Deep Learning for Malignancy Risk Estimation of Pulmonary Nodules Detected at Low-Dose Screening CT

Impact on Image Quality and Diagnostic Performance of Dual-Layer Detector Spectral CT for Pulmonary Subsolid Nodules: Comparison With Hybrid and Model-Based Iterative Reconstruction

OBJECTIVE

To evaluate the image quality and diagnostic performance of pulmonary subsolid nodules on conventional iterative algorithms, virtual monoenergetic images (VMIs), and electron density mapping (EDM) using a dual-layer detector spectral CT (DLSCT).

METHODS

This retrospective study recruited 270 patients who underwent DLSCT scan for lung nodule screening or follow-up. All CT examinations with subsolid nodules (pure ground-glass nodules [GGNs] or part-solid nodules) were reconstructed with hybrid and model-based iterative reconstruction, VMI at 40, 70, 100, and 130 keV levels, and EDM. The CT number, objective image noise, signal-to-noise ratio, contrast-to-noise ratio, diameter, and volume of subsolid nodules were measured for quantitative analysis. The overall image quality, image noise, visualization of nodules, artifact, and sharpness were subjectively rated by 2 thoracic radiologists on a 5-point scale (1 = unacceptable, 5 = excellent) in consensus. The objective image quality measurements, diameter, and volume were compared among the 7 groups with a repeated 1-way analysis of variance. The subjective scores were compared with Kruskal-Wallis test.

RESULTS

A total of 198 subsolid nodules, including 179 pure GGNs, and 19 part-solid nodules were identified. Based on the objective analysis, EDM had the highest signal-to-noise ratio (164.71 ± 133.60; P < 0.001) and contrast-to-noise ratio (227.97 ± 161.96; P < 0.001) among all image sets. Furthermore, EDM had a superior mean subjective rating score (4.80 ± 0.42) for visualization of GGNs compared to other reconstructed images (all P < 0.001), although the model-based iterative reconstruction had superior subjective scores of overall image quality. For pure GGNs, the measured diameter and volume did not significantly differ among different reconstructions (both P > 0.05).

CONCLUSIONS

EDM derived from DLSCT enabled improved image quality and lesion conspicuity for the evaluation of lung subsolid nodules compared to conventional iterative reconstruction algorithms and VMIs.

Deep learning for malignancy risk estimation of incidental sub-centimeter pulmonary nodules on CT images

OBJECTIVES

To establish deep learning models for malignancy risk estimation of sub-centimeter pulmonary nodules incidentally detected by chest CT and managed in clinical settings.

MATERIALS AND METHODS

Four deep learning models were trained using CT images of sub-centimeter pulmonary nodules from West China Hospital, internally tested, and externally validated on three cohorts. The four models respectively learned 3D deep features from the baseline whole lung region, baseline image patch where the nodule located, baseline nodule box, and baseline plus follow-up nodule boxes. All regions of interest were automatically segmented except that the nodule boxes were additionally manually checked. The performance of models was compared with each other and that of three respiratory clinicians.

RESULTS

There were 1822 nodules (981 malignant) in the training set, 806 (416 malignant) in the testing set, and 357 (253 malignant) totally in the external sets. The area under the curve (AUC) in the testing set was 0.754, 0.855, 0.928, and 0.942, respectively, for models derived from baseline whole lung, image patch, nodule box, and the baseline plus follow-up nodule boxes. When baseline models externally validated (follow-up images not available), the nodule-box model outperformed the other two with AUC being 0.808, 0.848, and 0.939 respectively in the three external datasets. The resident, junior, and senior clinicians achieved an accuracy of 67.0%, 82.5%, and 90.0%, respectively, in the testing set. The follow-up model performed comparably to the senior clinician.

CONCLUSION

The deep learning algorithms solely mining nodule information can efficiently predict malignancy of incidental sub-centimeter pulmonary nodules.

CLINICAL RELEVANCE STATEMENT

The established models may be valuable for supporting clinicians in routine clinical practice, potentially reducing the number of unnecessary examinations and also delays in diagnosis.

KEY POINTS

• According to different regions of interest, four deep learning models were developed and compared to evaluate the malignancy of sub-centimeter pulmonary nodules by CT images. • The models derived from baseline nodule box or baseline plus follow-up nodule boxes demonstrated sufficient diagnostic accuracy (86.4% and 90.4% in the testing set), outperforming the respiratory resident (67.0%) and junior clinician (82.5%). • The proposed deep learning methods may aid clinicians in optimizing follow-up recommendations for sub-centimeter pulmonary nodules and may lead to fewer unnecessary diagnostic interventions.

Computational risk model for predicting 2-year malignancy of pulmonary nodules using demographic and radiographic characteristics

OBJECTIVES

To determine whether discriminatory performance of a computational risk model in classifying pulmonary lesion malignancy using demographic, radiographic, and clinical characteristics is superior to the opinion of experienced providers. We hypothesized that computational risk models would outperform providers.

METHODS

Outcome of malignancy was obtained from selected patients enrolled in the NAVIGATE trial (NCT02410837). Five predictive risk models were developed using an 80:20 train-test split: univariable logistic regression model based solely on provider opinion, multivariable logistic regression model, random forest classifier, extreme gradient boosting model, and artificial neural network. Area under the receiver operating characteristic curve achieved during testing of the predictive models was compared to that of prebiopsy provider opinion baseline using the DeLong test with 10,000 bootstrapped iterations.

RESULTS

The cohort included 984 patients, 735 (74.7%) of which were diagnosed with malignancy. Factors associated with malignancy from multivariable logistic regression included age, history of cancer, largest lesion size, lung zone, and positron-emission tomography positivity. Testing area under the receiver operating characteristic curve were 0.830 for provider opinion baseline, 0.770 for provider opinion univariable logistic regression, 0.659 for multivariable logistic regression model, 0.743 for random forest classifier, 0.740 for extreme gradient boosting, and 0.679 for artificial neural network. Provider opinion baseline was determined to be the best predictive classification system.

CONCLUSIONS

Computational models predicting malignancy of pulmonary lesions using clinical, demographic, and radiographic characteristics are inferior to provider opinion. This study questions the ability of these models to provide additional insight into patient care. Expert clinician evaluation of pulmonary lesion malignancy is paramount.

Leveraging Serial Low-Dose CT Scans in Radiomics-based Reinforcement Learning to Improve Early Diagnosis of Lung Cancer at Baseline Screening

Purpose To evaluate the feasibility of leveraging serial low-dose CT (LDCT) scans to develop a radiomics-based reinforcement learning (RRL) model for improving early diagnosis of lung cancer at baseline screening. Materials and Methods In this retrospective study, 1951 participants (female patients, 822; median age, 61 years [range, 55-74 years]) (male patients, 1129; median age, 62 years [range, 55-74 years]) were randomly selected from the National Lung Screening Trial between August 2002 and April 2004. An RRL model using serial LDCT scans (S-RRL) was trained and validated using data from 1404 participants (372 with lung cancer) containing 2525 available serial LDCT scans up to 3 years. A baseline RRL (B-RRL) model was trained with only LDCT scans acquired at baseline screening for comparison. The 547 held-out individuals (150 with lung cancer) were used as an independent test set for performance evaluation. The area under the receiver operating characteristic curve (AUC) and the net reclassification index (NRI) were used to assess the performances of the models in the classification of screen-detected nodules. Results Deployment to the held-out baseline scans showed that the S-RRL model achieved a significantly higher test AUC (0.88 [95% CI: 0.85, 0.91]) than both the Brock model (AUC, 0.84 [95% CI: 0.81, 0.88]; P = .02) and the B-RRL model (AUC, 0.86 [95% CI: 0.83, 0.90]; P = .02). Lung cancer risk stratification was significantly improved by the S-RRL model as compared with Lung CT Screening Reporting and Data System (NRI, 0.29; P < .001) and the Brock model (NRI, 0.12; P = .008). Conclusion The S-RRL model demonstrated the potential to improve early diagnosis and risk stratification for lung cancer at baseline screening as compared with the B-RRL model and clinical models. Keywords: Radiomics-based Reinforcement Learning, Lung Cancer Screening, Low-Dose CT, Machine Learning © RSNA, 2024 Supplemental material is available for this article.

Deep Learning Models for Predicting Malignancy Risk in CT-Detected Pulmonary Nodules: A Systematic Review and Meta-analysis

BACKGROUND

There has been growing interest in using artificial intelligence/deep learning (DL) to help diagnose prevalent diseases earlier. In this study we sought to survey the landscape of externally validated DL-based computer-aided diagnostic (CADx) models, and assess their diagnostic performance for predicting the risk of malignancy in computed tomography (CT)-detected pulmonary nodules.

METHODS

An electronic search was performed in four databases (from inception to 10 August 2023). Studies were eligible if they were peer-reviewed experimental or observational articles comparing the diagnostic performance of externally validated DL-based CADx models with models widely used in clinical practice to predict the risk of malignancy. A bivariate random-effect approach for the meta-analysis on the included studies was used.

