Stereotactic Arrhythmia Radioablation (STAR) is a promising treatment for refractory ventricular tachycardia. However, its precision may be hampered by cardiac and respiratory motions. Multiple techniques exist to mitigate the effects of these displacements. The purpose of this work was, based on cardiac and respiratory dynamic CT scans, to generate a patient-specific dynamic model of the structures of interest, that enables simulation of treatments for evaluation of motion management methods. Deep learning-based segmentation was used to extract the geometry of the cardiac structures, whose deformations and displacements were assessed using deformable and rigid image registrations. The combination of the model with dose maps enabled to evaluate the dose locally accumulated during the treatment. The reproducibility of each step was evaluated considering expert references, and treatment simulations were evaluated using data of a physical phantom. The exploitation of the model was illustrated on the data of nine patients, demonstrating that the impact of cardiorespiratory dynamics is potentially important and highly patient-specific, and allowing for future evaluations of motion management methods.

Magnetic resonance imaging (MRI) of patients with cardiac implantable electronic devices (CIEDs) is challenged by susceptibility artifacts. We propose a simulation-to-learning framework for artifact suppression in cine MRI. Subjectspecific digital phantoms were derived from the ACDC dataset by encoding proton density (PD), longitudinal relaxation (T1), transverse relaxation (T2), effective transverse relaxation (T2*), and off-resonance ($\Delta \omega$). Susceptibility effects were modeled as randomized dipole perturbations applied to the off-resonance map, yielding perfectly paired-MRI images artifact-free and artifact-present. MRI reconstruction was conducted with KomaMRI. In order to remove artifacts effects, approximately 32,500 paired images were generated to train a Residual U-Net. It resulted in a improved SSIM from 0.63 (artifacted) to 0.71 (denoised). After transfer learning on a small real coronal cine cohort, high-frequency energy decreased by ~ 52% and edge-aware sharpness showed a reduction (-23%, $\sigma=1$), indicating substantial noise suppression with preserved edges. These results demonstrate that digital phantoms and paired simulations enable artifact correction and provide a scalable pathway toward artifactresilient cine MRI in patients with CIEDs.

PURPOSE:The purpose of this study was to develop an artificial intelligence (AI) tool to assist recognition of three major interstitial lung disease (ILD) patterns on high-resolution computed tomography (HRCT) and to evaluate its added value in supporting decision-making for non-specialist radiologists. MATERIAL AND METHODS:This retrospective, multicenter study included 1097 HRCT examinations. Of these, 989 (90.15%) were used for development and 108 (9.85%) for external testing. A two-stage architecture inspired by domain-specific pretraining was employed. The encoder of a three-dimensional ILD segmentation model was kept to extract 7168 disease-specific features per HRCT, which were combined with age and sex in a deep learning model to predict three radiological patterns (usual interstitial pneumonia, non-specific interstitial pneumonia and fibrotic bronchiolocentric interstitial pneumonia) as diagnosed in multidisciplinary discussions (MDD). The external test dataset was interpreted by seven thoracic radiologists to establish a second reference (majority's vote) and by eight radiology residents with and without AI assistance. Accuracy, sensitivity and specificity were calculated for each pattern. RESULTS:The AI system achieved 77.8% accuracy on the external test dataset using MMD as a reference standard, within the range of thoracic experts (median, 75.6%; range: 61.1-81.5). AI assistance improved residents' median accuracy (+14.8% of absolute increase) and reduced reading time by 20.7% (P < 0.001). Six out of eight residents assisted by AI (75%) performed worse than AI alone. CONCLUSION:AI can accurately classify major ILD patterns and help less-experienced readers improve their performance. However, the level of improvement was inconsistent, and non-specialists rarely equaled the performance of AI alone.

