CT venography for the diagnosis of postpartum venous thromboembolism: a prospective multi-center cohort study.

OBJECTIVES: To assess the role of CT venography (CTV) in the diagnosis of venous thromboembolism (VTE) during the postpartum period. MATERIALS AND METHODS: This multicenter prospective cohort study was conducted between April 2016 and April 2020 in 14 university hospitals. All women referred for CT pulmonary angiography (CTPA) for suspected pulmonary embolism (PE) within the first 6 weeks postpartum were eligible. All CTPAs were performed on multidetector CT machines with the usual parameters and followed by CTV of the abdomen, pelvis, and proximal lower limbs. On-site reports were compared to expert consensus reading, and the added value of CTV was assessed for both. RESULTS: The final study population consisted of 123 women. On-site CTPA reports mentioned PE in seven women (7/123, 5.7%), all confirmed following expert consensus reading, three involving proximal pulmonary arteries and four limited to distal arteries. Positive CTV was reported on-site in nine women, five of whom had negative and two indeterminate CTPAs, bringing the VTE detection rate to 11.4% (14/123) (95%CI: 6.4-18.4, p = 0.03). Expert consensus reading confirmed all positive on-site CTV results, but detected a periuterine vein thrombosis in an additional woman who had a negative CTPA, increasing the VTE detection rate to 12.2% (15/123) (95%CI: 7.0-19.3, p = 0.008). Follow-up at 3 months revealed no adverse events in this woman, who was left untreated. Median Dose-Length-Product was 117 mGy.cm for CTPA and 675 mGy.cm for CTPA + CTV. CONCLUSION: Performing CTV in women suspected of postpartum PE doubles the detection of venous thromboembolism, at the cost of increased radiation exposure. CLINICAL RELEVANCE STATEMENT: CTV can help in the decision-making process concerning curative anticoagulation in women with suspected postpartum PE, particularly those whose CTPA results are indeterminate or whose PE is limited to the subsegmental level. KEY POINTS: Postpartum women are at risk of pulmonary embolism, and CT pulmonary angiography can give equivocal results. CT venography (CTV) positivity increased the venous thromboembolism detection rate from 5.7 to 11.4%. CTV may help clinical decision-making, especially in women with indeterminate CTPA results or subsegmental emboli.

Anti-U1RNP antibodies are associated with a distinct clinical phenotype and a worse survival in patients with systemic sclerosis.

OBJECTIVES: To clarify the impact of anti-U1RNP antibodies on the clinical features and prognosis of patients with SSc. METHODS: We conducted a monocentric case-control, retrospective, longitudinal study. For each patient with SSc and anti-U1RNP antibodies (SSc-RNP+), one patient with mixed connective tissue disease (MCTD) and 2 SSc patients without anti-U1RNP antibodies (SSc-RNP-) were matched for age, sex, and date of inclusion. RESULTS: Sixty-four SSc-RNP+ patients were compared to 128 SSc-RNP- and 64 MCTD patients. Compared to SSc-RNP-, SSc-RNP+ patients were more often of Afro-Caribbean origin (31.3% vs. 11%, p < 0.01), and more often had an overlap syndrome than SSc-RNP- patients (53.1 % vs. 22.7%, p < 0.0001), overlapping with Sjögren's syndrome (n = 23, 35.9%) and/or systemic lupus erythematosus (n = 19, 29.7%). SSc-RNP+ patients were distinctly different from MCTD patients but less often had joint involvement (p < 0.01). SSc-RNP+ patients more frequently developed interstitial lung disease (ILD) (73.4% vs. 55.5% vs. 31.3%, p < 0.05), pulmonary fibrosis (PF) (60.9% vs. 37.5% vs. 10.9%, p < 0.0001), SSc associated myopathy (29.7% vs. 6.3% vs. 7.8%, p < 0.0001), and kidney involvement (10.9% vs. 2.3% vs. 1.6%, p < 0.05). Over a 200-month follow-up period, SSc-RNP+ patients had worse overall survival (p < 0.05), worse survival without PF occurrence (p < 0.01), ILD or PF progression (p < 0.01 and p < 0.0001). CONCLUSIONS: In SSc patients, anti-U1RNP antibodies are associated with a higher incidence of overlap syndrome, a distinct clinical phenotype, and poorer survival compared to SSc-RNP- and MCTD patients. Our study suggests that SSc-RNP+ patients should be separated from MCTD patients and may constitute an enriched population for progressive lung disease.