RESULTS

Seventeen studies were included, comprising 8553 participants and 9884 nodules. Pooled analyses showed DL-based CADx models were 11.6% more sensitive than physician judgement alone, and 14.5% more than clinical risk models alone. They had a similar pooled specificity to physician judgement alone [0.77 (95% CI 0.68-0.84) v 0.81 (95% CI 0.71-0.88)], and were 7.4% more specific than clinical risk models alone. They had superior pooled areas under the receiver operating curve (AUC), with relative pooled AUCs of 1.03 (95% CI 1.00-1.07) and 1.10 (95% CI 1.07-1.13) versus physician judgement and clinical risk models alone, respectively.

CONCLUSION

DL-based models are already used in clinical practice in certain settings for nodule management. Our results show their diagnostic performance potentially justifies wider, more routine deployment alongside experienced physician readers to help inform multidisciplinary team decision-making.

Standalone deep learning versus experts for diagnosis lung cancer on chest computed tomography: a systematic review

PURPOSE

To compare the diagnostic performance of standalone deep learning (DL) algorithms and human experts in lung cancer detection on chest computed tomography (CT) scans.

MATERIALS AND METHODS

This study searched for studies on PubMed, Embase, and Web of Science from their inception until November 2023. We focused on adult lung cancer patients and compared the efficacy of DL algorithms and expert radiologists in disease diagnosis on CT scans. Quality assessment was performed using QUADAS-2, QUADAS-C, and CLAIM. Bivariate random-effects and subgroup analyses were performed for tasks (malignancy classification vs invasiveness classification), imaging modalities (CT vs low-dose CT [LDCT] vs high-resolution CT), study region, software used, and publication year.

RESULTS

We included 20 studies on various aspects of lung cancer diagnosis on CT scans. Quantitatively, DL algorithms exhibited superior sensitivity (82%) and specificity (75%) compared to human experts (sensitivity 81%, specificity 69%). However, the difference in specificity was statistically significant, whereas the difference in sensitivity was not statistically significant. The DL algorithms' performance varied across different imaging modalities and tasks, demonstrating the need for tailored optimization of DL algorithms. Notably, DL algorithms matched experts in sensitivity on standard CT, surpassing them in specificity, but showed higher sensitivity with lower specificity on LDCT scans.

CONCLUSION

DL algorithms demonstrated improved accuracy over human readers in malignancy and invasiveness classification on CT scans. However, their performance varies by imaging modality, underlining the importance of continued research to fully assess DL algorithms' diagnostic effectiveness in lung cancer.

CLINICAL RELEVANCE STATEMENT

DL algorithms have the potential to refine lung cancer diagnosis on CT, matching human sensitivity and surpassing in specificity. These findings call for further DL optimization across imaging modalities, aiming to advance clinical diagnostics and patient outcomes.

KEY POINTS

Lung cancer diagnosis by CT is challenging and can be improved with AI integration. DL shows higher accuracy in lung cancer detection on CT than human experts. Enhanced DL accuracy could lead to improved lung cancer diagnosis and outcomes.

A Self-supervised Learning-Based Fine-Grained Classification Model for Distinguishing Malignant From Benign Subcentimeter Solid Pulmonary Nodules

RATIONALE AND OBJECTIVES

Diagnosing subcentimeter solid pulmonary nodules (SSPNs) remains challenging in clinical practice. Deep learning may perform better than conventional methods in differentiating benign and malignant pulmonary nodules. This study aimed to develop and validate a model for differentiating malignant and benign SSPNs using CT images.

MATERIALS AND METHODS

This retrospective study included consecutive patients with SSPNs detected between January 2015 and October 2021 as an internal dataset. Malignancy was confirmed pathologically; benignity was confirmed pathologically or via follow-up evaluations. The SSPNs were segmented manually. A self-supervision pre-training-based fine-grained network was developed for predicting SSPN malignancy. The pre-trained model was established using data from the National Lung Screening Trial, Lung Nodule Analysis 2016, and a database of 5478 pulmonary nodules from the previous study, with subsequent fine-tuning using the internal dataset. The model's efficacy was investigated using an external cohort from another center, and its accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve (AUC) were determined.

RESULTS

Overall, 1276 patients (mean age, 56 ± 10 years; 497 males) with 1389 SSPNs (mean diameter, 7.5 ± 2.0 mm; 625 benign) were enrolled. The internal dataset was specifically enriched for malignancy. The model's performance in the internal testing set (316 SSPNs) was: AUC, 0.964 (95% confidence interval (95%CI): 0.942-0.986); accuracy, 0.934; sensitivity, 0.965; and specificity, 0.908. The model's performance in the external test set (202 SSPNs) was: AUC, 0.945 (95% CI: 0.910-0.979); accuracy, 0.911; sensitivity, 0.977; and specificity, 0.860.

CONCLUSION

This deep learning model was robust and exhibited good performance in predicting the malignancy of SSPNs, which could help optimize patient management.

Enhancing cancer prediction in challenging screen-detected incident lung nodules using time-series deep learning

Lung cancer screening (LCS) using annual computed tomography (CT) scanning significantly reduces mortality by detecting cancerous lung nodules at an earlier stage. Deep learning algorithms can improve nodule malignancy risk stratification. However, they have typically been used to analyse single time point CT data when detecting malignant nodules on either baseline or incident CT LCS rounds. Deep learning algorithms have the greatest value in two aspects. These approaches have great potential in assessing nodule change across time-series CT scans where subtle changes may be challenging to identify using the human eye alone. Moreover, they could be targeted to detect nodules developing on incident screening rounds, where cancers are generally smaller and more challenging to detect confidently. Here, we show the performance of our Deep learning-based Computer-Aided Diagnosis model integrating Nodule and Lung imaging data with clinical Metadata Longitudinally (DeepCAD-NLM-L) for malignancy prediction. DeepCAD-NLM-L showed improved performance (AUC = 88%) against models utilizing single time-point data alone. DeepCAD-NLM-L also demonstrated comparable and complementary performance to radiologists when interpreting the most challenging nodules typically found in LCS programs. It also demonstrated similar performance to radiologists when assessed on out-of-distribution imaging dataset. The results emphasize the advantages of using time-series and multimodal analyses when interpreting malignancy risk in LCS.

Attention pyramid pooling network for artificial diagnosis on pulmonary nodules

The development of automated tools using advanced technologies like deep learning holds great promise for improving the accuracy of lung nodule classification in computed tomography (CT) imaging, ultimately reducing lung cancer mortality rates. However, lung nodules can be difficult to detect and classify, from CT images since different imaging modalities may provide varying levels of detail and clarity. Besides, the existing convolutional neural network may struggle to detect nodules that are small or located in difficult-to-detect regions of the lung. Therefore, the attention pyramid pooling network (APPN) is proposed to identify and classify lung nodules. First, a strong feature extractor, named vgg16, is used to obtain features from CT images. Then, the attention primary pyramid module is proposed by combining the attention mechanism and pyramid pooling module, which allows for the fusion of features at different scales and focuses on the most important features for nodule classification. Finally, we use the gated spatial memory technique to decode the general features, which is able to extract more accurate features for classifying lung nodules. The experimental results on the LIDC-IDRI dataset show that the APPN can achieve highly accurate and effective for classifying lung nodules, with sensitivity of 87.59%, specificity of 90.46%, accuracy of 88.47%, positive predictive value of 95.41%, negative predictive value of 76.29% and area under receiver operating characteristic curve of 0.914.

Development and external validation of a multimodal integrated feature neural network (MIFNN) for the diagnosis of malignancy in small pulmonary nodules (≤10 mm)

Objectives. Current lung cancer screening protocols primarily evaluate pulmonary nodules, yet often neglect the malignancy risk associated with small nodules (≤10 mm). This study endeavors to optimize the management of pulmonary nodules in this population by devising and externally validating a Multimodal Integrated Feature Neural Network (MIFNN). We hypothesize that the fusion of deep learning algorithms with morphological nodule features will significantly enhance diagnostic accuracy.Materials and Methods. Data were retrospectively collected from the Lung Nodule Analysis 2016 (LUNA16) dataset and four local centers in Beijing, China. The study includes patients with small pulmonary nodules (≤10 mm). We developed a neural network, termed MIFNN, that synergistically combines computed tomography (CT) images and morphological characteristics of pulmonary nodules. The network is designed to acquire clinically relevant deep learning features, thereby elevating the diagnostic accuracy of existing models. Importantly, the network's simple architecture and use of standard screening variables enable seamless integration into standard lung cancer screening protocols.Results. In summary, the study analyzed a total of 382 small pulmonary nodules (85 malignant) from the LUNA16 dataset and 101 small pulmonary nodules (33 malignant) obtained from four specialized centers in Beijing, China, for model training and external validation. Both internal and external validation metrics indicate that the MIFNN significantly surpasses extant state-of-the-art models, achieving an internal area under the curve (AUC) of 0.890 (95% CI: 0.848-0.932) and an external AUC of 0.843 (95% CI: 0.784-0.891).Conclusion. The MIFNN model significantly enhances the diagnostic accuracy of small pulmonary nodules, outperforming existing benchmarks by Zhanget alwith a 6.34% improvement for nodules less than 10 mm. Leveraging advanced integration techniques for imaging and clinical data, MIFNN increases the efficiency of lung cancer screenings and optimizes nodule management, potentially reducing false positives and unnecessary biopsies.Clinical relevance statement. The MIFNN enhances lung cancer screening efficiency and patient management for small pulmonary nodules, while seamlessly integrating into existing workflows due to its reliance on standard screening variables.

