Search

Im Zuge der rasanten Entwicklung der Computertechnik ist künstliche Intelligenz heute zu einem wesentlichen Bestandteil des täglichen Lebens geworden. Auch die Bereiche Medizin und Lungengesundheit stellen in diesem Zusammenhang keine Ausnahme dar. Wissenschaftliche Forschung auf der Grundlage von Big Data durchzuführen, bedeutet nicht, einfach eine große Menge an Daten zu sammeln und dann den Maschinen die Arbeit zu überlassen. Wissenschaftler müssen vielmehr Probleme identifizieren, deren Lösung sich positiv auf die Versorgung der Patienten auswirken würde. In dieser Übersichtsarbeit wird die Rolle der künstlichen Intelligenz sowohl aus physiologischer als auch aus anatomischer Sicht erörtert. Ausgehend von der automatischen quantitativen Bewertung anatomischer Strukturen mithilfe der Bildgebung der Lunge betrachten wir die Themenbereiche der Erkennung von Krankheiten und der Einschätzung der Prognose auf der Grundlage von maschinellem Lernen. Die Bewertung der aktuellen Stärken und Einschränkungen wird es uns ermöglichen, einen breiteren Blick auf zukünftige Entwicklungen zu werfen.

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.

ObjectivesWhile chest high-resolution CT (HRCT) is correlated to severity and prognosis in asthma, it has not been studied in eosinophilic granulomatosis with polyangiitis (EGPA). Our objective is to study the prognostic value of baseline HRCT in EGPA patients.MethodsRetrospective, multicentre observational study in three French hospitals, including EGPA patients with available chest HRCT before any systemic treatment. Two experienced radiologists blinded to clinical data evaluated HRCT images using semi-quantitative scoring. HRCT characteristics were correlated with clinical features and outcome.ResultsAmong 46 patients, 38 (82.6%) had abnormal parenchymal findings on HRCT, including bronchial wall thickening (69.6%), mosaic perfusion (63.0%), ground-glass opacities (32.6%), bronchiectasis (30.4%), mucous plugging (21.7%) and consolidations (17.4%). Patients were clustered into three groups depending on HRCT features: ground-glass pattern, i.e. with ground-glass opacities with or without bronchial abnormalities (group 1, 28.3%), bronchial pattern (group 2, 41.3%) and extra-pulmonary pattern with no significant abnormality (group 3, 30.4%). Group 2 showed less frequent cardiac involvement (31.6 vs 46.2 and 42.9% in groups 1 and 3), more frequent positive ANCA (52.6 vs 0.0 and 14.3%) and higher eosinophil count (median 7510 vs 4000 and 4250/mm3). Group 1 showed worse prognosis with more frequent steroid-dependency (58.3 vs 11.1 and 28.6%) and requirement for mepolizumab (25.0 vs 11.1 and 7.1%). Conversely, group 2 showed a better outcome with higher rates of remission (88.9 vs 41.6 and 71.4%).ConclusionChest HRCT at diagnosis of EGPA may have prognostic value and help clinicians better manage these patients.

IntroductionLung cancer screening (LCS) using low-dose CT has been demonstrated to reduce lung cancer-related mortality in large randomised controlled trials. Moving from trials to practice requires answering practical questions about the level of expertise of CT readers, the need for double reading as in trials and the potential role of artificial intelligence (AI). In addition, most LCS studies have predominantly included male participants with women being under-represented, even though the benefit of screening is greater for them. Thus, this study aims to compare the performance of a single CT reading by general radiologists trained in LCS using AI as a second reader to that of a double reading by expert thoracic radiologists, in a campaign for low-dose CT screening in high-risk women.Methods and analysisThis observational cohort study will recruit 2400 asymptomatic women aged between 50 and 74 years, current or former smokers with at least a 20 pack-year smoking history, in 4 different French district areas. Assistance with smoking cessation will be offered to current smokers. An initial low-dose CT scan will be performed, with subsequent follow-ups at 1 year and 2 years. The primary objective is to compare CT scan readings by a single LCS-trained, AI-assisted radiologist to that of an expert double reading. The secondary objectives are: to evaluate the performance of AI as a stand-alone reader; the adherence to screening of female participants; the influence on smoking cessation; the psychological consequences of screening; the detection of chronic obstructive pulmonary disease (COPD), coronary artery disease and osteoporosis on low-dose CT scans and the costs incurred by screening.Ethics and disseminationEthics approval was obtained from the Comité de Protection des Personnes Sud-Est 1 (ethics approval number: 2021-A02265-36 with an amendment on 15 July 2022). Trial results will be disseminated at conferences, through relevant patient groups and published in peer-reviewed journals.Trial registration numberNCT05195385.

