Deglutition and the Regulation of the Swallow Motor Pattern.

Despite centuries of investigation, questions and controversies remain regarding the fundamental genesis and motor pattern of swallow. Two significant topics include inspiratory muscle activity during swallow (Schluckatmung, i.e., "swallow-breath") and anatomical boundaries of the swallow pattern generator. We discuss the long history of reports regarding the presence or absence of Schluckatmung and the possible advantages of and neural basis for such activity, leading to current theories and novel experimental directions.

Alveolar membrane and capillary function in COVID-19 convalescents: insights from chest MRI.

OBJECTIVES: To investigate potential presence and resolution of longer-term pulmonary diffusion limitation and microvascular perfusion impairment in COVID-19 convalescents. MATERIALS AND METHODS: This prospective, longitudinal study was carried out between May 2020 and April 2023. COVID-19 convalescents repeatedly and age/sex-matched healthy controls once underwent MRI including hyperpolarized 129Xe MRI. Blood samples were obtained in COVID-19 convalescents for immunophenotyping. Ratios of 129Xe in red blood cells (RBC), tissue/plasma (TP), and gas phase (GP) as well as lung surface-volume ratio were quantified and correlations with CD4+/CD8+ T cell frequencies were assessed using Pearson's correlation coefficient. Signed-rank tests were used for longitudinal and U tests for group comparisons. RESULTS: Thirty-five participants were recruited. Twenty-three COVID-19 convalescents (age 52.1 ± 19.4 years, 13 men) underwent baseline MRI 12.6 ± 4.2 weeks after symptom onset. Fourteen COVID-19 convalescents underwent follow-up MRI and 12 were included for longitudinal comparison (baseline MRI at 11.5 ± 2.7 weeks and follow-up 38.0 ± 5.5 weeks). Twelve matched controls were included for comparison. In COVID-19 convalescents, RBC-TP was increased at follow-up (p = 0.04). Baseline RBC-TP was lower in patients treated on intensive care unit (p = 0.03) and in patients with severe/critical disease (p = 0.006). RBC-TP correlated with CD4+/CD8+ T cell frequencies (R = 0.61/ - 0.60) at baseline. RBC-TP was not significantly different compared to matched controls at follow-up (p = 0.25). CONCLUSION: Impaired microvascular pulmonary perfusion and alveolar membrane function persisted 12 weeks after symptom onset and resolved within 38 weeks after COVID-19 symptom onset. CLINICAL RELEVANCE STATEMENT: 129Xe MRI shows improvement of microvascular pulmonary perfusion and alveolar membrane function between 11.5 ± 2.7 weeks and 38.0 ± 5.5 weeks after symptom onset in patients after COVID-19, returning to normal in subjects without significant prior disease. KEY POINTS: • The study aims to investigate long-term effects of COVID-19 on lung function, in particular gas uptake efficiency, and on the cardiovascular system. • In COVID-19 convalescents, the ratio of 129Xe in red blood cells/tissue plasma increased longitudinally (p = 0.04), but was not different from matched controls at follow-up (p = 0.25). • Microvascular pulmonary perfusion and alveolar membrane function are impaired 11.5 weeks after symptom onset in patients after COVID-19, returning to normal in subjects without significant prior disease at 38.0 weeks.

Eco-geographic and sexual variation of the ribcage in Homo sapiens.

Up to now, Allen and Bergmann's rules have been studied in modern humans by analyzing differences in limb length, height, or body mass. However, there are no publications studying the effects of latitude in the 3D configuration of the ribcage. To assess this issue, we digitally reconstructed the ribcages of a balanced sample of 109 adult individuals of global distribution. Shape and size of the ribcage was quantified using geometric morphometrics. Our results show that the ribcage belonging to tropical individuals is smaller and slenderer compared to others living in higher latitudes, which is in line with Allen and Bergmann's rules and suggests an allometric relationship between size and shape. Although sexual dimorphism was observed in the whole sample, significant differences were only found in tropical populations. Our proposal is that, apart from potential sexual selection, avoiding heat loss might be the limiting factor for sexual dimorphism in cold-adapted populations.

The lung ultrasound in children with SARS-COV-2 infection: a national multicenter prospective study.