Prognostication of lung adenocarcinomas using CT-based deep learning of morphological and histopathological features: a retrospective dual-institutional study

OBJECTIVES

To develop and validate CT-based deep learning (DL) models that learn morphological and histopathological features for lung adenocarcinoma prognostication, and to compare them with a previously developed DL discrete-time survival model.

METHODS

DL models were trained to simultaneously predict five morphological and histopathological features using preoperative chest CT scans from patients with resected lung adenocarcinomas. The DL score was validated in temporal and external test sets, with freedom from recurrence (FFR) and overall survival (OS) as outcomes. Discrimination was evaluated using the time-dependent area under the receiver operating characteristic curve (TD-AUC) and compared with the DL discrete-time survival model. Additionally, we performed multivariable Cox regression analysis.

RESULTS

In the temporal test set (640 patients; median age, 64 years), the TD-AUC was 0.79 for 5-year FFR and 0.73 for 5-year OS. In the external test set (846 patients; median age, 65 years), the TD-AUC was 0.71 for 5-year OS, equivalent to the pathologic stage (0.71 vs. 0.71 [p = 0.74]). The prognostic value of the DL score was independent of clinical factors (adjusted per-percentage hazard ratio for FFR (temporal test), 1.02 [95% CI: 1.01-1.03; p < 0.001]; OS (temporal test), 1.01 [95% CI: 1.002-1.02; p = 0.01]; OS (external test), 1.01 [95% CI: 1.005-1.02; p < 0.001]). Our model showed a higher TD-AUC than the DL discrete-time survival model, but without statistical significance (2.5-year OS: 0.73 vs. 0.68; p = 0.13).

CONCLUSION

The CT-based prognostic score from collective deep learning of morphological and histopathological features showed potential in predicting survival in lung adenocarcinomas.

CLINICAL RELEVANCE STATEMENT

Collective CT-based deep learning of morphological and histopathological features presents potential for enhancing lung adenocarcinoma prognostication and optimizing pre-/postoperative management.

KEY POINTS

• A CT-based prognostic model was developed using collective deep learning of morphological and histopathological features from preoperative CT scans of 3181 patients with resected lung adenocarcinoma. • The prognostic performance of the model was comparable-to-higher performance than the pathologic T category or stage. • Our approach yielded a higher discrimination performance than the direct survival prediction model, but without statistical significance (0.73 vs. 0.68; p=0.13).

Deep Learning Assessment of Small Renal Masses at Contrast-enhanced Multiphase CT

Background Accurate characterization of suspicious small renal masses is crucial for optimized management. Deep learning (DL) algorithms may assist with this effort. Purpose To develop and validate a DL algorithm for identifying benign small renal masses at contrast-enhanced multiphase CT. Materials and Methods Surgically resected renal masses measuring 3 cm or less in diameter at contrast-enhanced CT were included. The DL algorithm was developed by using retrospective data from one hospital between 2009 and 2021, with patients randomly allocated in a training and internal test set ratio of 8:2. Between 2013 and 2021, external testing was performed on data from five independent hospitals. A prospective test set was obtained between 2021 and 2022 from one hospital. Algorithm performance was evaluated by using the area under the receiver operating characteristic curve (AUC) and compared with the results of seven clinicians using the DeLong test. Results A total of 1703 patients (mean age, 56 years ± 12 [SD]; 619 female) with a single renal mass per patient were evaluated. The retrospective data set included 1063 lesions (874 in training set, 189 internal test set); the multicenter external test set included 537 lesions (12.3%, 66 benign) with 89 subcentimeter (≤1 cm) lesions (16.6%); and the prospective test set included 103 lesions (13.6%, 14 benign) with 20 (19.4%) subcentimeter lesions. The DL algorithm performance was comparable with that of urological radiologists: for the external test set, AUC was 0.80 (95% CI: 0.75, 0.85) versus 0.84 (95% CI: 0.78, 0.88) (P = .61); for the prospective test set, AUC was 0.87 (95% CI: 0.79, 0.93) versus 0.92 (95% CI: 0.86, 0.96) (P = .70). For subcentimeter lesions in the external test set, the algorithm and urological radiologists had similar AUC of 0.74 (95% CI: 0.63, 0.83) and 0.81 (95% CI: 0.68, 0.92) (P = .78), respectively. Conclusion The multiphase CT-based DL algorithm showed comparable performance with that of radiologists for identifying benign small renal masses, including lesions of 1 cm or less. Published under a CC BY 4.0 license. Supplemental material is available for this article.

Assistive AI in Lung Cancer Screening: A Retrospective Multinational Study in the United States and Japan

Purpose To evaluate the impact of an artificial intelligence (AI) assistant for lung cancer screening on multinational clinical workflows. Materials and Methods An AI assistant for lung cancer screening was evaluated on two retrospective randomized multireader multicase studies where 627 (141 cancer-positive cases) low-dose chest CT cases were each read twice (with and without AI assistance) by experienced thoracic radiologists (six U.S.-based or six Japan-based radiologists), resulting in a total of 7524 interpretations. Positive cases were defined as those within 2 years before a pathology-confirmed lung cancer diagnosis. Negative cases were defined as those without any subsequent cancer diagnosis for at least 2 years and were enriched for a spectrum of diverse nodules. The studies measured the readers' level of suspicion (on a 0-100 scale), country-specific screening system scoring categories, and management recommendations. Evaluation metrics included the area under the receiver operating characteristic curve (AUC) for level of suspicion and sensitivity and specificity of recall recommendations. Results With AI assistance, the radiologists' AUC increased by 0.023 (0.70 to 0.72; P = .02) for the U.S. study and by 0.023 (0.93 to 0.96; P = .18) for the Japan study. Scoring system specificity for actionable findings increased 5.5% (57% to 63%; P < .001) for the U.S. study and 6.7% (23% to 30%; P < .001) for the Japan study. There was no evidence of a difference in corresponding sensitivity between unassisted and AI-assisted reads for the U.S. (67.3% to 67.5%; P = .88) and Japan (98% to 100%; P > .99) studies. Corresponding stand-alone AI AUC system performance was 0.75 (95% CI: 0.70, 0.81) and 0.88 (95% CI: 0.78, 0.97) for the U.S.- and Japan-based datasets, respectively. Conclusion The concurrent AI interface improved lung cancer screening specificity in both U.S.- and Japan-based reader studies, meriting further study in additional international screening environments. Keywords: Assistive Artificial Intelligence, Lung Cancer Screening, CT Supplemental material is available for this article. Published under a CC BY 4.0 license.

Predicting Invasiveness of Lung Adenocarcinoma at Chest CT with Deep Learning Ternary Classification Models

Background Preoperative discrimination of preinvasive, minimally invasive, and invasive adenocarcinoma at CT informs clinical management decisions but may be challenging for classifying pure ground-glass nodules (pGGNs). Deep learning (DL) may improve ternary classification. Purpose To determine whether a strategy that includes an adjudication approach can enhance the performance of DL ternary classification models in predicting the invasiveness of adenocarcinoma at chest CT and maintain performance in classifying pGGNs. Materials and Methods In this retrospective study, six ternary models for classifying preinvasive, minimally invasive, and invasive adenocarcinoma were developed using a multicenter data set of lung nodules. The DL-based models were progressively modified through framework optimization, joint learning, and an adjudication strategy (simulating a multireader approach to resolving discordant nodule classifications), integrating two binary classification models with a ternary classification model to resolve discordant classifications sequentially. The six ternary models were then tested on an external data set of pGGNs imaged between December 2019 and January 2021. Diagnostic performance including accuracy, specificity, and sensitivity was assessed. The χ2 test was used to compare model performance in different subgroups stratified by clinical confounders. Results A total of 4929 nodules from 4483 patients (mean age, 50.1 years ± 9.5 [SD]; 2806 female) were divided into training (n = 3384), validation (n = 579), and internal (n = 966) test sets. A total of 361 pGGNs from 281 patients (mean age, 55.2 years ± 11.1 [SD]; 186 female) formed the external test set. The proposed strategy improved DL model performance in external testing (P < .001). For classifying minimally invasive adenocarcinoma, the accuracy was 85% and 79%, sensitivity was 75% and 63%, and specificity was 89% and 85% for the model with adjudication (model 6) and the model without (model 3), respectively. Model 6 showed a relatively narrow range (maximum minus minimum) across diagnostic indexes (accuracy, 1.7%; sensitivity, 7.3%; specificity, 0.9%) compared with the other models (accuracy, 0.6%-10.8%; sensitivity, 14%-39.1%; specificity, 5.5%-17.9%). Conclusion Combining framework optimization, joint learning, and an adjudication approach improved DL classification of adenocarcinoma invasiveness at chest CT. Published under a CC BY 4.0 license. Supplemental material is available for this article. See also the editorial by Sohn and Fields in this issue.