Reversal of cylindrical bronchial dilatations in a subset of adults with cystic fibrosis treated with elexacaftor-tezacaftor-ivacaftor.

BACKGROUND: This study sought to evaluate the impact of elexacaftor-tezacaftor-ivacaftor (ETI) on lung structural abnormalities in adults with cystic fibrosis (awCF) with a specific focus on the reversal of bronchial dilatations. METHODS: Chest computed tomography (CT) performed prior to, and ≥12 months after initiation of ETI were visually reviewed for possible reversal of bronchial dilatations. AwCF with and without reversal of bronchial dilatation (the latter served as controls with 3 controls per case) were selected. Visual Brody score, bronchial and arterial diameters, and lung volume were measured on CT. RESULTS: Reversal of bronchial dilatation was found in 12/235 (5%) awCF treated with ETI. Twelve awCF with and 36 without reversal of bronchial dilatations were further analyzed (male=56%, mean age=31.6±8.5 years, F508del/F508del CFTR =54% and mean %predicted forced expiratory volume in 1 s=58.8%±22.3). The mean±sd Brody score improved overall from 79.4±29.8 to 54.8±32.3 (p<0.001). Reversal of bronchial dilatations was confirmed by a decrease in bronchial lumen diameter in cases from 3.9±0.9 mm to 3.2±1.1 mm (p<0.001), whereas it increased in awCF without reversal of bronchial dilatation (from 3.5±1.1 mm to 3.6±1.2 mm, p=0.002). Reversal of bronchial dilatations occurred in cylindrical (not varicose or saccular) bronchial dilatations. Lung volumes decreased by -6.6±10.7% in awCF with reversal of bronchial dilatation but increased by +2.3±9.6% in controls (p=0.007). CONCLUSION: Although bronchial dilatations are generally considered irreversible, ETI was associated with reversal, which was limited to the cylindrical bronchial dilatations subtype, and occurred in a small subset of awCF. Initiating ETI earlier in life may reverse early bronchial dilatations.

Detection and severity quantification of pulmonary embolism with 3D CT data using an automated deep learning-based artificial solution.

PURPOSE: The purpose of this study was to propose a deep learning-based approach to detect pulmonary embolism and quantify its severity using the Qanadli score and the right-to-left ventricle diameter (RV/LV) ratio on three-dimensional (3D) computed tomography pulmonary angiography (CTPA) examinations with limited annotations. MATERIALS AND METHODS: Using a database of 3D CTPA examinations of 1268 patients with image-level annotations, and two other public datasets of CTPA examinations from 91 (CAD-PE) and 35 (FUME-PE) patients with pixel-level annotations, a pipeline consisting of: (i), detecting blood clots; (ii), performing PE-positive versus negative classification; (iii), estimating the Qanadli score; and (iv), predicting RV/LV diameter ratio was followed. The method was evaluated on a test set including 378 patients. The performance of PE classification and severity quantification was quantitatively assessed using an area under the curve (AUC) analysis for PE classification and a coefficient of determination (R²) for the Qanadli score and the RV/LV diameter ratio. RESULTS: Quantitative evaluation led to an overall AUC of 0.870 (95% confidence interval [CI]: 0.850-0.900) for PE classification task on the training set and an AUC of 0.852 (95% CI: 0.810-0.890) on the test set. Regression analysis yielded R² value of 0.717 (95% CI: 0.668-0.760) and of 0.723 (95% CI: 0.668-0.766) for the Qanadli score and the RV/LV diameter ratio estimation, respectively on the test set. CONCLUSION: This study shows the feasibility of utilizing AI-based assistance tools in detecting blood clots and estimating PE severity scores with 3D CTPA examinations. This is achieved by leveraging blood clots and cardiac segmentations. Further studies are needed to assess the effectiveness of these tools in clinical practice.

Lung fibrosis is uncommon in primary Sjögren's disease: A retrospective analysis of computed tomography features in 77 patients.