Computed tomography (CT) analysis of lung morphology has significantly advanced our understanding of acute respiratory distress syndrome (ARDS). During the Coronavirus Disease 2019 (COVID-19) pandemic, CT imaging was widely utilized to evaluate lung injury and was suggested as a tool for predicting patient outcomes. However, data specifically focused on patients with ARDS admitted to intensive care units (ICUs) remain limited. This retrospective study analyzed patients admitted to ICUs between March 2020 and November 2022 with moderate to severe COVID-19 ARDS. All CT scans performed within 48 h of ICU admission were independently reviewed by three experts. Lung injury severity was quantified using the CT Severity Score (CT-SS; range 0–25). Patients were categorized as having severe disease (CT-SS ≥ 18) or non-severe disease (CT-SS < 18). The primary outcome was all-cause mortality at 90 days. Secondary outcomes included ICU mortality and medical complications during the ICU stay. Additionally, we evaluated a computer-assisted CT-score assessment using artificial intelligence software (CT Pneumonia Analysis®, SIEMENS Healthcare) to explore the feasibility of automated measurement and routine implementation. A total of 215 patients with moderate to severe COVID-19 ARDS were included. The median CT-SS at admission was 18/25 [interquartile range, 15–21]. Among them, 120 patients (56

Pulmonary alveolar proteinosis is suspected when a "crazy paving" pattern is observed on a chest CT scan. This diagnosis is confirmed by the presence of eosinophilic extracellular material that shows positive staining with Periodic Acid Schiff on bronchoalveolar lavage samples. The autoimmune form of pulmonary alveolar proteinosis is confirmed by detecting anti-granulocyte-macrophage colony-stimulating factor antibodies in the patient's serum. The historical first-line treatment for autoimmune pulmonary alveolar proteinosis is whole lung lavage, which should only be performed in expert centers. It remains the preferred treatment for patients experiencing respiratory failure, especially at the time of diagnosis. Inhaled granulocyte-macrophage colony-stimulating factor supplementation with molgramostim or sargramostim is now considered a first-line treatment in the international guidelines for autoimmune pulmonary alveolar proteinosis, following the positive results of recent randomized placebo-controlled studies. Rituximab and plasmapheresis can be prescribed as third- and fourth-line treatments, respectively. Lung transplantation may be considered for eligible patients experiencing terminal respiratory failure. A deeper understanding of the pathogenesis of autoimmune pulmonary alveolar proteinosis has opened up new therapeutic avenues, such as the use of PPARγ agonists or statins.

PURPOSE/OBJECTIVE:This study proposes an implementation of the mid-position (MidP) approach to compensate for cardio-respiratory motions in the context of Stereotactic Arrhythmia Radioablation (STAR) and evaluates its benefits compared to an internal target volume (ITV) approach. MATERIALS AND METHODS:Fifteen patients who underwent STAR for refractory ventricular tachycardia in our institution were included in this retrospective planning study. For each patient, a cardiac-gated four-dimensional computed tomography (4D-CTcard) scan and a respiratory-gated four-dimensional computed tomography (4D-CTresp) were acquired. All patients were treated using a volumetric modulated arc therapy technique using an in-treatment Cone-Beam CT (CBCT) image guidance. The MidP approach was implemented to compensate for uncertainties, including cardio-respiratory motions characterized using the 4D-CTcard and 4D-CTresp scans, and the inter-fraction motions measured using the CBCT scans. For comparison purposes, the ITV approach was also implemented. Both approaches were compared in terms of planning target volume (PTV) volumes, doses to organs-at-risk, and clinical target volume (CTV) doses, assessed using a 4D modeling method that estimates the accumulated dose. RESULTS:Compared with the ITV method, the MidP approach resulted in a mean [min-max] relative PTV volume reduction of 30% [19%, 48%] (p < 0.001, Wilcoxon signed rank test). The mean [min-max] D95% CTV coverage was 105% [101%-114%] and 107% [101%-117%] of the prescription dose for MidP and ITV-based plans, respectively. The median dose to the whole heart was significantly lower with MidP-based plans with a mean difference of -0.5 Gy (p = 0.0084). The near-maximum dose (D1%) delivered to left coronary arteries, aorta, and stomach was systematically lower with the MidP-based plans. CONCLUSION:Compared to ITV based approach, the use of MidP strategy for treatment planning of STAR leads to significantly smaller PTV and lower surrounding OAR doses while still achieving a clinically acceptable CTV coverage.