Radiomics analysis to predict pulmonary nodule malignancy using machine learning approaches

BACKGROUND

Low-dose CT screening can reduce lung cancer-related mortality. However, most screen-detected pulmonary abnormalities do not develop into cancer and it often remains challenging to identify malignant nodules, particularly among indeterminate nodules. We aimed to develop and assess prediction models based on radiological features to discriminate between benign and malignant pulmonary lesions detected on a baseline screen.

METHODS

Using four international lung cancer screening studies, we extracted 2060 radiomic features for each of 16 797 nodules (513 malignant) among 6865 participants. After filtering out low-quality radiomic features, 642 radiomic and 9 epidemiological features remained for model development. We used cross-validation and grid search to assess three machine learning (ML) models (eXtreme Gradient Boosted Trees, random forest, least absolute shrinkage and selection operator (LASSO)) for their ability to accurately predict risk of malignancy for pulmonary nodules. We report model performance based on the area under the curve (AUC) and calibration metrics in the held-out test set.

RESULTS

The LASSO model yielded the best predictive performance in cross-validation and was fit in the full training set based on optimised hyperparameters. Our radiomics model had a test-set AUC of 0.93 (95% CI 0.90 to 0.96) and outperformed the established Pan-Canadian Early Detection of Lung Cancer model (AUC 0.87, 95% CI 0.85 to 0.89) for nodule assessment. Our model performed well among both solid (AUC 0.93, 95% CI 0.89 to 0.97) and subsolid nodules (AUC 0.91, 95% CI 0.85 to 0.95).

CONCLUSIONS

We developed highly accurate ML models based on radiomic and epidemiological features from four international lung cancer screening studies that may be suitable for assessing indeterminate screen-detected pulmonary nodules for risk of malignancy.

AI-Driven Models for Diagnosing and Predicting Outcomes in Lung Cancer: A Systematic Review and Meta-Analysis

(1) Background: Lung cancer's high mortality due to late diagnosis highlights a need for early detection strategies. Artificial intelligence (AI) in healthcare, particularly for lung cancer, offers promise by analyzing medical data for early identification and personalized treatment. This systematic review evaluates AI's performance in early lung cancer detection, analyzing its techniques, strengths, limitations, and comparative edge over traditional methods. (2) Methods: This systematic review and meta-analysis followed the PRISMA guidelines rigorously, outlining a comprehensive protocol and employing tailored search strategies across diverse databases. Two reviewers independently screened studies based on predefined criteria, ensuring the selection of high-quality data relevant to AI's role in lung cancer detection. The extraction of key study details and performance metrics, followed by quality assessment, facilitated a robust analysis using R software (Version 4.3.0). The process, depicted via a PRISMA flow diagram, allowed for the meticulous evaluation and synthesis of the findings in this review. (3) Results: From 1024 records, 39 studies met the inclusion criteria, showcasing diverse AI model applications for lung cancer detection, emphasizing varying strengths among the studies. These findings underscore AI's potential for early lung cancer diagnosis but highlight the need for standardization amidst study variations. The results demonstrate promising pooled sensitivity and specificity of 0.87, signifying AI's accuracy in identifying true positives and negatives, despite the observed heterogeneity attributed to diverse study parameters. (4) Conclusions: AI demonstrates promise in early lung cancer detection, showing high accuracy levels in this systematic review. However, study variations underline the need for standardized protocols to fully leverage AI's potential in revolutionizing early diagnosis, ultimately benefiting patients and healthcare professionals. As the field progresses, validated AI models from large-scale perspective studies will greatly benefit clinical practice and patient care in the future.

[The contribution of artificial intelligence (AI) subsequent to the processing of thoracic imaging]

The contribution of artificial intelligence (AI) to medical imaging is currently the object of widespread experimentation. The development of deep learning (DL) methods, particularly convolution neural networks (CNNs), has led to performance gains often superior to those achieved by conventional methods such as machine learning. Radiomics is an approach aimed at extracting quantitative data not accessible to the human eye from images expressing a disease. The data subsequently feed machine learning models and produce diagnostic or prognostic probabilities. As for the multiple applications of AI methods in thoracic imaging, they are undergoing evaluation. Chest radiography is a practically ideal field for the development of DL algorithms able to automatically interpret X-rays. Current algorithms can detect up to 14 different abnormalities present either in isolation or in combination. Chest CT is another area offering numerous AI applications. Various algorithms have been specifically formed and validated for the detection and characterization of pulmonary nodules and pulmonary embolism, as well as segmentation and quantitative analysis of the extent of diffuse lung diseases (emphysema, infectious pneumonias, interstitial lung disease). In addition, the analysis of medical images can be associated with clinical, biological, and functional data (multi-omics analysis), the objective being to construct predictive approaches regarding disease prognosis and response to treatment.

Supervised training models with or without manual lesion delineation outperform clinicians in distinguishing pulmonary cryptococcosis from lung adenocarcinoma on chest CT

BACKGROUND

The role of artificial intelligence (AI) in the discrimination between pulmonary cryptococcosis (PC) and lung adenocarcinoma (LA) warrants further research.

OBJECTIVES

To compare the performances of AI models with clinicians in distinguishing PC from LA on chest CT.

METHODS

Patients diagnosed with confirmed PC or LA were retrospectively recruited from three tertiary hospitals in Guangzhou. A deep learning framework was employed to develop two models: an undelineated supervised training (UST) model utilising original CT images, and a delineated supervised training (DST) model utilising CT images with manual lesion annotations provided by physicians. A subset of 20 cases was randomly selected from the entire dataset and reviewed by clinicians through a network questionnaire. The sensitivity, specificity and accuracy of the models and the clinicians were calculated.

RESULTS

A total of 395 PC cases and 249 LA cases were included in the final analysis. The internal validation results for the UST model showed a sensitivity of 85.3%, specificity of 81.0%, accuracy of 83.6% and an area under the curve (AUC) of 0.93. Similarly, the DST model exhibited a sensitivity of 88.2%, specificity of 88.1%, accuracy of 88.2% and an AUC of 0.94. The external validation of the two models yielded AUC values of 0.74 and 0.77, respectively. The average sensitivity, specificity and accuracy of 102 clinicians were determined to be 63.1%, 53.7% and 59.3%, respectively.

CONCLUSIONS

Both models outperformed the clinicians in distinguishing between PC and LA on chest CT, with the UST model exhibiting comparable performance to the DST model.

ERS International Congress 2023: highlights from the Clinical Techniques, Imaging and Endoscopy Assembly

The Clinical Techniques, Imaging and Endoscopy Assembly is involved in the diagnosis and treatment of several pulmonary diseases, as demonstrated at the 2023 European Respiratory Society (ERS) International Congress in Milan, Italy. From interventional pulmonology, the congress included several exciting results for the use of bronchoscopy in lung cancer, including augmented fluoroscopy, robotic-assisted bronchoscopy and cryobiopsies. In obstructive lung disease, the latest results on bronchoscopic treatment of emphysema with hyperinflation and chronic bronchitis were presented. Research on using cryobiopsies to diagnose interstitial lung disease was further explored, with the aims of elevating diagnostic yield and minimising risk. For imaging, the latest updates in using artificial intelligence to overcome the increased workload of radiologists were of great interest. Novel imaging in sarcoidosis explored the use of magnetic resonance imaging, photon-counting computed tomography and positron emission tomography/computed tomography in the diagnostic work-up. Lung cancer screening is still a hot topic and new results were presented regarding incorporation of biomarkers, identifying knowledge gaps and improving screening programmes. The use of ultrasound in respiratory medicine is an expanding field, which was demonstrated by the large variety in studies presented at the 2023 ERS Congress. Ultrasound of the diaphragm in patients with amyotrophic lateral sclerosis and myasthenia gravis was used to assess movements and predict respiratory fatigue. Furthermore, studies using ultrasound to diagnose or monitor pulmonary disease were presented. The congress also included studies regarding the training and assessment of competencies as an important part of implementing ultrasound in clinical practice.

Deep learning algorithms for classification and detection of recurrent aphthous ulcerations using oral clinical photographic images

Background/purpose

The application of artificial intelligence diagnosis based on deep learning in the medical field has been widely accepted. We aimed to evaluate convolutional neural networks (CNNs) for automated classification and detection of recurrent aphthous ulcerations (RAU), normal oral mucosa, and other common oral mucosal diseases in clinical oral photographs.