PURPOSE: The purpose of this study was to describe lung abnormalities observed on computed tomography (CT) in patients meeting the 2016 American College of Rheumatology/European League Against Rheumatism (EULAR) classification criteria for primary Sjögren's disease (pSD). MATERIALS AND METHODS: All patients with pSD seen between January 2009 and December 2020 in the day care centre of our National Reference Center for rare systemic autoimmune diseases, who had at least one chest CT examination available for review and for whom the cumulative EULAR Sjögren's Syndrome Disease Activity Index (cumESSDAI) could be calculated were retrospectively evaluated. CT examinations were reviewed, together with clinical symptoms and pulmonary functional results. RESULTS: Seventy-seven patients (73 women, four men) with a median age of 51 years at pSD diagnosis (age range: 17-79 years), a median follow-up time of 6 years and a median cumESSDAI of 7 were included. Sixty-six patients (86%) had anti-SSA antibodies. Thirty-three patients (33/77; 43%) had respiratory symptoms, without significant alteration in pulmonary function tests. Forty patients (40/77; 52%) had abnormal lung CT findings of whom almost half of them had no respiratory symptoms. Abnormalities on chest CT were more frequently observed in patients with anti-SSA positivity and a history of lymphoma. Air cysts (28/77; 36%) and mosaic perfusion (35/77; 35%) were the predominant abnormalities, whereas lung fibrosis was observed in five patients (5/77; 6%). CONCLUSION: More than half of patients with pSD have abnormal CT findings, mainly air cysts and mosaic perfusion, indicative of small airways disease, whereas lung fibrosis is rare, observed in less than 10% of such patients.

Prognostic value of automated assessment of interstitial lung disease on CT in systemic sclerosis.

OBJECTIVE: Stratifying the risk of death in SSc-related interstitial lung disease (SSc-ILD) is a challenging issue. The extent of lung fibrosis on high-resolution CT (HRCT) is often assessed by a visual semiquantitative method that lacks reliability. We aimed to assess the potential prognostic value of a deep-learning-based algorithm enabling automated quantification of ILD on HRCT in patients with SSc. METHODS: We correlated the extent of ILD with the occurrence of death during follow-up, and evaluated the additional value of ILD extent in predicting death based on a prognostic model including well-known risk factors in SSc. RESULTS: We included 318 patients with SSc, among whom 196 had ILD; the median follow-up was 94 months (interquartile range 73-111). The mortality rate was 1.6% at 2 years and 26.3% at 10 years. For each 1% increase in the baseline ILD extent (up to 30% of the lung), the risk of death at 10 years was increased by 4% (hazard ratio 1.04, 95% CI 1.01, 1.07, P = 0.004). We constructed a risk prediction model that showed good discrimination for 10-year mortality (c index 0.789). Adding the automated quantification of ILD significantly improved the model for 10-year survival prediction (P = 0.007). Its discrimination was only marginally improved, but it improved prediction of 2-year mortality (difference in time-dependent area under the curve 0.043, 95% CI 0.002, 0.084, P = 0.040). CONCLUSION: The deep-learning-based, computer-aided quantification of ILD extent on HRCT provides an effective tool for risk stratification in SSc. It might help identify patients at short-term risk of death.

Using AI to Improve Radiologist Performance in Detection of Abnormalities on Chest Radiographs.

Background Chest radiography remains the most common radiologic examination, and interpretation of its results can be difficult. Purpose To explore the potential benefit of artificial intelligence (AI) assistance in the detection of thoracic abnormalities on chest radiographs by evaluating the performance of radiologists with different levels of expertise, with and without AI assistance. Materials and Methods Patients who underwent both chest radiography and thoracic CT within 72 hours between January 2010 and December 2020 in a French public hospital were screened retrospectively. Radiographs were randomly included until reaching 500 radiographs, with about 50% of radiographs having abnormal findings. A senior thoracic radiologist annotated the radiographs for five abnormalities (pneumothorax, pleural effusion, consolidation, mediastinal and hilar mass, lung nodule) based on the corresponding CT results (ground truth). A total of 12 readers (four thoracic radiologists, four general radiologists, four radiology residents) read half the radiographs without AI and half the radiographs with AI (ChestView; Gleamer). Changes in sensitivity and specificity were measured using paired t tests. Results The study included 500 patients (mean age, 54 years ± 19 [SD]; 261 female, 239 male), with 522 abnormalities visible on 241 radiographs. On average, for all readers, AI use resulted in an absolute increase in sensitivity of 26% (95% CI: 20, 32), 14% (95% CI: 11, 17), 12% (95% CI: 10, 14), 8.5% (95% CI: 6, 11), and 5.9% (95% CI: 4, 8) for pneumothorax, consolidation, nodule, pleural effusion, and mediastinal and hilar mass, respectively (P < .001). Specificity increased with AI assistance (3.9% [95% CI: 3.2, 4.6], 3.7% [95% CI: 3, 4.4], 2.9% [95% CI: 2.3, 3.5], and 2.1% [95% CI: 1.6, 2.6] for pleural effusion, mediastinal and hilar mass, consolidation, and nodule, respectively), except in the diagnosis of pneumothorax (-0.2%; 95% CI: -0.36, -0.04; P = .01). The mean reading time was 81 seconds without AI versus 56 seconds with AI (31% decrease, P < .001). Conclusion AI-assisted chest radiography interpretation resulted in absolute increases in sensitivity for all radiologists of various levels of expertise and reduced the reading times; specificity increased with AI, except in the diagnosis of pneumothorax. © RSNA, 2023 Supplemental material is available for this article.

