodule detection and segmentation tasks in terms of FROC, IoU, and DSC metrics. Our method reports an average FROC score of 88.11% in lung nodule detection. For the lung nodule segmentation, the results reach an average IoU score of 71.29% and a DSC score of 82.74%. The ablation study also shows the effectiveness of the new modules which can be integrated into other UNet-based models.

CONCLUSIONS

The experiments demonstrated our method with multi-branch attention auxiliary learning ability are a promising approach for detecting and segmenting the pulmonary nodule instances compared to the original UNet design.

Radiomic-Based Prediction of Lesion-Specific Systemic Treatment Response in Metastatic Disease

Despite sharing the same histologic classification, individual tumors in multi metastatic patients may present with different characteristics and varying sensitivities to anticancer therapies. In this study, we investigate the utility of radiomic biomarkers for prediction of lesion-specific treatment resistance in multi metastatic leiomyosarcoma patients. Using a dataset of n=202 lung metastases (LM) from n=80 patients with 1648 pre-treatment computed tomography (CT) radiomics features and LM progression determined from follow-up CT, we developed a radiomic model to predict the progression of each lesion. Repeat experiments assessed the relative predictive performance across LM volume groups. Lesion-specific radiomic models indicate up to a 5-fold increase in predictive capacity compared with a no-skill classifier, with an area under the precision-recall curve of 0.79 for the most precise model (FDR = 0.01). Precision varied by administered drug and LM volume. The effect of LM volume was controlled by removing radiomic features at a volume-correlation coefficient threshold of 0.20. Predicting lesion-specific responses using radiomic features represents a novel strategy by which to assess treatment response that acknowledges biological diversity within metastatic subclones, which could facilitate management strategies involving selective ablation of resistant clones in the setting of systemic therapy.

AI-based computer-aided diagnostic system of chest digital tomography synthesis: Demonstrating comparative advantage with X-ray-based AI systems

BACKGROUND

Compared with chest X-ray (CXR) imaging, which is a single image projected from the front of the patient, chest digital tomosynthesis (CDTS) imaging can be more advantageous for lung lesion detection because it acquires multiple images projected from multiple angles of the patient. Various clinical comparative analysis and verification studies have been reported to demonstrate this, but there is no artificial intelligence (AI)-based comparative analysis studies. Existing AI-based computer-aided detection (CAD) systems for lung lesion diagnosis have been developed mainly based on CXR images; however, CAD-based on CDTS, which uses multi-angle images of patients in various directions, has not been proposed and verified for its usefulness compared to CXR-based counterparts.

BACKGROUND AND OBJECTIVE

This study develops and tests a CDTS-based AI CAD system to detect lung lesions to demonstrate performance improvements compared to CXR-based AI CAD.

METHODS

We used multiple (e.g., five) projection images as input for the CDTS-based AI model and a single-projection image as input for the CXR-based AI model to compare and evaluate the performance between models. Multiple/single projection input images were obtained by virtual projection on the three-dimensional (3D) stack of computed tomography (CT) slices of each patient's lungs from which the bed area was removed. These multiple images result from shooting from the front and left and right 30/60∘. The projected image captured from the front was used as the input for the CXR-based AI model. The CDTS-based AI model used all five projected images. The proposed CDTS-based AI model consisted of five AI models that received images in each of the five directions, and obtained the final prediction result through an ensemble of five models. Each model used WideResNet-50. To train and evaluate CXR- and CDTS-based AI models, 500 healthy data, 206 tuberculosis data, and 242 pneumonia data were used, and three three-fold cross-validation was applied.

RESULTS

The proposed CDTS-based AI CAD system yielded sensitivities of 0.782 and 0.785 and accuracies of 0.895 and 0.837 for the (binary classification) performance of detecting tuberculosis and pneumonia, respectively, against normal subjects. These results show higher performance than the sensitivity of 0.728 and 0.698 and accuracies of 0.874 and 0.826 for detecting tuberculosis and pneumonia through the CXR-based AI CAD, which only uses a single projection image in the frontal direction. We found that CDTS-based AI CAD improved the sensitivity of tuberculosis and pneumonia by 5.4% and 8.7% respectively, compared to CXR-based AI CAD without loss of accuracy.

CONCLUSIONS

This study comparatively proves that CDTS-based AI CAD technology can improve performance more than CXR. These results suggest that we can enhance the clinical application of CDTS. Our code is available at https://github.com/kskim-phd/CDTS-CAD-P.

[Patient-centered, value-based management of incidental findings in radiology]

As a byproduct of the increased use of high-resolution radiological imaging, the prevalence of incidental findings (IFs) has been increasing for years. The discovery of an incidental finding can allow early treatment of a potentially health-threatening disease and thus decisively change the course of the disease. However, many incidental findings are of low risk with little or no health impact, and yet their discovery often leads to a cascade of additional investigations. It is undisputed that incidental findings can have a direct impact on the life of the person and that not only psychosocial aspects such as worries and anxiety due to false-positive findings play a role, but that insurance, legal or professional problems can also occur under certain circumstances, which is why the correct handling of incidental findings and the accompanying ethical challenges that apply to them regularly give rise to discussions. General principles to consider when managing incidental findings are responsibility for the well-being of the patient/study participant and of society. In order to avoid overdiagnosis and overtreatment and to achieve high patient benefit, radiologists and clinicians must know how to properly deal with IFs. In recent years, various national and international societies have published important guidelines ("white papers") on how to deal with the management of IFs. It is important that radiologists are fully aware of and follow these guidelines and are also available to referring physicians for further discussions and advice. The most important fact is that the well-being of the patient must always be at the center of all decisions.