Materials and methods

The study included 785 clinical oral photographs, which was divided into 251 images of RAU, 271 images of the normal oral mucosa, and 263 images of other common oral mucosal diseases. Four and three CNN models were used for the classification and detection tasks, respectively. 628 images were randomly selected as training data. In addition, 78 and 79 images were assigned as validating and testing data. Main outcome measures included precision, recall, F1, specificity, sensitivity and area under the receiver operating characteristics curve (AUC).

Results

In the classification task, the Pretrained ResNet50 model had the best performance with a precision of 92.86%, a recall of 91.84%, an F1 score of 92.24%, a specificity of 96.41%, a sensitivity of 91.84% and an AUC of 98.95%. In the detection task, the Pretrained YOLOV5 model had the best performance with a precision of 98.70%, a recall of 79.51%, an F1 score of 88.07% and an AUC of Precision-Recall curve 90.89%.

Conclusion

The Pretrained ResNet50 and the Pretrained YOLOV5 algorithms were shown to have superior performance and acceptable potential in the classification and detection of RAU lesions based on non-invasive oral images, which may prove useful in clinical practice.

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

Purpose

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

Materials and methods

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

Results

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

Conclusion

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

Novel research and future prospects of artificial intelligence in cancer diagnosis and treatment

Research into the potential benefits of artificial intelligence for comprehending the intricate biology of cancer has grown as a result of the widespread use of deep learning and machine learning in the healthcare sector and the availability of highly specialized cancer datasets. Here, we review new artificial intelligence approaches and how they are being used in oncology. We describe how artificial intelligence might be used in the detection, prognosis, and administration of cancer treatments and introduce the use of the latest large language models such as ChatGPT in oncology clinics. We highlight artificial intelligence applications for omics data types, and we offer perspectives on how the various data types might be combined to create decision-support tools. We also evaluate the present constraints and challenges to applying artificial intelligence in precision oncology. Finally, we discuss how current challenges may be surmounted to make artificial intelligence useful in clinical settings in the future.

Deep learning for the detection of benign and malignant pulmonary nodules in non-screening chest CT scans

BACKGROUND

Outside a screening program, early-stage lung cancer is generally diagnosed after the detection of incidental nodules in clinically ordered chest CT scans. Despite the advances in artificial intelligence (AI) systems for lung cancer detection, clinical validation of these systems is lacking in a non-screening setting.

METHOD

We developed a deep learning-based AI system and assessed its performance for the detection of actionable benign nodules (requiring follow-up), small lung cancers, and pulmonary metastases in CT scans acquired in two Dutch hospitals (internal and external validation). A panel of five thoracic radiologists labeled all nodules, and two additional radiologists verified the nodule malignancy status and searched for any missed cancers using data from the national Netherlands Cancer Registry. The detection performance was evaluated by measuring the sensitivity at predefined false positive rates on a free receiver operating characteristic curve and was compared with the panel of radiologists.

RESULTS

On the external test set (100 scans from 100 patients), the sensitivity of the AI system for detecting benign nodules, primary lung cancers, and metastases is respectively 94.3% (82/87, 95% CI: 88.1-98.8%), 96.9% (31/32, 95% CI: 91.7-100%), and 92.0% (104/113, 95% CI: 88.5-95.5%) at a clinically acceptable operating point of 1 false positive per scan (FP/s). These sensitivities are comparable to or higher than the radiologists, albeit with a slightly higher FP/s (average difference of 0.6).

CONCLUSIONS

The AI system reliably detects benign and malignant pulmonary nodules in clinically indicated CT scans and can potentially assist radiologists in this setting.

Expanding Role of Advanced Image Analysis in CT-detected Indeterminate Pulmonary Nodules and Early Lung Cancer Characterization

The implementation of low-dose chest CT for lung screening presents a crucial opportunity to advance lung cancer care through early detection and interception. In addition, millions of pulmonary nodules are incidentally detected annually in the United States, increasing the opportunity for early lung cancer diagnosis. Yet, realization of the full potential of these opportunities is dependent on the ability to accurately analyze image data for purposes of nodule classification and early lung cancer characterization. This review presents an overview of traditional image analysis approaches in chest CT using semantic characterization as well as more recent advances in the technology and application of machine learning models using CT-derived radiomic features and deep learning architectures to characterize lung nodules and early cancers. Methodological challenges currently faced in translating these decision aids to clinical practice, as well as the technical obstacles of heterogeneous imaging parameters, optimal feature selection, choice of model, and the need for well-annotated image data sets for the purposes of training and validation, will be reviewed, with a view toward the ultimate incorporation of these potentially powerful decision aids into routine clinical practice.

Artificial Intelligence in Oncological Hybrid Imaging

BACKGROUND

 Artificial intelligence (AI) applications have become increasingly relevant across a broad spectrum of settings in medical imaging. Due to the large amount of imaging data that is generated in oncological hybrid imaging, AI applications are desirable for lesion detection and characterization in primary staging, therapy monitoring, and recurrence detection. Given the rapid developments in machine learning (ML) and deep learning (DL) methods, the role of AI will have significant impact on the imaging workflow and will eventually improve clinical decision making and outcomes.

METHODS AND RESULTS

 The first part of this narrative review discusses current research with an introduction to artificial intelligence in oncological hybrid imaging and key concepts in data science. The second part reviews relevant examples with a focus on applications in oncology as well as discussion of challenges and current limitations.

CONCLUSION

 AI applications have the potential to leverage the diagnostic data stream with high efficiency and depth to facilitate automated lesion detection, characterization, and therapy monitoring to ultimately improve quality and efficiency throughout the medical imaging workflow. The goal is to generate reproducible, structured, quantitative diagnostic data for evidence-based therapy guidance in oncology. However, significant challenges remain regarding application development, benchmarking, and clinical implementation.

KEY POINTS

  · Hybrid imaging generates a large amount of multimodality medical imaging data with high complexity and depth.. · Advanced tools are required to enable fast and cost-efficient processing along the whole radiology value chain.. · AI applications promise to facilitate the assessment of oncological disease in hybrid imaging with high quality and efficiency for lesion detection, characterization, and response assessment. The goal is to generate reproducible, structured, quantitative diagnostic data for evidence-based oncological therapy guidance.. · Selected applications in three oncological entities (lung, prostate, and neuroendocrine tumors) demonstrate how AI algorithms may impact imaging-based tasks in hybrid imaging and potentially guide clinical decision making..

Performance of deep learning-based autodetection of arterial stenosis on head and neck CT angiography: an independent external validation study

PURPOSE

To externally validate the performance of automated stenosis detection on head and neck CT angiography (CTA) and investigate the impact factors using an independent bi-center dataset with digital subtraction angiography (DSA) as the ground truth.

MATERIAL AND METHODS

Patients who underwent head and neck CTA and DSA between January 2019 and December 2021 were retrospectively included. The degree of stenosis was automatically evaluated using CerebralDoc based on CTA. The performance of CerebralDoc across levels (per-patient, per-region, per-vessel, and per-segment) and thresholds (≥ 50%, ≥ 70%, and = 100%) was evaluated. Logistic regression was performed to identify independent factors associated with false negative results.

RESULTS

296 patients were analyzed. Specificity across levels and thresholds was high, exceeding 92%. The area under the curve ranged from poor (0.615, 95% CI: 0.544, 0.686; at the region-based analysis for stenosis ≥ 70%) to excellent (0.945, 95% CI: 0.905, 0.985; at the patient-based analysis for stenosis ≥ 50%). Sensitivity ranged from 0.714 (95% CI: 0.675, 0.750) at the segment-based analysis for stenosis ≥ 70% to 0.895 (95% CI: 0.849, 0.919) at the patient-based analysis for stenosis ≥ 50%. The multiple logistic regression analysis revealed that false negative results were primarily more likely to specific stenosis locations (particularly the M2 segment and skull base segment of the internal carotid artery) and occlusion.

CONCLUSIONS

CerebralDoc has the potential to automated stenosis detection on head and neck CTA, but further efforts are needed to optimize its performance.