Ten reasons to screen women at risk of lung cancer.

This opinion piece reviews major reasons for promoting lung cancer screening in at-risk women who are smokers or ex-smokers, from the age of 50. The epidemiology of lung cancer in European women is extremely worrying, with lung cancer mortality expected to surpass breast cancer mortality in most European countries. There are conflicting data as to whether women are at increased risk of developing lung cancer compared to men who have a similar tobacco exposure. The sharp increase in the incidence of lung cancer in women exceeds the increase in their smoking exposure which is in favor of greater susceptibility. Lung and breast cancer screening could be carried out simultaneously, as the screening ages largely coincide. In addition, lung cancer screening could be carried out every 2 years, as is the case for breast cancer screening, if the baseline CT scan is negative.As well as detecting early curable lung cancer, screening can also detect coronary heart disease and osteoporosis induced by smoking. This enables preventive measures to be taken in addition to smoking cessation assistance, to reduce morbidity and mortality in the female population. Key points • The epidemiology of lung cancer in European women is very worrying.• Lung cancer is becoming the leading cause of cancer mortality in European women.• Women benefit greatly from screening in terms of reduced risk of death from lung cancer.

Comparative performances of machine learning algorithms in radiomics and impacting factors.

There are no current recommendations on which machine learning (ML) algorithms should be used in radiomics. The objective was to compare performances of ML algorithms in radiomics when applied to different clinical questions to determine whether some strategies could give the best and most stable performances regardless of datasets. This study compares the performances of nine feature selection algorithms combined with fourteen binary classification algorithms on ten datasets. These datasets included radiomics features and clinical diagnosis for binary clinical classifications including COVID-19 pneumonia or sarcopenia on CT, head and neck, orbital or uterine lesions on MRI. For each dataset, a train-test split was created. Each of the 126 (9 × 14) combinations of feature selection algorithms and classification algorithms was trained and tuned using a ten-fold cross validation, then AUC was computed. This procedure was repeated three times per dataset. Best overall performances were obtained with JMI and JMIM as feature selection algorithms and random forest and linear regression models as classification algorithms. The choice of the classification algorithm was the factor explaining most of the performance variation (10% of total variance). The choice of the feature selection algorithm explained only 2% of variation, while the train-test split explained 9%.

Learning from the machine: AI assistance is not an effective learning tool for resident education in chest x-ray interpretation.

OBJECTIVES: To assess whether a computer-aided detection (CADe) system could serve as a learning tool for radiology residents in chest X-ray (CXR) interpretation. METHODS: Eight radiology residents were asked to interpret 500 CXRs for the detection of five abnormalities, namely pneumothorax, pleural effusion, alveolar syndrome, lung nodule, and mediastinal mass. After interpreting 150 CXRs, the residents were divided into 2 groups of equivalent performance and experience. Subsequently, group 1 interpreted 200 CXRs from the "intervention dataset" using a CADe as a second reader, while group 2 served as a control by interpreting the same CXRs without the use of CADe. Finally, the 2 groups interpreted another 150 CXRs without the use of CADe. The sensitivity, specificity, and accuracy before, during, and after the intervention were compared. RESULTS: Before the intervention, the median individual sensitivity, specificity, and accuracy of the eight radiology residents were 43% (range: 35-57%), 90% (range: 82-96%), and 81% (range: 76-84%), respectively. With the use of CADe, residents from group 1 had a significantly higher overall sensitivity (53% [n = 431/816] vs 43% [n = 349/816], p < 0.001), specificity (94% [i = 3206/3428] vs 90% [n = 3127/3477], p < 0.001), and accuracy (86% [n = 3637/4244] vs 81% [n = 3476/4293], p < 0.001), compared to the control group. After the intervention, there were no significant differences between group 1 and group 2 regarding the overall sensitivity (44% [n = 309/696] vs 46% [n = 317/696], p = 0.666), specificity (90% [n = 2294/2541] vs 90% [n = 2285/2542], p = 0.642), or accuracy (80% [n = 2603/3237] vs 80% [n = 2602/3238], p = 0.955). CONCLUSIONS: Although it improves radiology residents' performances for interpreting CXRs, a CADe system alone did not appear to be an effective learning tool and should not replace teaching. CLINICAL RELEVANCE STATEMENT: Although the use of artificial intelligence improves radiology residents' performance in chest X-rays interpretation, artificial intelligence cannot be used alone as a learning tool and should not replace dedicated teaching. KEY POINTS: • With CADe as a second reader, residents had a significantly higher sensitivity (53% vs 43%, p < 0.001), specificity (94% vs 90%, p < 0.001), and accuracy (86% vs 81%, p < 0.001), compared to residents without CADe. • After removing access to the CADe system, residents' sensitivity (44% vs 46%, p = 0.666), specificity (90% vs 90%, p = 0.642), and accuracy (80% vs 80%, p = 0.955) returned to that of the level for the group without CADe.