History A 50-year-old woman presented to the emergency department of our hospital with a 2-day history of lower limb pain associated with unusual asthenia and diffuse arthralgia over the past 3 weeks. She was a native of Guinea and had lived in France for most of her life, working as a personal care assistant. Her only medical history of note was an occurrence of fetal death at 12 weeks gestation when she was 35 years old. She had bilateral lower limb swelling, without changes in skin temperature or color. All proximal and distal arterial pulses were felt. General physical examination findings were otherwise unremarkable. Her laboratory tests showed a decreased hemoglobin concentration of 8.9 g/dL (normal range, 12-16 g/dL), a decreased platelet count of 45 × 109/L (normal range, [150-400] × 109/L), a C-reactive protein level of 158 mg/L (normal range

SARS-Cov-2 infects the alveolar epithelial cells causing COVID-19 pneumonia of varying severity [1, 2]. Fifteen to 30% of patients develop acute respiratory distress syndrome (ARDS) requiring hospitalisation in intensive care units (ICU) and mechanical ventilation [2, 3]. At 3 months, there are persisting CT abnormalities in 17 to 91% of discharged COVID-19 patients [4–8], mainly consistent with an organising pneumonia pattern. These anomalies are more frequently reported in patients who were admitted to ICU [9]. Pulmonary fibrosis has been reported at autopsy of patients deceased from COVID-19 pneumonia, along with pulmonary microvascular thrombosis [10]. Fibrotic-like changes have also been reported at 6-month CT follow-up in COVID-19 survivors [11]. A proportion of survivors from the first SARS-Cov outbreak of 2003 had residual impairment of lung function and abnormal chest radiograph (CXR) findings at 6 months [12]. However, little is known about long-term changes on CT following severe SARS-Cov-2 pneumonia, or after ARDS in general, most descriptions being limited to CXR.

Lung cancer is the leading cause of cancer death in France and worldwide (20 % of cancer deaths). This mortality is partly linked to an overrepresentation of metastatic stages at diagnosis (approximately 55 % of lung cancers at diagnosis). Low-dose chest CT in a target population to detect early forms accessible to radical treatment has been evaluated through multiple randomized trials (NLST, NELSON, MILD, DANTE…). These trials demonstrated a reduction in lung cancer specific mortality. The current problem is to integrate a CT screening policy CT at a national level, which should be both efficient and cost-effective, while presenting the least harms for the eligible population. Finally, it is necessary to optimize the participation of the eligible population and particularly in the most deprived areas and ensure the proper implementation of smoking cessation measures.

Purpose The purpose of this study was to evaluate the prevalence of an atypical, alveolar presentation of pulmonary metastases from pancreatic adenocarcinoma (PDAC) on computed tomography (CT) and to correlate CT features with those obtained at histopathologic analysis. Material and methods A total of 76 patients with lung metastases from PDAC over a 10-year period (2009–2019) in a French university hospital were retrospectively included. There were 34 men and 42 women with a mean age of 67.6 ± 11.3 (SD) years (range: 38–89 years). CT features of PDAC were classified according to their presentations as usual metastatic pattern or atypical alveolar pattern; the atypical alveolar pattern corresponding to either ground glass nodules or opacities, solid nodules with a halo sign, “air-space” nodules with air bronchogram, or parenchymal consolidation. Imaging-histopathologic correlation was performed when tissue samples were available. Results Pulmonary metastases were synchronous in 36 patients (36/76; 47%) and metachronous in 40 patients (40/76; 53%). A predominant alveolar presentation on CT was observed in 17 patients (17/76, 22%). Nodules with halo sign were the predominant alveolar pattern in 7 patients (7/17; 41%), air-space nodules were predominant in 4 patients (4/17; 24%) whereas pure ground glass nodules and consolidations were observed as predominant features in 3 patients (3/17; 18%) each. For 5 patients who had histopathological confirmation, alveolar metastases of PDAC were characterized by columnar tumor cells lining the alveolar wall, which was not seen in other radiological presentations, whereas there were no differences regarding mucin secretion between pulmonary metastases with alveolar presentation and those with typical pattern. Conclusions Lung metastases from PDAC may present with a so-called “alveolar” pattern on CT. This misleading CT features is found in 22% of patients with lung metastases from PDAC and is due to lepidic growth of the metastatic cells.