Prediction Variability to Identify Reduced AI Performance in Cancer Diagnosis at MRI and CT

Background A priori identification of patients at risk of artificial intelligence (AI) failure in diagnosing cancer would contribute to the safer clinical integration of diagnostic algorithms. Purpose To evaluate AI prediction variability as an uncertainty quantification (UQ) metric for identifying cases at risk of AI failure in diagnosing cancer at MRI and CT across different cancer types, data sets, and algorithms. Materials and Methods Multicenter data sets and publicly available AI algorithms from three previous studies that evaluated detection of pancreatic cancer on contrast-enhanced CT images, detection of prostate cancer on MRI scans, and prediction of pulmonary nodule malignancy on low-dose CT images were analyzed retrospectively. Each task's algorithm was extended to generate an uncertainty score based on ensemble prediction variability. AI accuracy percentage and partial area under the receiver operating characteristic curve (pAUC) were compared between certain and uncertain patient groups in a range of percentile thresholds (10%-90%) for the uncertainty score using permutation tests for statistical significance. The pulmonary nodule malignancy prediction algorithm was compared with 11 clinical readers for the certain group (CG) and uncertain group (UG). Results In total, 18 022 images were used for training and 838 images were used for testing. AI diagnostic accuracy was higher for the cases in the CG across all tasks (P < .001). At an 80% threshold of certain predictions, accuracy in the CG was 21%-29% higher than in the UG and 4%-6% higher than in the overall test data sets. The lesion-level pAUC in the CG was 0.25-0.39 higher than in the UG and 0.05-0.08 higher than in the overall test data sets (P < .001). For pulmonary nodule malignancy prediction, accuracy of AI was on par with clinicians for cases in the CG (AI results vs clinician results, 80% [95% CI: 76, 85] vs 78% [95% CI: 70, 87]; P = .07) but worse for cases in the UG (AI results vs clinician results, 50% [95% CI: 37, 64] vs 68% [95% CI: 60, 76]; P < .001). Conclusion An AI-prediction UQ metric consistently identified reduced performance of AI in cancer diagnosis. © RSNA, 2023 Supplemental material is available for this article. See also the editorial by Babyn in this issue.

Novel tools for early diagnosis and precision treatment based on artificial intelligence

Lung cancer has the highest mortality rate among all cancers in the world. Hence, early diagnosis and personalized treatment plans are crucial to improving its 5-year survival rate. Chest computed tomography (CT) serves as an essential tool for lung cancer screening, and pathology images are the gold standard for lung cancer diagnosis. However, medical image evaluation relies on manual labor and suffers from missed diagnosis or misdiagnosis, and physician heterogeneity. The rapid development of artificial intelligence (AI) has brought a whole novel opportunity for medical task processing, demonstrating the potential for clinical application in lung cancer diagnosis and treatment. AI technologies, including machine learning and deep learning, have been deployed extensively for lung nodule detection, benign and malignant classification, and subtype identification based on CT images. Furthermore, AI plays a role in the non-invasive prediction of genetic mutations and molecular status to provide the optimal treatment regimen, and applies to the assessment of therapeutic efficacy and prognosis of lung cancer patients, enabling precision medicine to become a reality. Meanwhile, histology-based AI models assist pathologists in typing, molecular characterization, and prognosis prediction to enhance the efficiency of diagnosis and treatment. However, the leap to extensive clinical application still faces various challenges, such as data sharing, standardized label acquisition, clinical application regulation, and multimodal integration. Nevertheless, AI holds promising potential in the field of lung cancer to improve cancer care.

DNA-Guided Extracellular-Vesicle Metallization with High Catalytic Activity for Accurate Diagnosis of Pulmonary Nodules

Sensitive and specific analysis of extracellular vesicles (EVs) offers a promising minimally invasive way to identify malignant pulmonary nodules from benign lesions. However, accurate analysis of EVs is subject to free target proteins in blood samples, which compromises the clinical diagnosis value of EVs. Here a DNA-guided extracellular-vesicle metallization (DEVM) strategy is described for ultrasensitive and specific analysis of EV protein biomarkers and classification of pulmonary nodules. The facile DEVM process mainly includes the incorporation of DNA labeled with cholesterol and thiol groups into EV membranes and subsequent deposition of Au3+ and Pt4+ to allow the DNA-functionalized EVs to be encapsulated with AuPt nanoshells. It is found that the synthesized AuPt-metallized EVs possess extrinsic peroxidase-like activity. Utilizing the feature of the catalytic metal nanoshells just growth on the EV membranes, the DEVM method enables multiparametric recognition of target proteins and EV membranes and can produce an amplified colorimetric signal, avoiding the interference of free proteins. By profiling four surface proteins of EVs from 48 patients with pulmonary nodules, the highest area under the receiver operating characteristic curve (0.9983) is obtained. Therefore, this work provides a feasible EVs analysis tool for accurate pulmonary nodules management.

A diagnostic classification of lung nodules using multiple-scale residual network

Computed tomography (CT) scans have been shown to be an effective way of improving diagnostic efficacy and reducing lung cancer mortality. However, distinguishing benign from malignant nodules in CT imaging remains challenging. This study aims to develop a multiple-scale residual network (MResNet) to automatically and precisely extract the general feature of lung nodules, and classify lung nodules based on deep learning. The MResNet aggregates the advantages of residual units and pyramid pooling module (PPM) to learn key features and extract the general feature for lung nodule classification. Specially, the MResNet uses the ResNet as a backbone network to learn contextual information and discriminate feature representation. Meanwhile, the PPM is used to fuse features under four different scales, including the coarse scale and the fine-grained scale to obtain more general lung features of the CT image. MResNet had an accuracy of 99.12%, a sensitivity of 98.64%, a specificity of 97.87%, a positive predictive value (PPV) of 99.92%, and a negative predictive value (NPV) of 97.87% in the training set. Additionally, its area under the receiver operating characteristic curve (AUC) was 0.9998 (0.99976-0.99991). MResNet's accuracy, sensitivity, specificity, PPV, NPV, and AUC in the testing set were 85.23%, 92.79%, 72.89%, 84.56%, 86.34%, and 0.9275 (0.91662-0.93833), respectively. The developed MResNet performed exceptionally well in estimating the malignancy risk of pulmonary nodules found on CT. The model has the potential to provide reliable and reproducible malignancy risk scores for clinicians and radiologists, thereby optimizing lung cancer screening management.

An ensemble deep learning model for risk stratification of invasive lung adenocarcinoma using thin-slice CT

Lung cancer screening using computed tomography (CT) has increased the detection rate of small pulmonary nodules and early-stage lung adenocarcinoma. It would be clinically meaningful to accurate assessment of the nodule histology by CT scans with advanced deep learning algorithms. However, recent studies mainly focus on predicting benign and malignant nodules, lacking of model for the risk stratification of invasive adenocarcinoma. We propose an ensemble multi-view 3D convolutional neural network (EMV-3D-CNN) model to study the risk stratification of lung adenocarcinoma. We include 1075 lung nodules (≤30 mm and ≥4 mm) with preoperative thin-section CT scans and definite pathology confirmed by surgery. Our model achieves a state-of-art performance of 91.3% and 92.9% AUC for diagnosis of benign/malignant and pre-invasive/invasive nodules, respectively. Importantly, our model outperforms senior doctors in risk stratification of invasive adenocarcinoma with 77.6% accuracy [i.e., Grades 1, 2, 3]). It provides detailed predictive histological information for the surgical management of pulmonary nodules. Finally, for user-friendly access, the proposed model is implemented as a web-based system ( https://seeyourlung.com.cn ).

ERS International Congress 2022: highlights from the Thoracic Oncology Assembly

Thoracic malignancies are associated with a substantial public health burden. Lung cancer is the leading cause of cancer-related mortality worldwide, with significant impact on patients' quality of life. Following 2 years of virtual European Respiratory Society (ERS) Congresses due to the COVID-19 pandemic, the 2022 hybrid ERS Congress in Barcelona, Spain allowed peers from all over the world to meet again and present their work. Thoracic oncology experts presented best practices and latest developments in lung cancer screening, lung cancer diagnosis and management. Early lung cancer diagnosis, subsequent pros and cons of aggressive management, identification and management of systemic treatments' side-effects, and the application of artificial intelligence and biomarkers across all aspects of the thoracic oncology pathway were among the areas that triggered specific interest and will be summarised here.

Lung Cancer Screening with Low-Dose CT: What We Have Learned in Two Decades of ITALUNG and What Is Yet to Be Addressed

The ITALUNG trial started in 2004 and compared lung cancer (LC) and other-causes mortality in 55-69 years-aged smokers and ex-smokers who were randomized to four annual chest low-dose CT (LDCT) or usual care. ITALUNG showed a lower LC and cardiovascular mortality in the screened subjects after 13 years of follow-up, especially in women, and produced many ancillary studies. They included recruitment results of a population-based mimicking approach, development of software for computer-aided diagnosis (CAD) and lung nodules volumetry, LDCT assessment of pulmonary emphysema and coronary artery calcifications (CAC) and their relevance to long-term mortality, results of a smoking-cessation intervention, assessment of the radiations dose associated with screening LDCT, and the results of biomarkers assays. Moreover, ITALUNG data indicated that screen-detected LCs are mostly already present at baseline LDCT, can present as lung cancer associated with cystic airspaces, and can be multiple. However, several issues of LC screening are still unaddressed. They include the annual vs. biennial pace of LDCT, choice between opportunistic or population-based recruitment. and between uni or multi-centre screening, implementation of CAD-assisted reading, containment of false positive and negative LDCT results, incorporation of emphysema. and CAC quantification in models of personalized LC and mortality risk, validation of ultra-LDCT acquisitions, optimization of the smoking-cessation intervention. and prospective validation of the biomarkers.