Detection and quantification of pulmonary embolism with artificial intelligence: The SFR 2022 artificial intelligence data challenge.

PURPOSE: In 2022, the French Society of Radiology together with the French Society of Thoracic Imaging and CentraleSupelec organized their 13th data challenge. The aim was to aid in the diagnosis of pulmonary embolism, by identifying the presence of pulmonary embolism and by estimating the ratio between right and left ventricular (RV/LV) diameters, and an arterial obstruction index (Qanadli's score) using artificial intelligence. MATERIALS AND METHODS: The data challenge was composed of three tasks: the detection of pulmonary embolism, the RV/LV diameter ratio, and Qanadli's score. Sixteen centers all over France participated in the inclusion of the cases. A health data hosting certified web platform was established to facilitate the inclusion process of the anonymized CT examinations in compliance with general data protection regulation. CT pulmonary angiography images were collected. Each center provided the CT examinations with their annotations. A randomization process was established to pool the scans from different centers. Each team was required to have at least a radiologist, a data scientist, and an engineer. Data were provided in three batches to the teams, two for training and one for evaluation. The evaluation of the results was determined to rank the participants on the three tasks. RESULTS: A total of 1268 CT examinations were collected from the 16 centers following the inclusion criteria. The dataset was split into three batches of 310, 580 and 378 C T examinations provided to the participants respectively on September 5, 2022, October 7, 2022 and October 9, 2022. Seventy percent of the data from each center were used for training, and 30% for the evaluation. Seven teams with a total of 48 participants including data scientists, researchers, radiologists and engineering students were registered for participation. The metrics chosen for evaluation included areas under receiver operating characteristic curves, specificity and sensitivity for the classification task, and the coefficient of determination r2 for the regression tasks. The winning team achieved an overall score of 0.784. CONCLUSION: This multicenter study suggests that the use of artificial intelligence for the diagnosis of pulmonary embolism is possible on real data. Moreover, providing quantitative measures is mandatory for the interpretability of the results, and is of great aid to the radiologists especially in emergency settings.

The association between air pollution and the severity at diagnosis and progression of systemic sclerosis-associated interstitial lung disease: results from the retrospective ScleroPol study.

OBJECTIVE: To investigate the association of air pollution exposure with the severity of interstitial lung disease (ILD) at diagnosis and ILD progression among patients with systemic sclerosis (SSc)-associated ILD. METHODS: We conducted a retrospective two-center study of patients with SSc-associated ILD diagnosed between 2006 and 2019. Exposure to the air pollutants particulate matter of up to 10 and 2.5 µm in diameter (PM10, PM2.5), nitrogen dioxide (NO2), and ozone (O3) was assessed at the geolocalization coordinates of the patients' residential address. Logistic regression models were used to evaluate the association between air pollution and severity at diagnosis according to the Goh staging algorithm, and progression at 12 and 24 months. RESULTS: We included 181 patients, 80% of whom were women; 44% had diffuse cutaneous scleroderma, and 56% had anti-topoisomerase I antibodies. ILD was extensive, according to the Goh staging algorithm, in 29% of patients. O3 exposure was associated with the presence of extensive ILD at diagnosis (adjusted OR: 1.12, 95% CI 1.05-1.21; p value = 0.002). At 12 and 24 months, progression was noted in 27/105 (26%) and 48/113 (43%) patients, respectively. O3 exposure was associated with progression at 24 months (adjusted OR: 1.10, 95% CI 1.02-1.19; p value = 0.02). We found no association between exposure to other air pollutants and severity at diagnosis and progression. CONCLUSION: Our findings suggest that high levels of O3 exposure are associated with more severe SSc-associated ILD at diagnosis, and progression at 24 months.