Background: Computed tomography (CT) is widely used in the early stages of acute respiratory distress syndrome (ARDS) for diagnose and patient management. Recent ARDS guidelines have questioned the utility of identifying ARDS subphenotype to improve prognostic and guide ventilation strategies. Excluding COVID-19 related ARDS, the predictive value of CT during the early phase of ARDS remains unclear. Methods: We performed a 7-year retrospective study on patients admitted in the medical intensive care unit (ICU) of a tertiary teaching hospital, from January 2016 to January 2023. All patients with ARDS unrelated to COVID-19 infection who underwent a chest CT scan within 48 hours of ARDS onset were included. Lung injury severity was assessed using a semi-quantitative CT severity score (CT-SS, range: 0-25 points), evaluated by two trained intensivists. The primary outcome was 90 days all-cause mortality. Secondary outcomes included 28-days mortality, duration of mechanical ventilation, and medical complications during the ICU stay. We also investigated the relationship between ventilatory variables and CT-SS. Results: We included 114 patients with moderate to severe ARDS. The median CT severity score at admission was 18 [IQR:14-22]. 58 patients (50.1%) were classified as having severe lung injury and 56 patients (49.9%) as non-severe. The 90-day all-cause mortality was 41.2% (47/114), with no significant difference in survival between the severe and non-severe CT groups (p = 0.84, log-rank test). CT severity was also not associated with the occurrence of complications during the ICU stay. Regarding ventilatory parameters, patients with a severe CT-score had significantly higher plateau pressures (26 [23-28] cmH2O vs 25 [20-26] cmH2O, p = 0.01) and lower static compliance (28.8 [23.1- 36.1] ml/cmH2O vs 32.7 [25.8-38.3] ml/cmH2O, p = 0.05). No strong correlation was observed between the CT-score and other ventilatory variables. Conclusion: We found that early assessment of CT severity in ARDS was not associated with 90-days mortality and showed no clear relationship with ventilatory impairment. Initial CT imaging did not appear to predict ICU outcomes. These findings question the utility of routine CT use in the early management of ARDS and are consistent with recent expert guidelines, which do not support its widespread use in this context.

BackgroundAutoimmune pulmonary alveolar proteinosis (aPAP) is a rare disease that may progress towards pulmonary fibrosis. Data about fibrosis prevalence and risk factors are lacking.MethodsIn this retrospective multicentre nationwide cohort, we included patients newly diagnosed with aPAP between 2008 and 2018 in France and Belgium. Data were collected from medical records using a standardised questionnaire.Results61 patients were included in the final analysis. We identified 5 patients (8%) with fibrosis on initial computed tomography (CT) and 16 patients (26%) with fibrosis on final CT after a median time of 3.6 years. Dust exposure was associated with pulmonary fibrosis occurrence (OR 4.3; p=0.038). aPAP patients treated with whole-lung lavage, rituximab or granulocyte–monocyte colony-stimulating factor therapy did not have more fibrotic evolution than patients who did not receive these treatments (n=25 out of 45, 57%versusn=10 out of 16, 62%; p=0.69). All-cause mortality was significantly higher in fibrotic than in nonfibrotic cases (n=4 out of 16, 25%versusn=2 out of 45, 4.4%; p=0.036, respectively).ConclusionIn our population, a quarter of aPAP patients progressed towards pulmonary fibrosis. Dust exposure seems to be an important factor associated with this complication. More studies are needed to analyse precisely the impact of dust exposure impact, especially silica, in patients with aPAP.