Imaging of lung cancer

Lung cancer is the leading cause of cancer-related mortality globally. Imaging is essential in the screening, diagnosis, staging, response assessment, and surveillance of patients with lung cancer. Subtypes of lung cancer can have distinguishing imaging appearances. The most frequently used imaging modalities include chest radiography, computed tomography, magnetic resonance imaging, and positron emission tomography. Artificial intelligence algorithms and radiomics are emerging technologies with potential applications in lung cancer imaging.

Serum laser Raman spectroscopy as a potential diagnostic tool to discriminate the benignancy or malignancy of pulmonary nodules

It has been proved that Raman spectral intensities could be used to diagnose lung cancer patients. However, the application of Raman spectroscopy in identifying the patients with pulmonary nodules was barely studied. In this study, we revealed that Raman spectra of serum samples from healthy participants and patients with benign and malignant pulmonary nodules were significantly different. A support vector machine (SVM) model was developed for the classification of Raman spectra with wave points, according to ANOVA test results. It got a good performance with a median area under the curve (AUC) of 0.89, when the SVM model was applied in discriminating benign from malignant individuals. Compared with three common clinical models, the SVM model showed a better discriminative ability and added more net benefits to participants, which were also excellent in the small-size nodules. Thus, the Raman spectroscopy could be a less-invasive and low-costly liquid biopsy.

The Effects of Artificial Intelligence Assistance on the Radiologists' Assessment of Lung Nodules on CT Scans: A Systematic Review

To reduce the number of missed or misdiagnosed lung nodules on CT scans by radiologists, many Artificial Intelligence (AI) algorithms have been developed. Some algorithms are currently being implemented in clinical practice, but the question is whether radiologists and patients really benefit from the use of these novel tools. This study aimed to review how AI assistance for lung nodule assessment on CT scans affects the performances of radiologists. We searched for studies that evaluated radiologists' performances in the detection or malignancy prediction of lung nodules with and without AI assistance. Concerning detection, radiologists achieved with AI assistance a higher sensitivity and AUC, while the specificity was slightly lower. Concerning malignancy prediction, radiologists achieved with AI assistance generally a higher sensitivity, specificity and AUC. The radiologists' workflows of using the AI assistance were often only described in limited detail in the papers. As recent studies showed improved performances of radiologists with AI assistance, AI assistance for lung nodule assessment holds great promise. To achieve added value of AI tools for lung nodule assessment in clinical practice, more research is required on the clinical validation of AI tools, impact on follow-up recommendations and ways of using AI tools.

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

OBJECTIVE

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

METHODS

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

RESULTS

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

CONCLUSIONS

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

KEY POINTS

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

Artificial Intelligence in CT and MR Imaging for Oncological Applications

Cancer care increasingly relies on imaging for patient management. The two most common cross-sectional imaging modalities in oncology are computed tomography (CT) and magnetic resonance imaging (MRI), which provide high-resolution anatomic and physiological imaging. Herewith is a summary of recent applications of rapidly advancing artificial intelligence (AI) in CT and MRI oncological imaging that addresses the benefits and challenges of the resultant opportunities with examples. Major challenges remain, such as how best to integrate AI developments into clinical radiology practice, the vigorous assessment of quantitative CT and MR imaging data accuracy, and reliability for clinical utility and research integrity in oncology. Such challenges necessitate an evaluation of the robustness of imaging biomarkers to be included in AI developments, a culture of data sharing, and the cooperation of knowledgeable academics with vendor scientists and companies operating in radiology and oncology fields. Herein, we will illustrate a few challenges and solutions of these efforts using novel methods for synthesizing different contrast modality images, auto-segmentation, and image reconstruction with examples from lung CT as well as abdome, pelvis, and head and neck MRI. The imaging community must embrace the need for quantitative CT and MRI metrics beyond lesion size measurement. AI methods for the extraction and longitudinal tracking of imaging metrics from registered lesions and understanding the tumor environment will be invaluable for interpreting disease status and treatment efficacy. This is an exciting time to work together to move the imaging field forward with narrow AI-specific tasks. New AI developments using CT and MRI datasets will be used to improve the personalized management of cancer patients.

Deep learning-enabled fully automated pipeline system for segmentation and classification of single-mass breast lesions using contrast-enhanced mammography: a prospective, multicentre study

Background

Breast cancer is the leading cause of cancer-related deaths in women. However, accurate diagnosis of breast cancer using medical images heavily relies on the experience of radiologists. This study aimed to develop an artificial intelligence model that diagnosed single-mass breast lesions on contrast-enhanced mammography (CEM) for assisting the diagnostic workflow.

Methods

A total of 1912 women with single-mass breast lesions on CEM images before biopsy or surgery were included from June 2017 to October 2022 at three centres in China. Samples were divided into training and validation sets, internal testing set, pooled external testing set, and prospective testing set. A fully automated pipeline system (FAPS) using RefineNet and the Xception + Pyramid pooling module (PPM) was developed to perform the segmentation and classification of breast lesions. The performances of six radiologists and adjustments in Breast Imaging Reporting and Data System (BI-RADS) category 4 under the FAPS-assisted strategy were explored in pooled external and prospective testing sets. The segmentation performance was assessed using the Dice similarity coefficient (DSC), and the classification was assessed using heatmaps, area under the receiver operating characteristic curve (AUC), sensitivity, and specificity. The radiologists' reading time was recorded for comparison with the FAPS. This trial is registered with China Clinical Trial Registration Centre (ChiCTR2200063444).

Findings

The FAPS-based segmentation task achieved DSCs of 0.888 ± 0.101, 0.820 ± 0.148 and 0.837 ± 0.132 in the internal, pooled external and prospective testing sets, respectively. For the classification task, the FAPS achieved AUCs of 0.947 (95% confidence interval [CI]: 0.916-0.978), 0.940 (95% [CI]: 0.894-0.987) and 0.891 (95% [CI]: 0.816-0.945). It outperformed radiologists in terms of classification efficiency based on single lesions (6 s vs 3 min). Moreover, the FAPS-assisted strategy improved the performance of radiologists. BI-RADS category 4 in 12.4% and 13.3% of patients was adjusted in two testing sets with the assistance of FAPS, which may play an important guiding role in the selection of clinical management strategies.

Interpretation

The FAPS based on CEM demonstrated the potential for the segmentation and classification of breast lesions, and had good generalisation ability and clinical applicability.

Funding

This study was supported by the Taishan Scholar Foundation of Shandong Province of China (tsqn202211378), National Natural Science Foundation of China (82001775), Natural Science Foundation of Shandong Province of China (ZR2021MH120), and Special Fund for Breast Disease Research of Shandong Medical Association (YXH2021ZX055).

Lung cancer screening

Randomised controlled trials, including the National Lung Screening Trial (NLST) and the NELSON trial, have shown reduced mortality with lung cancer screening with low-dose CT compared with chest radiography or no screening. Although research has provided clarity on key issues of lung cancer screening, uncertainty remains about aspects that might be critical to optimise clinical effectiveness and cost-effectiveness. This Review brings together current evidence on lung cancer screening, including an overview of clinical trials, considerations regarding the identification of individuals who benefit from lung cancer screening, management of screen-detected findings, smoking cessation interventions, cost-effectiveness, the role of artificial intelligence and biomarkers, and current challenges, solutions, and opportunities surrounding the implementation of lung cancer screening programmes from an international perspective. Further research into risk models for patient selection, personalised screening intervals, novel biomarkers, integrated cardiovascular disease and chronic obstructive pulmonary disease assessments, smoking cessation interventions, and artificial intelligence for lung nodule detection and risk stratification are key opportunities to increase the efficiency of lung cancer screening and ensure equity of access.

Effects of Expert-Determined Reference Standards in Evaluating the Diagnostic Performance of a Deep Learning Model: A Malignant Lung Nodule Detection Task on Chest Radiographs

OBJECTIVE

Little is known about the effects of using different expert-determined reference standards when evaluating the performance of deep learning-based automatic detection (DLAD) models and their added value to radiologists. We assessed the concordance of expert-determined standards with a clinical gold standard (herein, pathological confirmation) and the effects of different expert-determined reference standards on the estimates of radiologists' diagnostic performance to detect malignant pulmonary nodules on chest radiographs with and without the assistance of a DLAD model.

MATERIALS AND METHODS

This study included chest radiographs from 50 patients with pathologically proven lung cancer and 50 controls. Five expert-determined standards were constructed using the interpretations of 10 experts: individual judgment by the most experienced expert, majority vote, consensus judgments of two and three experts, and a latent class analysis (LCA) model. In separate reader tests, additional 10 radiologists independently interpreted the radiographs and then assisted with the DLAD model. Their diagnostic performance was estimated using the clinical gold standard and various expert-determined standards as the reference standard, and the results were compared using the t test with Bonferroni correction.