Coronavirus disease-19 (COVID-19) caused hospitalizations, severe disease, and deaths in any age, including in the youngest children. The aim of this multicenter national study is to characterize the clinical and the prognostic role of lung ultrasound (LU) in children with COVID-19. We enrolled children between 1 month and 18 years of age with severe acute respiratory syndrome coronavirus 2 (SARS-CoV2) infection who underwent a LU within 6 h from the first medical evaluation. A total of 213 children were enrolled, 51.6% were male, median age was 2 years and 5 months (interquartile range (IQR) 4 months -11 years and 4 months). One hundred and forty-eight (69.4%) children were admitted in hospital, 9 (6.1%) in pediatric intensive care unit. We found an inverse correlation between the lung ultrasound score (LUS) and the oxygen saturation at the first clinical evaluation (r = -0.16; p = 0.019). Moreover, LUS was significantly higher in patients requiring oxygen supplementation (8 (IQR 3-19) vs 2 (IQR 0-4); p = 0.001). Among LU pathological findings, irregular pleural lines, subpleural consolidations, and pleural effusions were significantly more frequent in patients needing oxygen supplementation (p = 0.007, p = 0.006, and p = 0.001, respectively).  Conclusion: This multicenter study showed that LU in children with COVID-19 can highlight pleural line irregularities, vertical artifacts, and subpleural consolidation. Notably, children with higher LUS have a higher risk of hospitalization and need for oxygen supplementation, supporting LU as a valid and safe point-of-care first level tool for the clinical evaluation of children with COVID-19. What is Known: • Few children infected with SARS-CoV-2 develop a severe disease and need oxygen therapy. • Lung ultrasound can easily detect low respiratory tract infection during SARS-CoV-2. What is New: • Children with higher lung ultrasound score have a higher risk of need for oxygen supplementation. • Irregular pleural line, sub-pleural consolidations and pleural effusions were significantly more frequent in patients needing oxygen supplementation.

Impact Location Dependence of Behind Armor Blunt Trauma Injury Assessed Using a Human Body Finite Element Model.

Behind armor blunt trauma (BABT), resulting from dynamic deformation of protective ballistic armor into the thorax, is currently assessed assuming a constant threshold of maximum backface deformation (BFDs) (44 mm). Although assessed for multiple impacts on the same armor, testing is focused on armor performance (shot-to-edge and shot-to-shot) without consideration of the underlying location on the thorax. Previous studies identified the importance of impacts on organs of animal surrogates wearing soft armor. However, the effect of impact location was not quantified outside the threshold of 44 mm. In the present study, a validated biofidelic advanced human thorax model (50th percentile male) was utilized to assess the BABT outcome from varying impact location. The thorax model was dynamically loaded using a method developed for recreating BABT impacts, and BABT events within the range of real-world impact severities and locations were simulated. It was found that thorax injury depended on impact location for the same BFDs. Generally, impacts over high compliance locations (anterolateral rib cage) yielded increased thoracic compression and loading on the lungs leading to pulmonary lung contusion (PLC). Impacts at low compliance locations (top of sternum) yielded hard tissue fractures. Injuries to the sternum, ribs, and lungs were predicted at BFDs lower than 44 mm for low compliance locations. Location-based injury risk curves demonstrated greater accuracy in injury prediction. This study quantifies the importance of impact location on BABT injury severity and demonstrates the need for consideration of location in future armor design and assessment.

Diagnostic imaging descriptions and prevalence of presumed phrenic lymph nodes in dogs.

The phrenic lymph node (PLN) has been described in bovine, equine, and feline anatomic references but descriptions in canine anatomic references are currently lacking. Authors have observed a well-defined, soft tissue attenuating, contrast-enhancing structure in a location consistent with previous anatomic descriptions of the PLN in dogs that underwent thoracic CT for neoplastic staging. The aims of this two-part, retrospective/prospective, anatomic, prevalence study were (1) to describe the presence of a soft tissue structure close to the thoracic caudal vena cava, defined as the presumed PLN, in a series of dogs that underwent CT as part of the staging for metastatic disease; (2) to confirm the lymphatic origin of the presumed PLN in a dog through postmortem examination; (3) to assess the prevalence of the presumed PLN in a population of dogs that underwent thoracic CT or MRI for different clinical purposes; and (4) to assess the possibility to visualize the presumed PLN with ultrasonography. The lymphatic origin of the presumed PLN was confirmed by postmortem examination in one dog. The presumed PLN was visible in 29 of 777 canine CT examinations (prevalence 3.7%). The presumed PLN was not visible in 9 of 10 prospectively recruited ultrasound cases. Most dogs with visible presumed PLNs were large-medium breeds that were presented for neoplastic staging purposes. Findings indicated that a structure consistent with the previously reported anatomic features of PLN in cattle, horses, and cats may be detected with a low prevalence in canine CT and MRI examinations.

Computer tomography measurements of the airway and thoracic cavity do not provide support for bronchial conformation as a predisposing factor of left cranial lung lobe torsion in pugs.

The objective of this retrospective clinical study was to determine if airway or thoracic cavity measurements in pugs, particularly the left cranial lung lobe, were significantly different from brachycephalic and mesocephalic control. Thoracic computed tomographic studies of 10 pugs, French bulldogs (FB), and Jack Russell Terriers (JRT) were analyzed. Thoracic height: width ratio (H:W), cross-sectional areas of the left mainstem bronchus (CSA LMB), left cranial lung lobe bronchus (CSA LCrBr), left caudal lung lobe bronchus (CSA LCauBr), CSA LCrBr relative to length (CSA LCrBr/length) and CSA LCauBr/length were measured and adjusted to body weight (/kg). CSA LMB/kg, CSA LCauBr/length/kg, and CSA LCrBr/length /kg were smaller in pugs and FB compared with JRT (P < .05), but no differences were found between pugs and FB. Cross-sectional areas of left cranial lung lobe bronchus /kg and CSA LCauBr/kg were smaller in pugs than JRT (P < .05), but no differences were found between pugs and FB or FB and JRT. No difference was found in thoracic H:W between any breeds. This demonstrated that pugs and FB had significantly narrower bronchi CSA/lengths ratios compared with JRT, but this was not limited to the LCBr. Airway measurements were not significantly different between brachycephalic breeds; therefore, the pugs' predisposition to left cranial lung lobe torsion cannot be solely explained by narrower lower airways.