BACKGROUND:Clinical course prediction of patients with interstitial lung disease (ILD) admitted to the intensive care unit (ICU) for acute respiratory failure (ARF) can be challenging. This study aimed to characterize the prognostic value of admission chest CT-scan in this situation. METHODS:We retrospectively included ILD patients admitted to a French ICU for acute respiratory failure requiring oxygen. Patients with lymphangitis carcinomatosis and ANCA vasculitis were excluded. We analyzed every admission chest CT-scan using two different approaches: a visual analysis (grading the extent of traction bronchiectasis, ground glass and honeycomb) and an automated analysis (grading the extent of ground glass and consolidation with a dedicated software). The primary outcome was ICU mortality. RESULTS:Between January 2014 and October 2020, 81 patients presented an acute respiratory failure with ILD on the admission chest CT-scan. In univariate analysis, only the main pulmonary artery diameter differed between patients who survived and those who died in ICU (30 vs 32 mm, p = 0.021). In multivariate analysis, none of the radiological funding was associated with ICU mortality. Visual and automated analyses did not yield different results, with a strong correlation between the two methods. However, the identification of an UIP pattern (and the presence of honeycomb) was associated with a poorer response to corticosteroid therapy. CONCLUSION:Our study showed that the extent of radiological findings and the severity of fibrosis indices on admission chest CT scans of ILD patients admitted to the ICU for ARF were not associated with subsequent deterioration.

To assess the role of CT venography (CTV) in the diagnosis of venous thromboembolism (VTE) during the postpartum period. 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. The final study population consisted of 123 women. On-site CTPA reports mentioned PE in seven women (7/123, 5.7

La fibrose pulmonaire idiopathique (FPI) est une pathologie interstitielle pulmonaire grave dont le pronostic est extrêmement mauvais et qui se caractérise par un pattern radiologique et histologique de pneumonie interstitielle commune. Certains patients atteints de FPI peuvent développer une exacerbation aiguë de la fibrose pulmonaire idiopathique (EA-FPI) avec un pronostic extrêmement défavorable. L'objectif de cette étude était de déterminer les facteurs prédictifs et pronostiques d'EA-FPI. Nous avons analysé rétrospectivement des patients atteints de FPI suivis au CHU de Rennes et dans quatre hôpitaux publics locaux. L'incidence et les résultats des EA-FPI ont été étudiés. Des analyses univariées et multivariées ont été utilisées pour identifier les variables indépendantes associées à l'EA FPI. Parmi les 307 patients inclus, 81 (26,4%) ont développé une EA-FPI dans un délai médian de 2,2 ± 2,0 ans après le diagnostic. Les facteurs associés de manière indépendante à l'apparition d'une EA-FPI étaient une capacité vitale forcée (CVF) plus basse HR 0,94 [0,91; 0,97] (p = 0,0003), des taux de neutrophiles et de lymphocytes sanguins plus élevés au diagnostic: respectivement HR 1,01 [1,00; 1,01], (p = 0,0047), HR 1,13 [1,04; 1,22] (p = 0,0028). La survie médiane après une EA-FPI était de 5,39 mois [2,6; 9,80]. L'EA-FPI a été associée à une survie médiane plus courte: 2,84 ans contre 4,33 ans (p < 0,0001). Les facteurs de risque d'EA-FPI identifiés dans cette cohorte étaient une CVF plus basse et un taux de neutrophiles et de lymphocytes sanguin plus élevé au diagnostic. Il serait intéressant d'accorder plus d'attention à la numération des leucocytes au moment du diagnostic. Des recherches supplémentaires sont nécessaires pour confirmer ces résultats, idéalement dans le cadre d'études prospectives.

Background: The present article is an English-language version of the French National Diagnostic and Care Protocol, a pragmatic tool to optimize and harmonize the diagnosis, care pathway, management and follow-up of lymphangioleiomyomatosis in France. Methods: Practical recommendations were developed in accordance with the method for developing a National Diagnosis and Care Protocol for rare diseases of the Haute Autorit ' e de Sant ' e and following international guide-lines and literature on lymphangioleiomyomatosis. It was developed by a multidisciplinary group, with the help of patient representatives and of RespiFIL, the rare disease network on respiratory diseases. Results: Lymphangioleiomyomatosis is a rare lung disease characterised by a proliferation of smooth muscle cells that leads to the formation of multiple lung cysts. It occurs sporadically or as part of a genetic disease called tuberous sclerosis complex (TSC). The document addresses multiple aspects of the disease, to guide the clinicians regarding when to suspect a diagnosis of lymphangioleiomyomatosis, what to do in case of recurrent pneumothorax or angiomyolipomas, what investigations are needed to make the diagnosis of lym-phangioleiomyomatosis, what the diagnostic criteria are for lymphangioleiomyomatosis, what the principles of management are, and how follow-up can be organised. Recommendations are made regarding the use of pharmaceutical specialties and treatment other than medications. Conclusion: These recommendations are intended to guide the diagnosis and practical management of pul-monary lymphangioleiomyomatosis. (c) 2023 SPLF and Elsevier Masson SAS. All rights reserved.