RESULTS

The LCA model (sensitivity, 72.6%; specificity, 100%) was most similar to the clinical gold standard. When expert-determined standards were used, the sensitivities of radiologists and DLAD model alone were overestimated, and their specificities were underestimated (all p-values < 0.05). DLAD assistance diminished the overestimation of sensitivity but exaggerated the underestimation of specificity (all p-values < 0.001). The DLAD model improved sensitivity and specificity to a greater extent when using the clinical gold standard than when using the expert-determined standards (all p-values < 0.001), except for sensitivity with the LCA model (p = 0.094).

CONCLUSION

The LCA model was most similar to the clinical gold standard for malignant pulmonary nodule detection on chest radiographs. Expert-determined standards caused bias in measuring the diagnostic performance of the artificial intelligence model.

Clinical Impact and Generalizability of a Computer-Assisted Diagnostic Tool to Risk-Stratify Lung Nodules With CT

OBJECTIVE

To evaluate whether an imaging classifier for radiology practice can improve lung nodule classification and follow-up.

METHODS

A machine learning classifier was developed and trained using imaging data from the National Lung Screening Trial (NSLT) to produce a malignancy risk score (malignancy Similarity Index [mSI]) for individual lung nodules. In addition to NLST cohorts, external cohorts were developed from a tertiary referral lung cancer screening program data set and an external nonscreening data set of all nodules detected on CT. Performance of the mSI combined with Lung-RADS was compared with Lung-RADS alone and the Mayo and Brock risk calculators.

RESULTS

We analyzed 963 subjects and 1,331 nodules across these cohorts. The mSI was comparable in accuracy (area under the curve = 0.89) to existing clinical risk models (area under the curve = 0.86-0.88) and independently predictive in the NLST cohort of 704 nodules. When compared with Lung-RADS, the mSI significantly increased sensitivity across all cohorts (25%-117%), with significant increases in specificity in the screening cohorts (17%-33%). When used in conjunction with Lung-RADS, use of mSI would result in earlier diagnoses and reduced follow-up across cohorts, including the potential for early diagnosis in 42% of malignant NLST nodules from prior-year CT scans.

CONCLUSION

A computer-assisted diagnosis software improved risk classification from chest CTs of screening and incidentally detected lung nodules compared with Lung-RADS. mSI added predictive value independent of existing radiological and clinical variables. These results suggest the generalizability and potential clinical impact of a tool that is straightforward to implement in practice.

Incidental findings in medical imaging

The success of medical imaging as a diagnostic tool has resulted in a continuing increase in its use. Technological advances mean that images are now acquired at higher resolution and in greater volumes than ever before. This has led to an increase in the detection of findings which do not appear to be related to the primary purpose of the examination and have been termed "incidental". Many of these will be harmless but some will carry significant implications for the patient's health. Determining which of these findings are significant and which may be safely disregarded is an increasing problem in radiology practice. Radiologists should familiarise themselves with the more common incidental findings in order to make the best possible estimation of their importance in each case and to allow them to make appropriate recommendations for further investigation where this is indicated. The decision to advise further investigation carries implications for the patient and the service as a whole and requires careful consideration.

Artificial Intelligence in Oncological Hybrid Imaging

BACKGROUND

Artificial intelligence (AI) applications have become increasingly relevant across a broad spectrum of settings in medical imaging. Due to the large amount of imaging data that is generated in oncological hybrid imaging, AI applications are desirable for lesion detection and characterization in primary staging, therapy monitoring, and recurrence detection. Given the rapid developments in machine learning (ML) and deep learning (DL) methods, the role of AI will have significant impact on the imaging workflow and will eventually improve clinical decision making and outcomes.

METHODS AND RESULTS

The first part of this narrative review discusses current research with an introduction to artificial intelligence in oncological hybrid imaging and key concepts in data science. The second part reviews relevant examples with a focus on applications in oncology as well as discussion of challenges and current limitations.

CONCLUSION

AI applications have the potential to leverage the diagnostic data stream with high efficiency and depth to facilitate automated lesion detection, characterization, and therapy monitoring to ultimately improve quality and efficiency throughout the medical imaging workflow. The goal is to generate reproducible, structured, quantitative diagnostic data for evidence-based therapy guidance in oncology. However, significant challenges remain regarding application development, benchmarking, and clinical implementation.

KEY POINTS

· Hybrid imaging generates a large amount of multimodality medical imaging data with high complexity and depth.. · Advanced tools are required to enable fast and cost-efficient processing along the whole radiology value chain.. · AI applications promise to facilitate the assessment of oncological disease in hybrid imaging with high quality and efficiency for lesion detection, characterization, and response assessment. The goal is to generate reproducible, structured, quantitative diagnostic data for evidence-based oncological therapy guidance.. · Selected applications in three oncological entities (lung, prostate, and neuroendocrine tumors) demonstrate how AI algorithms may impact imaging-based tasks in hybrid imaging and potentially guide clinical decision making..

CITATION FORMAT

· Feuerecker B, Heimer M, Geyer T et al. Artificial Intelligence in Oncological Hybrid Imaging. Fortschr Röntgenstr 2023; 195: 105 - 114.

Diagnostic test accuracy of artificial intelligence-based imaging for lung cancer screening: A systematic review and meta-analysis

BACKGROUND

Lung cancer is the principal cause of cancer-related deaths worldwide. Early detection of lung cancer with screening is indispensable to reduce the high morbidity and mortality rates. Artificial intelligence (AI) is widely utilised in healthcare, including in the assessment of medical images. A growing number of reviews studied the application of AI in lung cancer screening, but no overarching meta-analysis has examined the diagnostic test accuracy (DTA) of AI-based imaging for lung cancer screening.

OBJECTIVE

To systematically review the DTA of AI-based imaging for lung cancer screening.

METHODS

PubMed, EMBASE, Cochrane Library, CINAHL, IEEE Xplore, Web of Science, ACM Digital Library, Scopus, PsycINFO, and ProQuest Dissertations and Theses were searched from inception to date. Studies that were published in English and that evaluated the performance of AI-based imaging for lung cancer screening were included. Two independent reviewers screened titles and abstracts and used the Quality Assessment of Diagnostic Accuracy Studies-2 tool to appraise the quality of selected studies. Grading of Recommendations Assessment, Development, and Evaluation to diagnostic tests was used to assess the certainty of evidence.

RESULTS

Twenty-six studies with 150,721 imaging data were included. Hierarchical summary receiver-operating characteristic model used for meta-analysis demonstrated that the pooled sensitivity for AI-based imaging for lung cancer screening was 94.6 % (95 % CI: 91.4 % to 96.7 %) and specificity was 93.6 % (95 % CI: 88.5 % to 96.6 %). Subgroup analyses revealed that similar results were found among different types of AI, region, data source, and year of publication, but the overall quality of evidence was very low.

CONCLUSION

AI-based imaging could effectively detect lung cancer and be incorporated into lung cancer screening programs. Further high-quality DTA studies on large lung cancer screening populations are required to validate AI's role in early lung cancer detection.

Deep Learning to Classify AL versus ATTR Cardiac Amyloidosis MR Images

The aim of this work was to compare the classification of cardiac MR-images of AL versus ATTR amyloidosis by neural networks and by experienced human readers. Cine-MR images and late gadolinium enhancement (LGE) images of 120 patients were studied (70 AL and 50 TTR). A VGG16 convolutional neural network (CNN) was trained with a 5-fold cross validation process, taking care to strictly distribute images of a given patient in either the training group or the test group. The analysis was performed at the patient level by averaging the predictions obtained for each image. The classification accuracy obtained between AL and ATTR amyloidosis was 0.750 for cine-CNN, 0.611 for Gado-CNN and between 0.617 and 0.675 for human readers. The corresponding AUC of the ROC curve was 0.839 for cine-CNN, 0.679 for gado-CNN (p < 0.004 vs. cine) and 0.714 for the best human reader (p < 0.007 vs. cine). Logistic regression with cine-CNN and gado-CNN, as well as analysis focused on the specific orientation plane, did not change the overall results. We conclude that cine-CNN leads to significantly better discrimination between AL and ATTR amyloidosis as compared to gado-CNN or human readers, but with lower performance than reported in studies where visual diagnosis is easy, and is currently suboptimal for clinical practice.

Application of artificial intelligence to the public health education

With the global outbreak of coronavirus disease 2019 (COVID-19), public health has received unprecedented attention. The cultivation of emergency and compound professionals is the general trend through public health education. However, current public health education is limited to traditional teaching models that struggle to balance theory and practice. Fortunately, the development of artificial intelligence (AI) has entered the stage of intelligent cognition. The introduction of AI in education has opened a new era of computer-assisted education, which brought new possibilities for teaching and learning in public health education. AI-based on big data not only provides abundant resources for public health research and management but also brings convenience for students to obtain public health data and information, which is conducive to the construction of introductory professional courses for students. In this review, we elaborated on the current status and limitations of public health education, summarized the application of AI in public health practice, and further proposed a framework for how to integrate AI into public health education curriculum. With the rapid technological advancements, we believe that AI will revolutionize the education paradigm of public health and help respond to public health emergencies.