Sarcopenia is observed in about 50% of cancer patients. Cancer-related sarcopenia negatively affects survival and is a predictive factor of anticancer drug toxicity. Sarcopenia diagnosis is challenging in routine care. We investigated whether plasma creatinine and cystatin C predict sarcopenia diagnosis in the specific population of cancer patients.Two common diagnostic criteria of cancer-related sarcopenia based on skeletal muscle mass ± handgrip strength were separately applied as the "gold standard" sarcopenia definition. Four sarcopenia indexes based on creatinine and cystatin C values were evaluated: Creatinine/Cystatin C, Glomerular Filter rate (GFR) Cockroft-Gault/GFR CKD-EPI, GFR Cockroft-Gault/GFR Grubb and GFR Cockroft-Gault/GFR simple. The receiver operating characteristic (ROC) curves and the area under the ROC curves were applied to evaluate the sarcopenia diagnostic accuracy of the four different sarcopenia indexes.A total of 99 patients were included. Among them, 47.5% were overweight or obese. The ratio creatinine/cystatin C (ratio value at 0.8) more accurately predicts the diagnosis of sarcopenia in the entire population based on low skeletal muscle mass and low handgrip strength (sensitivity, specificity, accuracy and Youden index at 0.77, 0.57, 0.90, 0.34 respectively). The other evaluated ratios predict sarcopenia with a lower specificity in all conditions. In the overweight/obese group, the results are similar. The ratio creatinine/cystatin C (ratio value at 1) accurately predicts sarcopenia with a sensitivity, a specificity, an accuracy and a Youden index at 0.50, 0.86, 0.95, 0.36 respectively in overweight/obese population.The creatinine/cystatin C ratio is a useful and simple biomarker to predict sarcopenia in cancer patients. Moreover, this sarcopenia index also seems to be a strong sarcopenia diagnosis biomarker in overweight and obese cancer patients. Our results must be confirmed in a larger cohort.

Following the American National Lung Screening Trial results in 2011 a consortium of French experts met to edit a statement. Recent results of other randomized trials gave the opportunity for our group to meet again in order to edit updated guidelines. After literature review, we provide here a new update on lung cancer screening in France. Notably, in accordance with all international guidelines, the experts renew their recommendation in favor of individual screening for lung cancer in France as per the conditions laid out in this document. In addition, the experts recommend the very rapid organization and funding of prospective studies, which, if conclusive, will enable the deployment of lung cancer screening organized at the national level.

Purpose The purpose of this study was to analyze the impact of different options for reduced-dose computed tomography (CT) on image noise and visibility of pulmonary structures in order to define the best choice of parameters when performing ultra-low dose acquisitions of the chest in clinical routine. Materials and methods Using an anthropomorphic chest phantom, CT images were acquired at four defined low dose levels (computed tomography dose index [CTDIvol] = 0.15, 0.20, 0.30 and 0.40 mGy), by changing tube voltage, pitch factor, or rotation time and adapting tube current to reach the predefined CTDIvol-values. Images were reconstructed using two different levels of iteration (adaptive statistical iterative reconstruction [ASIR®]-v70% and ASIR®-v100%). Signal-to-noise ratio (SNR) as well as contrast-to-noise ratio (CNR) was calculated. Visibility of pulmonary structures (bronchi/vessels) were assessed by two readers on a 5-point-Likert scale. Results Best visual image assessments and CNR/SNR were obtained with high tube voltage, while lowest scores were reached with lower pitch factor followed by high tube current. Protocols favoring lower pitch factor resulted in decreased visibility of bronchi/vessels, especially in the periphery. Decreasing radiation dose from 0.40 to 0.30 mGy was not associated with a significant decrease in visual scores (P Conclusion Favoring high tube voltage is the best option for reduced-dose protocols. A decrease of SNR and CNR does not necessarily go along with reduced visibility of pulmonary structures.