A proposed methodology for detecting the malignant potential of pulmonary nodules in sarcoma using computed tomographic imaging and artificial intelligence-based models

The presence of lung metastases in patients with primary malignancies is an important criterion for treatment management and prognostication. Computed tomography (CT) of the chest is the preferred method to detect lung metastasis. However, CT has limited efficacy in differentiating metastatic nodules from benign nodules (e.g., granulomas due to tuberculosis) especially at early stages (<5 mm). There is also a significant subjectivity associated in making this distinction, leading to frequent CT follow-ups and additional radiation exposure along with financial and emotional burden to the patients and family. Even 18F-fluoro-deoxyglucose positron emission technology-computed tomography (18F-FDG PET-CT) is not always confirmatory for this clinical problem. While pathological biopsy is the gold standard to demonstrate malignancy, invasive sampling of small lung nodules is often not clinically feasible. Currently, there is no non-invasive imaging technique that can reliably characterize lung metastases. The lung is one of the favored sites of metastasis in sarcomas. Hence, patients with sarcomas, especially from tuberculosis prevalent developing countries, can provide an ideal platform to develop a model to differentiate lung metastases from benign nodules. To overcome the lack of optimal specificity of CT scan in detecting pulmonary metastasis, a novel artificial intelligence (AI)-based protocol is proposed utilizing a combination of radiological and clinical biomarkers to identify lung nodules and characterize it as benign or metastasis. This protocol includes a retrospective cohort of nearly 2,000-2,250 sample nodules (from at least 450 patients) for training and testing and an ambispective cohort of nearly 500 nodules (from 100 patients; 50 patients each from the retrospective and prospective cohort) for validation. Ground-truth annotation of lung nodules will be performed using an in-house-built segmentation tool. Ground-truth labeling of lung nodules (metastatic/benign) will be performed based on histopathological results or baseline and/or follow-up radiological findings along with clinical outcome of the patient. Optimal methods for data handling and statistical analysis are included to develop a robust protocol for early detection and classification of pulmonary metastasis at baseline and at follow-up and identification of associated potential clinical and radiological markers.

A graph neural network model for the diagnosis of lung adenocarcinoma based on multimodal features and an edge-generation network

Background

Lung cancer is a global disease with high lethality, with early screening being considerably helpful for improving the 5-year survival rate. Multimodality features in early screening imaging are an important part of the prediction for lung adenocarcinoma, and establishing a model for adenocarcinoma diagnosis based on multimodal features is an obvious clinical need. Through our practice and investigation, we found that graph neural networks (GNNs) are excellent platforms for multimodal feature fusion, and the data can be completed using the edge-generation network. Therefore, we propose a new lung adenocarcinoma multiclassification model based on multimodal features and an edge-generation network.

Methods

According to a ratio of 80% to 20%, respectively, the dataset of 338 cases was divided into the training set and the test set through 5-fold cross-validation, and the distribution of the 2 sets was the same. First, the regions of interest (ROIs) cropped from computed tomography (CT) images were separately fed into convolutional neural networks (CNNs) and radiomics processing platforms. The results of the 2 parts were then input into a graph embedding representation network to obtain the fused feature vectors. Subsequently, a graph database based on the clinical and semantic features was established, and the data were supplemented by an edge-generation network, with the fused feature vectors being used as the input of the nodes. This enabled us to clearly understand where the information transmission of the GNN takes place and improves the interpretability of the model. Finally, the nodes were classified using GNNs.

Results

On our dataset, the proposed method presented in this paper achieved superior results compared to traditional methods and showed some comparability with state-of-the-art methods for lung nodule classification. The results of our method are as follows: accuracy (ACC) =66.26% (±4.46%), area under the curve (AUC) =75.86% (±1.79%), F1-score =64.00% (±3.65%), and Matthews correlation coefficient (MCC) =48.40% (±5.07%). The model with the edge-generating network consistently outperformed the model without it in all aspects.

Conclusions

The experiments demonstrate that with appropriate data=construction methods GNNs can outperform traditional image processing methods in the field of CT-based medical image classification. Additionally, our model has higher interpretability, as it employs subjective clinical and semantic features as the data construction approach. This will help doctors better leverage human-computer interactions.

Diagnostic performance of artificial intelligence for pediatric pulmonary nodule detection in computed tomography of the chest

PURPOSE

To test the performance of a commercially available adult pulmonary nodule detection artificial intelligence (AI) tool in pediatric CT chests.

METHODS

30 consecutive chest CTs with or without contrast of patients ages 12-18 were included. Images were retrospectively reconstructed at 3 mm and 1 mm slice thickness. AI for detection of lung nodules in adults (Syngo CT Lung Computer Aided Detection (CAD)) was evaluated. 3 mm axial images were retrospectively reviewed by two pediatric radiologists (reference read) who determined the location, type, and size of nodules. Lung CAD results at 3 mm and 1 mm slice thickness were compared to reference read by two other pediatric radiologists. Sensitivity (Sn) and positive predictive value (PPV) were analyzed.

RESULTS

The radiologists identified 109 nodules. At 1 mm, CAD detected 70 nodules; 43 true positive (Sn = 39 %), 26 false positive (PPV = 62 %), and 1 nodule which had not been identified by radiologists. At 3 mm, CAD detected 60 nodules; 28 true positive (Sn = 26 %), 30 false positive (PPV = 48 %) and 2 nodules which had not been identified by radiologists. There were 103 solid nodules (47 measuring < 3 mm) and 6 subsolid nodules (5 measuring < 5 mm). When excluding 52 nodules (solid < 3 mm and subsolid < 5 mm) based on algorithm conditions, the Sn increased to 68 % at 1 mm and 49 % at 3 mm but there was no significant change in the PPV measuring 60 % at 1 mm and 48 % at 3 mm.

CONCLUSION

The adult Lung CAD showed low sensitivity in pediatric patients, but better performance at thinner slice thickness and when smaller nodules were excluded.

Metabolomic differentiation of benign vs malignant pulmonary nodules with high specificity via high-resolution mass spectrometry analysis of patient sera

Differential diagnosis of pulmonary nodules detected by computed tomography (CT) remains a challenge in clinical practice. Here, we characterize the global metabolomes of 480 serum samples including healthy controls, benign pulmonary nodules, and stage I lung adenocarcinoma. The adenocarcinoma demonstrates a distinct metabolomic signature, whereas benign nodules and healthy controls share major similarities in metabolomic profiles. A panel of 27 metabolites is identified in the discovery cohort (n = 306) to distinguish between benign and malignant nodules. The discriminant model achieves an AUC of 0.915 and 0.945 in the internal validation (n = 104) and external validation cohort (n = 111), respectively. Pathway analysis reveals elevation in glycolytic metabolites associated with decreased tryptophan in serum of lung adenocarcinoma vs benign nodules and healthy controls, and demonstrates that uptake of tryptophan promotes glycolysis in lung cancer cells. Our study highlights the value of the serum metabolite biomarkers in risk assessment of pulmonary nodules detected by CT screening.