Cardiac radioablation (CR) is a promising treatment for patients with refractory ventricular tachycardia. Several medical centers around the world reported patient treatments using Cyberknife. Some of these teams used the implantable cardioverter defibrillator (ICD) lead as the sole tracking surrogate during treatment. However, the motions of CR target and ICD lead may differ along the cardiac cycle. The objectives of this study were to evaluate: (i) the reliability of the ICD lead to track the target position along the cardiac cycle; (ii) the geometrical impact of the tracking uncertainties on the target volume. Contrast-enhanced breath-hold cardiac ECG-gated computed tomography (c4D-CT) of 15 patients suffering from arrhythmias were acquired and reconstructed in 10 3D images datasets. A CR target was delineated on one phase of the c4D-CT for each patient and propagated to the other phases of the c4D-CT, using deformable image registration. The tip of the ICD lead was manually located on each phase of the c4D-CT for each patient. Target and ICD lead motion discrepancies were evaluated by comparing the positions of the centroids of both objects (i.e., target and lead) across phases. At each phase of the cardiac cycle, we defined the motion vector of an object, as the distance between its current position and the average position of the object across all phases. The Mean Absolute Error (MAE) between motion vectors of the target and the ICD lead on the same phase was calculated for each patient as a metric of motion discrepancies. For each patient an Internal Target Volume (ITV) has been built as the volume encompassing all the positions of the target relatively to the lead along the cardiac cycle. The mean (min - max) MAE was of 3.3 (1.5–4.9) mm. The distance between the ICD lead and the target (d) has also been evaluated, and its mean (min–max) value was 65 (19–120) mm. No correlation was found between MAE and d (Pearson = 0.38, P = 0.16). The relative volume increase between CTV and ITV was 104 (39–151) %. Non-negligible discrepancies were observed between cardiac-induced motions of ICD lead and target in cardiac radioablation. An ITV that encompasses the relative positions of the target with respect to the ICD lead is, on average, twice the volume of the CTV.

La pneumopathie interstitielle diffuse est un événement indésirable associé à de nombreux médicaments anticancéreux, y compris certains anticorps conjugués récents utilisés dans le cancer du sein. En l’absence de signes cliniques ou radiologiques spécifiques, le diagnostic d’une pneumopathie interstitielle diffuse médicamenteuse repose souvent sur un diagnostic d’exclusion. Lorsqu’ils sont présents, les symptômes les plus fréquents sont des signes respiratoires (toux, dyspnée, douleur thoracique) et des signes généraux (fatigue, fièvre). Toute suspicion de pneumopathie interstitielle diffuse doit être explorée par imagerie et, en cas de doute, le scanner doit être relu par un pneumologue et un radiologue. L’interruption temporaire ou définitive du traitement est fonction de la sévérité de la pneumopathie interstitielle diffuse et du type d’anticorps conjugué. Pour les cas asymptomatiques (grade 1), l’efficacité des corticostéroïdes n’est pas clairement établie ; pour les grades supérieurs, la balance bénéfice/risque d’une corticothérapie à long terme doit être prise en compte pour définir la dose et la durée du traitement. Une hospitalisation et une oxygénothérapie sont indiquées pour les cas graves (grades 3-4). Le suivi du patient nécessite l’expertise d’un pneumologue avec réalisation de scanners thoraciques répétés et évaluations régulières de la fonction pulmonaire (spirométrie et capacité de diffusion du monoxyde de carbone). La prévention des pneumopathies interstitielles diffuses induites par les anticorps conjugués et de leur évolution vers un grade plus sévère repose sur l’expertise d’un réseau multidisciplinaire d’experts. Ce réseau d’expertise permet d’évaluer les facteurs de risque individuels et de proposer une prise en charge précoce, un suivi étroit et une éducation thérapeutique des patients.

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.