Immune checkpoint inhibitors have emerged as a cornerstone of management in non-small-cell lung cancer (NSCLC). They are responsible for a wide spectrum of immune-related adverse effects, which greatly differ from the adverse events of standard chemotherapy. In this report, we describe a unique case of severe non-infectious cystitis in a patient receiving pembrolizumab for metastatic NSCLC. A 56-year-old female patient received first-line pembrolizumab for multimetastatic NSCLC, with a PD-L1 status of 90%. After 5 cycles, the patient presented with hematuria, urinary burning, urgent polyuria, painful urination and afebrile diarrhea. Urine was sterile, with a highly inflammatory cytology and high protein content. Diarrhea abated under symptomatic treatment but urinary symptoms persisted. Uroscanner found thickened, irregular bladder walls. Cystoscopy revealed a highly inflammatory and irregular mucosa, with diffuse inflammation and vasculo-exudative ulcerative remodeling on pathology. Grade 3 interstitial cystitis was diagnosed and attributed to immunotherapy. Following suspension of pembrolizumab and administration of systemic steroids, the patient had an excellent clinical and radiological response. Because of thoracic progression, oncological treatment was resumed with third-line paclitaxel-bevacizumab. Overall, this case highlights a novel toxicity of immune checkpoint inhibitors. Appropriate and timely diagnosis of rare immune-related adverse events is essential, as proper management of these toxicities may enable ICI continuation in patients who benefit the most from immunotherapy.

• Current guidelines recommend the use of unenhanced chest CT as first-line imaging in patients suspected of COVID-19 pneumonia. • In unselected COVID-19 patients, the prevalence of PE at baseline is no different from the one in COVID-19 negative patients. • There is no clear benefit to routinely perform CTPA as first-line imaging modality in patients suspected of COVID-19 pneumonia

The purpose of this study was to determine whether a single reconstruction kernel or both high and low frequency kernels should be used for training deep learning models for the segmentation of diffuse lung disease on chest computed tomography (CT).Two annotated datasets of COVID-19 pneumonia (323,960 slices) and interstitial lung disease (ILD) (4,284 slices) were used. Annotated CT images were used to train a U-Net architecture to segment disease. All CT slices were reconstructed using both a lung kernel (LK) and a mediastinal kernel (MK). Three different trainings, resulting in three different models were compared for each disease: training on LK only, MK only or LK+MK images. Dice similarity scores (DSC) were compared using the Wilcoxon signed-rank test.Models only trained on LK images performed better on LK images than on MK images (median DSC = 0.62 [interquartile range (IQR): 0.54, 0.69] vs. 0.60 [IQR: 0.50, 0.70], P0.001 for COVID-19 and median DSC = 0.62 [IQR: 0.56, 0.69] vs. 0.50 [IQR 0.43, 0.57], P0.001 for ILD). Similarly, models only trained on MK images performed better on MK images (median DSC = 0.62 [IQR: 0.53, 0.68] vs. 0.54 [IQR: 0.47, 0.63], P0.001 for COVID-19 and 0.69 [IQR: 0.61, 0.73] vs. 0.63 [IQR: 0.53, 0.70], P0.001 for ILD). Models trained on both kernels performed better or similarly than those trained on only one kernel. For COVID-19, median DSC was 0.67 (IQR: =0.59, 0.73) when applied on LK images and 0.67 (IQR: 0.60, 0.74) when applied on MK images (P0.001 for both). For ILD, median DSC was 0.69 (IQR: 0.63, 0.73) when applied on LK images (P = 0.006) and 0.68 (IQR: 0.62, 0.72) when applied on MK images (P0.99).Reconstruction kernels impact the performance of deep learning-based models for lung disease segmentation. Training on both LK and MK images improves the performance.