Size measurement of lung nodules on CT: which diameter is most stable to inter-observer variability?

Indeterminate lung nodules detected on CT are common findings in the clinical practice, and the correct assessment of their size is critical for patient evaluation and management. We compared the stability of three definitions of nodule diameter (Feret's mean diameter, Martin's mean diameter and area-equivalent diameter) to inter-observer variability on a population of 336 solid nodules from 207 subjects. We found that inter-observer agreement was highest with Martin's mean diameter (intra-class correlation coefficient = 0.977, 95% Confidence interval = 0.977-0.978), followed by area-equivalent diameter (0.972, 0.971-0.973) and Feret's mean diameter (0.965, 0.964-0.966). The differences were statistically significant. In conclusion, although all the three diameter definitions achieved very good inter-observer agreement (ICC > 0.96), Martin's mean diameter was significantly better than the others. Future guidelines may consider adopting Martin's mean diameter as an alternative to the currently used Feret's (caliper) diameter for assessing the size of lung nodules on CT.

Epidemiology, management and outcomes of Cryptococcus gattii infections: A 22-year cohort

BACKGROUND

Cryptococcus gattii is a globally endemic pathogen causing disease in apparently immune-competent hosts. We describe a 22-year cohort study from Australia's Northern Territory to evaluate trends in epidemiology and management, and outcome predictors.

METHODS

A retrospective cohort study of all C. gattii infections at the northern Australian referral hospital 1996-2018 was conducted. Cases were defined as confirmed (culture-positive) or probable. Demographic, clinical and outcome data were extracted from medical records.

RESULTS

45 individuals with C. gattii infection were included: 44 Aboriginal Australians; 35 with confirmed infection; none HIV positive out of 38 tested. Multifocal disease (pulmonary and central nervous system) occurred in 20/45 (44%). Nine people (20%) died within 12 months of diagnosis, five attributed directly to C. gattii. Significant residual disability was evident in 4/36 (11%) survivors. Predictors of mortality included: treatment before the year 2002 (4/11 versus 1/34); interruption to induction therapy (2/8 versus 3/37) and end-stage kidney disease (2/5 versus 3/40). Prolonged antifungal therapy was the standard approach in this cohort, with median treatment duration being 425 days (IQR 166-715). Ten individuals had adjunctive lung resection surgery for large pulmonary cryptococcomas (median diameter 6cm [range 2.2-10cm], versus 2.8cm [1.2-9cm] in those managed non-operatively). One died post-operatively, and 7 had thoracic surgical complications, but ultimately 9/10 (90%) treated surgically were cured compared with 10/15 (67%) who did not have lung surgery. Four patients were diagnosed with immune reconstitution inflammatory syndrome which was associated with age <40 years, brain cryptococcomas, high cerebrospinal fluid pressure, and serum cryptococcal antigen titre >1:512.

CONCLUSION

C. gattii infection remains a challenging condition but treatment outcomes have significantly improved over 2 decades, with eradication of infection the norm. Adjunctive surgery for the management of bulky pulmonary C. gattii infection appears to increase the likelihood of durable cure and likely reduces the required duration of antifungal therapy.

Contextualizing the Role of Volumetric Analysis in Pulmonary Nodule Assessment: AJR Expert Panel Narrative Review

Pulmonary nodules are managed on the basis of their size and morphologic characteristics. Radiologists are familiar with assessing nodule size by measuring diameter using manually deployed electronic calipers. Size may also be assessed with 3D volumetric measurements (referred to as volumetry) obtained with software. Nodule size and growth are more accurately assessed with volumetry than on the basis of diameter, and the evidence supporting clinical use of volumetry has expanded, driven by its use in lung cancer screening nodule management algorithms in Europe. The application of volumetry has the potential to reduce recommendations for imaging follow-up of indeterminate solid nodules without impacting cancer detection. Although changes in scanning conditions and volumetry software packages can lead to variation in volumetry results, ongoing technical advances have improved the reliability of calculated volumes. Volumetry is now the primary method for determining size of solid nodules in the European lung cancer screening position statement and British Thoracic Society recommendations. The purposes of this article are to review technical aspects, advantages, and limitations of volumetry and, by considering specific scenarios, to contextualize the use of volumetry with respect to its importance in morphologic evaluation, its role in predicting malignancy in risk models, and its practical impact on nodule management. Implementation challenges and areas requiring further evidence are also highlighted.

Detection of solid and subsolid pulmonary nodules with lung MRI: performance of UTE, T1 gradient-echo, and single-shot T2 fast spin echo

BACKGROUND

Although MRI is a radiation-free imaging modality, it has historically been limited in lung imaging due to inherent technical restrictions. The aim of this study is to explore the performance of lung MRI in detecting solid and subsolid pulmonary nodules using T1 gradient-echo (GRE) (VIBE, Volumetric interpolated breath-hold examination), ultrashort time echo (UTE) and T2 Fast Spin Echo (HASTE, Half fourier Single-shot Turbo spin-Echo).

METHODS

Patients underwent a lung MRI in a 3 T scanner as part of a prospective research project. A baseline Chest CT was obtained as part of their standard of care. Nodules were identified and measured on the baseline CT and categorized according to their density (solid and subsolid) and size (> 4 mm/ ≤ 4 mm). Nodules seen on the baseline CT were classified as present or absent on the different MRI sequences by two thoracic radiologists independently. Interobserver agreement was determined using the simple Kappa coefficient. Paired differences were compared using nonparametric Mann-Whitney U tests. The McNemar test was used to evaluate paired differences in nodule detection between MRI sequences.