Strategies to safely rule out pulmonary embolism in COVID-19 outpatients: a multicenter retrospective study.

OBJECTIVES: The objective was to define a safe strategy to exclude pulmonary embolism (PE) in COVID-19 outpatients, without performing CT pulmonary angiogram (CTPA). METHODS: COVID-19 outpatients from 15 university hospitals who underwent a CTPA were retrospectively evaluated. D-Dimers, variables of the revised Geneva and Wells scores, as well as laboratory findings and clinical characteristics related to COVID-19 pneumonia, were collected. CTPA reports were reviewed for the presence of PE and the extent of COVID-19 disease. PE rule-out strategies were based solely on D-Dimer tests using different thresholds, the revised Geneva and Wells scores, and a COVID-19 PE prediction model built on our dataset were compared. The area under the receiver operating characteristics curve (AUC), failure rate, and efficiency were calculated. RESULTS: In total, 1369 patients were included of whom 124 were PE positive (9.1%). Failure rate and efficiency of D-Dimer > 500 µg/l were 0.9% (95%CI, 0.2-4.8%) and 10.1% (8.5-11.9%), respectively, increasing to 1.0% (0.2-5.3%) and 16.4% (14.4-18.7%), respectively, for an age-adjusted D-Dimer level. D-dimer > 1000 µg/l led to an unacceptable failure rate to 8.1% (4.4-14.5%). The best performances of the revised Geneva and Wells scores were obtained using the age-adjusted D-Dimer level. They had the same failure rate of 1.0% (0.2-5.3%) for efficiency of 16.8% (14.7-19.1%), and 16.9% (14.8-19.2%) respectively. The developed COVID-19 PE prediction model had an AUC of 0.609 (0.594-0.623) with an efficiency of 20.5% (18.4-22.8%) when its failure was set to 0.8%. CONCLUSIONS: The strategy to safely exclude PE in COVID-19 outpatients should not differ from that used in non-COVID-19 patients. The added value of the COVID-19 PE prediction model is minor. KEY POINTS: • D-dimer level remains the most important predictor of pulmonary embolism in COVID-19 patients. • The AUCs of the revised Geneva and Wells scores using an age-adjusted D-dimer threshold were 0.587 (95%CI, 0.572 to 0.603) and 0.588 (95%CI, 0.572 to 0.603). • The AUC of COVID-19-specific strategy to rule out pulmonary embolism ranged from 0.513 (95%CI: 0.503 to 0.522) to 0.609 (95%CI: 0.594 to 0.623).

Artificial intelligence: A critical review of applications for lung nodule and lung cancer.

Artificial intelligence (AI) is a broad concept that usually refers to computer programs that can learn from data and perform certain specific tasks. In the recent years, the growth of deep learning, a successful technique for computer vision tasks that does not require explicit programming, coupled with the availability of large imaging databases fostered the development of multiple applications in the medical imaging field, especially for lung nodules and lung cancer, mostly through convolutional neural networks (CNN). Some of the first applications of AI is this field were dedicated to automated detection of lung nodules on X-ray and computed tomography (CT) examinations, with performances now reaching or exceeding those of radiologists. For lung nodule segmentation, CNN-based algorithms applied to CT images show excellent spatial overlap index with manual segmentation, even for irregular and ground glass nodules. A third application of AI is the classification of lung nodules between malignant and benign, which could limit the number of follow-up CT examinations for less suspicious lesions. Several algorithms have demonstrated excellent capabilities for the prediction of the malignancy risk when a nodule is discovered. These different applications of AI for lung nodules are particularly appealing in the context of lung cancer screening. In the field of lung cancer, AI tools applied to lung imaging have been investigated for distinct aims. First, they could play a role for the non-invasive characterization of tumors, especially for histological subtype and somatic mutation predictions, with a potential therapeutic impact. Additionally, they could help predict the patient prognosis, in combination to clinical data. Despite these encouraging perspectives, clinical implementation of AI tools is only beginning because of the lack of generalizability of published studies, of an inner obscure working and because of limited data about the impact of such tools on the radiologists' decision and on the patient outcome. Radiologists must be active participants in the process of evaluating AI tools, as such tools could support their daily work and offer them more time for high added value tasks.