RESULTS

Thirty-six patients were prospectively enrolled. One hundred forty-nine nodules (100 solid/49 subsolid) with mean size 10.8 mm (SD = 9.4) were included in the analysis. There was substantial interobserver agreement (k = 0.7, p = 0.05). Detection for all nodules, solid and subsolid nodules was respectively; UTE: 71.8%/71.0%/73.5%; VIBE: 61.6%/65%/55.1%; HASTE 72.4%/72.2%/72.7%. Detection rate was higher for nodules > 4 mm in all groups: UTE 90.2%/93.4%/85.4%, VIBE 78.4%/88.5%/63.4%, HASTE 89.4%/93.8%/83.8%. Detection of lesions ≤4 mm was low for all sequences. UTE and HASTE performed significantly better than VIBE for detection of all nodules and subsolid nodules (diff = 18.4 and 17.6%, p = < 0.01 and p = 0.03, respectively). There was no significant difference between UTE and HASTE. There were no significant differences amongst MRI sequences for solid nodules.

CONCLUSIONS

Lung MRI shows adequate performance for the detection of solid and subsolid pulmonary nodules larger than 4 mm and can serve as a promising radiation-free alternative to CT.

A radiomics-based decision support tool improves lung cancer diagnosis in combination with the Herder score in large lung nodules

BACKGROUND

Large lung nodules (≥15 mm) have the highest risk of malignancy, and may exhibit important differences in phenotypic or clinical characteristics to their smaller counterparts. Existing risk models do not stratify large nodules well. We aimed to develop and validate an integrated segmentation and classification pipeline, incorporating deep-learning and traditional radiomics, to classify large lung nodules according to cancer risk.

METHODS

502 patients from five U.K. centres were recruited to the large-nodule arm of the retrospective LIBRA study between July 2020 and April 2022. 838 CT scans were used for model development, split into training and test sets (70% and 30% respectively). An nnUNet model was trained to automate lung nodule segmentation. A radiomics signature was developed to classify nodules according to malignancy risk. Performance of the radiomics model, termed the large-nodule radiomics predictive vector (LN-RPV), was compared to three radiologists and the Brock and Herder scores.

FINDINGS

499 patients had technically evaluable scans (mean age 69 ± 11, 257 men, 242 women). In the test set of 252 scans, the nnUNet achieved a DICE score of 0.86, and the LN-RPV achieved an AUC of 0.83 (95% CI 0.77-0.88) for malignancy classification. Performance was higher than the median radiologist (AUC 0.75 [95% CI 0.70-0.81], DeLong p = 0.03). LN-RPV was robust to auto-segmentation (ICC 0.94). For baseline solid nodules in the test set (117 patients), LN-RPV had an AUC of 0.87 (95% CI 0.80-0.93) compared to 0.67 (95% CI 0.55-0.76, DeLong p = 0.002) for the Brock score and 0.83 (95% CI 0.75-0.90, DeLong p = 0.4) for the Herder score. In the international external test set (n = 151), LN-RPV maintained an AUC of 0.75 (95% CI 0.63-0.85). 18 out of 22 (82%) malignant nodules in the Herder 10-70% category in the test set were identified as high risk by the decision-support tool, and may have been referred for earlier intervention.

INTERPRETATION

The model accurately segments and classifies large lung nodules, and may improve upon existing clinical models.

FUNDING

This project represents independent research funded by: 1) Royal Marsden Partners Cancer Alliance, 2) the Royal Marsden Cancer Charity, 3) the National Institute for Health Research (NIHR) Biomedical Research Centre at the Royal Marsden NHS Foundation Trust and The Institute of Cancer Research, London, 4) the National Institute for Health Research (NIHR) Biomedical Research Centre at Imperial College London, 5) Cancer Research UK (C309/A31316).

Computer-Aided Diagnosis of Pulmonary Nodules in Rheumatoid Arthritis

(1) Background: Rheumatoid arthritis (RA) is considered a systemic inflammatory pathology characterized by symmetric polyarthritis associated with extra-articular manifestations, such as lung disease. The purpose of the present study is to use CAD in the detection of rheumatoid pulmonary nodules. In addition, we aim to identify the characteristics and associations between clinical, laboratory and imaging data in patients with rheumatoid arthritis and lung nodules. (2) Methods: The study included a number of 42 patients diagnosed with rheumatoid arthritis according to the 2010 American College of Rheumatology (ACR)/European League Against Rheumatism (EULAR) criteria, examined from January 2017 to November 2022 in the Departments of Rheumatology and Radiology and Medical Imaging of the University of Medicine and Pharmacy of Craiova. Medical records were reviewed. A retrospective blinded review of CT for biopsy-proven pulmonary nodules in RA using Veolity LungCAD software was performed (MeVis Medical Solutions AG, Bremen, Germany). Imaging was also reviewed by a senior radiologist. (3) Results: The interobserver agreement proved to be moderate (κ = 0.478) for the overall examined cases. CAD interpretation resulted in false positive results in the case of 12 lung nodules, whereas false negative results were reported in the case of 8 lung nodules. The mean time it took for the detection of lung nodules using CAD was 4.2 min per patient, whereas the detection of lung nodules by the radiologist was 8.1 min per patient. This resulted in a faster interpretation of lung CT scans, almost reducing the detection time by half (p < 0.001). (4) Conclusions: The CAD software is useful in identifying lung nodules, in shortening the interpretation time of the CT examination and also in aiding the radiologist in better assessing all the pulmonary lung nodules. However, the CAD software cannot replace the human eye yet due to the relative high rate of false positive and false negative results.

Incidental Pulmonary Nodules - What Do We Know in 2022

Lung cancer (LC) is the leading cause of cancer-related mortality worldwide, and early LC diagnosis can significantly improve outcomes and survival rates in affected patients. Implementation of LC screening programs using low-dose computed tomography CT in high-risk subjects aims to detect LC as early as possible, but so far, adoption of screening programs into routine clinical care has been very slow. In recent years, the use of CT has significantly increased the rate of incidentally detected pulmonary nodules. Although most of those incidental pulmonary nodules (IPNs) are benign, some of them represent early-stage LC. Given the large number of IPNs detected in the range of several millions each year, this represents an additional, maybe even larger, opportunity to drive stage shift in LC diagnosis, next to LC screening programs. Comprehensive evaluation and targeted work-up of IPNs are mandatory to identify the malignant nodules from the crowd, and several guidelines provide radiologists and physicians' guidance on IPN assessment and management. However, IPNs still seem to be inadequately processed due to various reasons including insufficient reporting in the radiological report, missing communication between stakeholders, absence of patient tracking systems, and uncertainty regarding responsibilities for the IPN management. In recent years, several approaches such as lung nodule programs, patient tracking software, artificial intelligence, and communication software were introduced into clinical practice to address those shortcomings. This review evaluates the current situation of IPN management and highlights recent developments in process improvement to achieve first steps toward stage shift in LC diagnosis.

Volume Doubling Time of Pulmonary Carcinoid Tumors Measured by Computed Tomography

INTRODUCTION

Pulmonary carcinoid tumor (PCT) is a rare neuroendocrine lung neoplasm comprising approximately 2% of lung cancer diagnoses. It is classified as either localized low-grade (typical) or intermediate-grade (atypical) subtypes. PCT is known clinically to be a slow-growing cancer, however few studies have established its true growth rate when followed over time by computed tomography (CT). Therefore, we sought to determine the volume doubling time for PCTs as visualized on CT imaging.

MATERIALS AND METHODS

We conducted a retrospective analysis of all PCTs treated at our institution between 2006 and 2020. Nodule dimensions were measured using a Picture Archiving and Communication System or retrieved from radiology reports. Volume doubling time was calculated using the Schwartz formula for PCTs followed by successive CT scans during radiographic surveillance. Consistent with Fleischner Society guidelines, tumors were considered to have demonstrated definitive growth by CT only when the interval change in tumor diameter was greater than or equal to 2 mm.

RESULTS

The median volume doubling time of 13 typical PCTs was 977 days, or 2.7 years. Five atypical PCTs were followed longitudinally, with a median doubling time of 327 days, or 0.9 years.

CONCLUSIONS

Typical pulmonary carcinoid features a remarkably slow growth rate as compared to more common lung cancers. Our analysis of atypical pulmonary carcinoid included too few cases to offer definitive conclusions. It is conceivable that clinicians following current nodule surveillance guidelines may mistake incidentally detected typical carcinoids for benign non-growing lesions when followed for less than 2 years in low-risk patients.

Prediction of Lung Nodule Progression with an Uncertainty-Aware Hierarchical Probabilistic Network

Predicting whether a lung nodule will grow, remain stable or regress over time, especially early in its follow-up, would help doctors prescribe personalized treatments and better surgical planning. However, the multifactorial nature of lung tumour progression hampers the identification of growth patterns. In this work, we propose a deep hierarchical generative and probabilistic network that, given an initial image of the nodule, predicts whether it will grow, quantifies its future size and provides its expected semantic appearance at a future time. Unlike previous solutions, our approach also estimates the uncertainty in the predictions from the intrinsic noise in medical images and the inter-observer variability in the annotations. The evaluation of this method on an independent test set reported a future tumour growth size mean absolute error of 1.74 mm, a nodule segmentation Dice’s coefficient of 78% and a tumour growth accuracy of 84% on predictions made up to 24 months ahead. Due to the lack of similar methods for providing future lung tumour growth predictions, along with their associated uncertainty, we adapted equivalent deterministic and alternative generative networks (i.e., probabilistic U-Net, Bayesian test dropout and Pix2Pix). Our method outperformed all these methods, corroborating the adequacy of our approach.

Prevalence and clinical characteristics of malignant lung nodules in tuberculosis endemic area in a single tertiary centre

BACKGROUND

Lung nodule management remains a challenge to clinicians, especially in endemic tuberculosis areas. Different guidelines are available with various recommendations; however, the suitability of these guidelines for the Asian population is still unclear. Our study described the prevalence of malignant lung nodules among nodules measuring 2-30 mm, the demographic and characteristics of lung nodules between benign and malignant groups, and the clinician's clinical practice in managing lung nodules.

METHOD

Retrospective review of lung nodules from the computed tomography archiving and communication system (PACS) database and clinical data from January 2019 to January 2022. The data was analysed by using chi square, mann whitney test and simple logistic regression.

RESULTS

There were 288 nodules measuring 2-30 mm identified; 49 nodules underwent biopsy. Twenty-seven (55%) biopsied nodules were malignant, (prevalence of 9.4%). Among the malignant lung nodules, 74% were adenocarcinoma (n = 20). The commonest benign nodules were granuloma n = 12 (55%). In nodules > 8 mm, the median age of malignant and benign was 72 ± 12 years and 66 ± 16 years, respectively (p = 0.024). There was a significant association of benign nodules (> 8 mm) in subjects with previous or concurrent tuberculosis (p = 0.008). Benign nodules are also associated with nodule size ≤ 8 mm, without spiculation (p < 0.001) and absence of emphysema (p = 0.007). The nodule size and the presence of spiculation are factors to make the clinicians proceed with tissue biopsy. Spiculated nodules and increased nodule size had 11 and 13 times higher chances of undergoing biopsy respectively (p < 0.001).) Previous history of tuberculosis had a 0.874 reduced risk of progression to malignant lung nodules (p = 0.013). These findings implied that these three factors are important risk factors for malignant lung nodules. There was no mortality association between benign and malignant. Using Brock's probability of malignancy, nodules ≤ 8 mm had a low probability of malignancy.

CONCLUSION

The prevalence of malignant lung nodules in our centre was comparatively lower than non-Asian countries. Older age, the presence of emphysema, and spiculation are associated with malignancy. Clinical judgment is of utmost importance in managing these patients. Fleishner guideline is still being used as a reference by our clinician.

Derivation and validation of a prediction model for patients with lung nodules malignancy regardless of mediastinal/hilar lymphadenopathy

BACKGROUND

Clinical prediction models to classify lung nodules often exclude patients with mediastinal/hilar lymphadenopathy, although the presence of mediastinal/hilar lymphadenopathy does not always indicate malignancy. Herein, we developed and validated a multimodal prediction model for lung nodules in which patients with mediastinal/hilar lymphadenopathy were included.

METHODS

A single-center retrospective study was conducted. We developed and validated a logistic regression model including patients with mediastinal/hilar lymphadenopathy. Discrimination of the model was assessed by area under the operating curve. Goodness of fit test was performed via the Hosmer-Lemeshow test, and a nomogram of the logistic regression model was drawn.

RESULTS

There were 311 cases included in the final analysis. A logistic regression model was developed and validated. There were nine independent variables included in the model. The aera under the curve (AUC) of the validation set was 0.91 (95% confidence interval [CI]: 0.85-0.98). In the validation set with mediastinal/hilar lymphadenopathy, the AUC was 0.95 (95% CI: 0.90-0.99). The goodness-of-fit test was 0.22.

CONCLUSIONS

We developed and validated a multimodal risk prediction model for lung nodules with excellent discrimination and calibration, regardless of mediastinal/hilar lymphadenopathy. This broadens the application of lung nodule prediction models. Furthermore, mediastinal/hilar lymphadenopathy added value for predicting lung nodule malignancy in clinical practice.

A predictive nomogram for lymph node metastasis in part-solid invasive lung adenocarcinoma: A complement to the IASLC novel grading system

Background

The International Association for the Study of Lung Cancer (IASLC) proposed a novel grading system for invasive lung adenocarcinoma, but lymphatic invasion was not evaluated. Meanwhile, the scope of lymph node dissection in part-solid invasive lung adenocarcinoma (PSILA) is still controversial. Therefore, this study aims to explore preoperative risk factors for lymph node metastasis in PSILA, to provide reference for intraoperative dissection of lymph nodes.

Methods

From 2018 to 2020, clinical data of patients (stage cN0) consecutively diagnosed as PSILA were retrospectively analyzed and classified according to the novel grading system. Logistic regression was conducted to screen the clinicopathological factors of lymph node metastasis in PSILA.

Results

A large cohort of 960 patients with PSILA who underwent lobectomy or sub-lobectomy were enrolled. By logistic regression analyses, solid part size, bronchial cutoff sign, spiculation, and carbohydrate antigen 199 (CA199) were eventually identified as independent risk factors for lymph node metastasis, based on which a nomogram was built to preoperatively predict the risk of lymph node metastasis [area under the receiver operating characteristic curve (AUC)=0.858; concordance index = 0.857; best cutoff, 0.027]. This suggests that intraoperative systematic lymph node dissection is recommended when the predicted risk value exceeds 0.027. Reproducibility of the novel grading system was verified.

Conclusions

The novel IASLC grading system was applicative in real world. The nomogram for preoperative prediction of lymph node metastasis may provide reference for the lymph node dissection strategy during PSILA surgeries.

Possible Bias in Supervised Deep Learning Algorithms for CT Lung Nodule Detection and Classification

Simple Summary

Artificial Intelligence (AI) algorithms can assist clinicians in their daily tasks by automatically detecting and/or classifying nodules in chest CT scans. Bias of such algorithms is one of the reasons why implementation of them in clinical practice is still not widely adopted. There is no published review on the bias that these algorithms may contain. This review aims to present different types of bias in such algorithms and present possible ways to mitigate them. Only then it would be possible to ensure that these algorithms work as intended under many different clinical settings.

Abstract

Artificial Intelligence (AI) algorithms for automatic lung nodule detection and classification can assist radiologists in their daily routine of chest CT evaluation. Even though many AI algorithms for these tasks have already been developed, their implementation in the clinical workflow is still largely lacking. Apart from the significant number of false-positive findings, one of the reasons for that is the bias that these algorithms may contain. In this review, different types of biases that may exist in chest CT AI nodule detection and classification algorithms are listed and discussed. Examples from the literature in which each type of bias occurs are presented, along with ways to mitigate these biases. Different types of biases can occur in chest CT AI algorithms for lung nodule detection and classification. Mitigation of them can be very difficult, if not impossible to achieve completely.

Form Factors as Potential Imaging Biomarkers to Differentiate Benign vs. Malignant Lung Lesions on CT Scans

Indeterminate lung nodules detected on CT scans are common findings in clinical practice. Their correct assessment is critical, as early diagnosis of malignancy is crucial to maximise the treatment outcome. In this work, we evaluated the role of form factors as imaging biomarkers to differentiate benign vs. malignant lung lesions on CT scans. We tested a total of three conventional imaging features, six form factors, and two shape features for significant differences between benign and malignant lung lesions on CT scans. The study population consisted of 192 lung nodules from two independent datasets, containing 109 (38 benign, 71 malignant) and 83 (42 benign, 41 malignant) lung lesions, respectively. The standard of reference was either histological evaluation or stability on radiological followup. The statistical significance was determined via the Mann-Whitney U nonparametric test, and the ability of the form factors to discriminate a benign vs. a malignant lesion was assessed through multivariate prediction models based on Support Vector Machines. The univariate analysis returned four form factors (Angelidakis compactness and flatness, Kong flatness, and maximum projection sphericity) that were significantly different between the benign and malignant group in both datasets. In particular, we found that the benign lesions were on average flatter than the malignant ones; conversely, the malignant ones were on average more compact (isotropic) than the benign ones. The multivariate prediction models showed that adding form factors to conventional imaging features improved the prediction accuracy by up to 14.5 pp. We conclude that form factors evaluated on lung nodules on CT scans can improve the differential diagnosis between benign and malignant lesions.

Postoperative computed tomography of insufflated lung specimens obtained by video-assisted thoracic surgery: detection and margin assessment of pulmonary nodules

Objective

To investigate the utility of computed tomography (CT) scans to detect and assess the margin status of pulmonary nodules that were insufflated after being resected by video-assisted thoracic surgery.

Materials and Methods

This was a novel multicenter study conducted at two national referral centers for thoracic diseases. Patients suspected of having lung cancer underwent video-assisted thoracic surgery for the resection of pulmonary nodules, which were submitted to postoperative CT. Measurements from the CT scans were compared with the results of the histopathological analysis.

Results

A total of 37 pulmonary nodules from 37 patients were evaluated. The mean age of the patients was 65 years (range, 36-84 years), and 27 (73%) were female. A CT analysis of insufflated specimens identified all 37 nodules, and 33 of those nodules were found to have tumor-free margins. The histopathological analysis revealed lung cancer in 30 of the nodules, all with tumor-free margins, and benign lesions in the seven remaining nodules.

Conclusion

Postoperative CT of insufflated suspicious lung lesions provides real-time detection of pulmonary nodules and satisfactory assessment of tumor margins. This initial study shows that CT of insufflated lung lesions can be a valuable tool at centers where intraoperative histopathological analysis is unavailable.

A Cost-Effective and Non-Invasive pfeRNA-Based Test Differentiates Benign and Suspicious Pulmonary Nodules from Malignant Ones

The ability to differentiate between benign, suspicious, and malignant pulmonary nodules is imperative for definitive intervention in patients with early stage lung cancers. Here, we report that plasma protein functional effector sncRNAs (pfeRNAs) serve as non-invasive biomarkers for determining both the existence and the nature of pulmonary nodules in a three-stage study that included the healthy group, patients with benign pulmonary nodules, patients with suspicious nodules, and patients with malignant nodules. Following the standards required for a clinical laboratory improvement amendments (CLIA)-compliant laboratory-developed test (LDT), we identified a pfeRNA classifier containing 8 pfeRNAs in 108 biospecimens from 60 patients by sncRNA deep sequencing, deduced prediction rules using a separate training cohort of 198 plasma specimens, and then applied the prediction rules to another 230 plasma specimens in an independent validation cohort. The pfeRNA classifier could (1) differentiate patients with or without pulmonary nodules with an average sensitivity and specificity of 96.2% and 97.35% and (2) differentiate malignant versus benign pulmonary nodules with an average sensitivity and specificity of 77.1% and 74.25%. Our biomarkers are cost-effective, non-invasive, sensitive, and specific, and the qPCR-based method provides the possibility for automatic testing of robotic applications.

Lung Nodule Malignancy Prediction in Sequential CT Scans: Summary of ISBI 2018 Challenge

Lung cancer is by far the leading cause of cancer death in the US. Recent studies have demonstrated the effectiveness of screening using low dose CT (LDCT) in reducing lung cancer related mortality. While lung nodules are detected with a high rate of sensitivity, this exam has a low specificity rate and it is still difficult to separate benign and malignant lesions. The ISBI 2018 Lung Nodule Malignancy Prediction Challenge, developed by a team from the Quantitative Imaging Network of the National Cancer Institute, was focused on the prediction of lung nodule malignancy from two sequential LDCT screening exams using automated (non-manual) algorithms. We curated a cohort of 100 subjects who participated in the National Lung Screening Trial and had established pathological diagnoses. Data from 30 subjects were randomly selected for training and the remaining was used for testing. Participants were evaluated based on the area under the receiver operating characteristic curve (AUC) of nodule-wise malignancy scores generated by their algorithms on the test set. The challenge had 17 participants, with 11 teams submitting reports with method description, mandated by the challenge rules. Participants used quantitative methods, resulting in a reporting test AUC ranging from 0.698 to 0.913. The top five contestants used deep learning approaches, reporting an AUC between 0.87 - 0.91. The team's predictor did not achieve significant differences from each other nor from a volume change estimate (p =.05 with Bonferroni-Holm's correction).

Undiagnosed solitary caseating granulomas: Is lung resection surgery a feasible method for diagnosis and treatment?

BACKGROUND

In many patients, low-dose computed tomography (CT) screening for lung cancer reveals asymptomatic pulmonary nodules. Lung resection surgery may be indicated in these patients; however, distinguishing malignancies from benign lesions preoperatively can be challenging.

METHODS

From 2013 to 2018, 4181 patients undergoing surgery for pulmonary nodules were reviewed at National Taiwan University Hospital, and 837 were diagnosed with benign pathologies. Only patients with pathological diagnosis as caseating granulomatous inflammation were included, sixty-nine patients were then analyzed for preoperative clinical and imaging characteristics, surgical methods and complications, pathogens, medical treatment and outcomes. Mycobacterial evidence was obtained from the culture of respiratory or surgical specimen.

RESULTS

Overall, 68% of the patients were asymptomatic before surgery. More than half of the nodules were in the upper lobes, and all patients underwent video-assisted thoracoscopic surgery (VATS). Some patients (14.5%) developed grade I complications, and the mean postoperative hospital stay was 4 days. The final pathology reports of 20% benign entities postoperatively, and caseating granulomatous inflammation accounted for a significant part. MTB and NTM were cultured from one-fourth of the patients respectively. All patients with confirmed MTB infection received antimycobacterial treatment, while the medical treatment in NTM-infected patients was decided by the infectious disease specialists. The mean follow-up period was 736 days, and no recurrence was found.

CONCLUSION

Lung resection surgery is an aggressive but safe and feasible method for diagnosing MTB- or NTM-associated pulmonary nodules, and, potentially, an effective therapeutic tool for patients with undiagnosed MTB- or NTM-associated pulmonary nodules.

Accelerated Stack-of-Spirals Free-Breathing Three-Dimensional Ultrashort Echo Time Lung Magnetic Resonance Imaging: A Feasibility Study in Patients With Breast Cancer

Purpose

To investigate the clinical feasibility of accelerated free-breathing stack-of-spirals (spiral) three-dimensional (3D) ultrashort echo time (UTE) lung magnetic resonance imaging (MRI) using iterative self‐consistent parallel imaging reconstruction from arbitrary k‐space (SPIRiT) algorithm in patients with breast cancer.

Methods

The institutional review board approved this prospective study and patients’ informed consents were obtained. Between June and August 2018, 29 female patients with breast cancer underwent 3-T MRI including accelerated free-breathing spiral 3D UTE (0.98-mm isotropic spatial resolution; echo time, 0.05 msec) of the lungs and thin-section chest computed tomography (CT). Two radiologists evaluated the image quality and pulmonary nodules on MRI were assessed and compared, CT as a reference.

Results

The pulmonary vessels and bronchi were visible consistently up to the sub-sub-segmental and sub-segmental branch levels, respectively, on accelerated spiral 3D UTE. The overall image quality was evaluated as good and excellent for 70.7% of accelerated spiral 3D UTE images (reviewer [R]1, 72.4% [21/29]; R2, 69.0% [20/29]) and acceptable for 20.7% (both R1 and R2, 20.7% [6/29]). Five patients on CT revealed 141 pulmonary metastatic nodules (5.3 ± 2.6 mm); the overall nodule detection rate of accelerated spiral 3D UTE was sensitivity of 90.8% (128/141), accuracy of 87.7%, and positive predictive value of 96.2%. In the Bland-Altman plot analysis comparing nodule size between CT and MRI, 132/141 nodules (93.6%) were inside the limits of agreement.

Conclusion

Accelerated free-breathing spiral 3D UTE using the SPIRiT algorithm could be a potential alternative to CT for oncology patients.

PULMONARY NODULE DETECTION IN CHEST CT USING A DEEP LEARNING-BASED RECONSTRUCTION ALGORITHM

This study's aim was to assess whether deep learning image reconstruction (DLIR) techniques are non-inferior to ASIR-V for the clinical task of pulmonary nodule detection in chest computed tomography. Up to 6 (range 3-6, mean 4.2) artificial lung nodules (diameter: 3, 5, 8 mm; density: -800, -630, +100 HU) were inserted at different locations in the Kyoto Kagaku Lungman phantom. In total, 16 configurations (10 abnormal, 6 normal) were scanned at 7.6, 3, 1.6 and 0.38 mGy CTDIvol (respectively 0, 60, 80 and 95% dose reduction). Images were reconstructed using 50% ASIR-V and a deep learning-based algorithm with low (DL-L), medium (DL-M) and high (DL-H) strength. Four chest radiologists evaluated 256 series by locating and scoring nodules on a five-point scale. No statistically significant difference was found among the reconstruction algorithms (p = 0.987, average across readers AUC: 0.555, 0.561, 0.557, 0.558 for ASIR-V, DL-L, DL-M, DL-H).

Deep Mining Generation of Lung Cancer Malignancy Models from Chest X-ray Images

Lung cancer is the leading cause of cancer death and morbidity worldwide. Many studies have shown machine learning models to be effective in detecting lung nodules from chest X-ray images. However, these techniques have yet to be embraced by the medical community due to several practical, ethical, and regulatory constraints stemming from the “black-box” nature of deep learning models. Additionally, most lung nodules visible on chest X-rays are benign; therefore, the narrow task of computer vision-based lung nodule detection cannot be equated to automated lung cancer detection. Addressing both concerns, this study introduces a novel hybrid deep learning and decision tree-based computer vision model, which presents lung cancer malignancy predictions as interpretable decision trees. The deep learning component of this process is trained using a large publicly available dataset on pathological biomarkers associated with lung cancer. These models are then used to inference biomarker scores for chest X-ray images from two independent data sets, for which malignancy metadata is available. Next, multi-variate predictive models were mined by fitting shallow decision trees to the malignancy stratified datasets and interrogating a range of metrics to determine the best model. The best decision tree model achieved sensitivity and specificity of 86.7% and 80.0%, respectively, with a positive predictive value of 92.9%. Decision trees mined using this method may be considered as a starting point for refinement into clinically useful multi-variate lung cancer malignancy models for implementation as a workflow augmentation tool to improve the efficiency of human radiologists.

Prediction of malignancy for solitary pulmonary nodules based on imaging, clinical characteristics and tumor marker levels

OBJECTIVE

To establish a prediction model of malignancy for solitary pulmonary nodules (SPNs) on the basis of imaging, clinical characteristics and tumor marker levels.

METHODS

Totally, 341 cases of SPNs were enrolled in this retrospective study, in which 70% were selected as the training group (n = 238) and the rest 30% as the verification group (n = 103). The imaging, clinical characteristics and tumor marker levels of patients with benign and malignant SPNs were compared. Influencing factors were identified using multivariate logistic regression analysis. The model was assessed by the area under the curve (AUC) of the receiver operating characteristic curve.

RESULTS

Differences were evident between patients with benign and malignant SPNs in age, gender, smoking history, carcinoembryonic antigen (CEA), neuron-specific enolase, nodule location, edge smoothing, spiculation, lobulation, vascular convergence sign, air bronchogram, ground-glass opacity, vacuole sign and calcification (all P < 0.05). Influencing factors for malignancy included age, gender, nodule location, spiculation, vacuole sign and CEA (all P < 0.05). The established model was as follows: Y = -5.368 + 0.055 × age + 1.012 × gender (female = 1, male = 0) + 1.302 × nodule location (right upper lobe = 1, others = 0) + 1.208 × spiculation (yes = 1, no = 0) + 2.164 × vacuole sign (yes = 1, no = 0) -0.054 × CEA. The AUC of the model with CEA was 0.818 (95% confidence interval, 0.763-0.865), with a sensitivity of 64.80% and a specificity of 84.96%, and the stability was better through internal verification.

CONCLUSIONS

The prediction model established in our study exhibits better accuracy and internal stability in predicting the probability of malignancy for SPNs.

[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。

Performance Evaluation of a Deep Learning System for Differential Diagnosis of Lung Cancer With Conventional CT and FDG PET/CT Using Transfer Learning and Metadata

PURPOSE

We aimed to evaluate the performance of a deep learning system for differential diagnosis of lung cancer with conventional CT and FDG PET/CT using transfer learning (TL) and metadata.

METHODS

A total of 359 patients with a lung mass or nodule who underwent noncontrast chest CT and FDG PET/CT prior to treatment were enrolled retrospectively. All pulmonary lesions were classified by pathology (257 malignant, 102 benign). Deep learning classification models based on ResNet-18 were developed using the pretrained weights obtained from ImageNet data set. We propose a deep TL model for differential diagnosis of lung cancer using CT imaging data and metadata with SUVmax and lesion size derived from PET/CT. The area under the receiver operating characteristic curve (AUC) of the deep learning model was measured as a performance metric and verified by 5-fold cross-validation.

RESULTS

The performance metrics of the conventional CT model were generally better than those of the CT of PET/CT model. Introducing metadata with SUVmax and lesion size derived from PET/CT into baseline CT models improved the diagnostic performance of the CT of PET/CT model (AUC = 0.837 vs 0.762) and the conventional CT model (AUC = 0.877 vs 0.817).

CONCLUSIONS

Deep TL models with CT imaging data provide good diagnostic performance for lung cancer, and the conventional CT model showed overall better performance than the CT of PET/CT model. Metadata information derived from PET/CT can improve the performance of deep learning systems.