Incidental findings of soft tissue gas in canine and feline thoracic computed tomographic examinations.

Computed tomography (CT) is a commonly used modality in small animal veterinary medicine. Anecdotally, gas bubbles are frequently identified in small animals undergoing thoracic CT examination. The aim of this retrospective observational study was to record the occurrence and prevalence of vascular and extravascular gas in routine thoracic CT examinations in dogs and cats. Patients with any clinical signs of diseases   related to soft tissue gas were excluded. A total of 84 canine studies and 90 feline studies were included, detecting gas in 66.7% of all studies, with gas more likely to be detected in canine studies (75.0%) than in feline cases (58.9%) (P = 0.009). Canine studies were more likely to have gas detected in postcontrast studies than in precontrast studies (P = 0.016). Intravascular gas was detected in 65.5% of all included studies (75% canine and 56.7% feline). Extravascular gas was detected in 13.8% of all studies (14.3% canine, 13.3% feline), with gas detected in the vertebral canal of 8.62% of all studies (canine 11.9%, feline 5.56%) and within fascial planes in 5.75% of all studies (canine 2.38%, feline 8.89%). In conclusion, gas bubbles were identified in the majority of canine and feline thoracic CT studies without clinical evidence of a pathologic etiology.

[Primary pulmonary myxoid sarcoma with EWSR1-CREB1 fusion]. / Pervichnaya legochnaya miksoidnaya sarkoma so sliyaniem genov EWSR1-CREB1.

Primary pulmonary myxoid sarcoma with EWSR1-CREB1 fusion is an extremely rare tumor. Its clinical manifestation is unspecific and only molecular genetic method can proof this diagnosis. This paper describes an unusual clinical presentation of primary pulmonary myxoid sarcoma in a 68-year-old patient with involvement of both lungs.

Texture analysis of chest X-ray images for the diagnosis of COVID-19 pneumonia.

Purpose: Medical imaging is one of the main methods of diagnosing COVID-19, along with real-time reverse trans-cription-polymerase chain reaction (RT-PCR) tests. The purpose of the study was to analyse the texture parameters of chest X-rays (CXR) of patients suspected of having COVID-19. Material and methods: Texture parameters of the CXRs of 70 patients with symptoms typical of COVID-19 infection were analysed using LIFEx software. The regions of interest (ROIs) included each lung separately, for which 57 para-meters were tested. The control group consisted of 30 healthy, age-matched patients with no pathological findings in CXRs. Results: According to the ROC analysis, 13 of the tested parameters differentiate the radiological image of lungs with COVID-19 features from the image of healthy lungs: GLRLM_LRHGE (AUC 0.91); DISCRETIZED_Q3 (AUC 0.90); GLZLM_HGZE (AUC 0.90); GLRLM_HGRE (AUC 0.89); DISCRETIZED_mean (AUC 0.89); DISCRETIZED_Q2 (AUC 0.61); GLRLM_SRHGE (AUC 0.87); GLZLM_LZHGE (AUC 0.87); GLZLM_SZHGE (AUC 0.84); DISCRETIZED_Q1 (AUC 0.81); NGLDM_Coarseness (AUC 0.70); DISCRETIZED_std (AUC 0.64); CONVENTIONAL_Q2 (AUC 0.61). Conclusions: Selected texture parameters of radiological CXRs make it possible to distinguish COVID-19 features from healthy ones.

Chronic inflammatory effects of in vivo irradiation of the murine heart on endothelial cells mimic mechanisms involved in atherosclerosis.

PURPOSE: Radiotherapy is a major pillar in the treatment of solid tumors including breast cancer. However, epidemiological studies have revealed an increase in cardiac diseases approximately a decade after exposure of the thorax to ionizing irradiation, which might be related to vascular inflammation. Therefore, chronic inflammatory effects were examined in primary heart and lung endothelial cells (ECs) of mice after local heart irradiation. METHODS: Long-lasting effects on primary ECs of the heart and lung were studied 20-50 weeks after local irradiation of the heart of mice (8 and 16 Gy) in vivo by multiparameter flow cytometry using antibodies directed against cell surface markers related to proliferation, stemness, lipid metabolism, and inflammation, and compared to those induced by occlusion of the left anterior descending coronary artery. RESULTS: In vivo irradiation of the complete heart caused long-lasting persistent upregulation of inflammatory (HCAM, ICAM­1, VCAM-1), proliferation (CD105), and lipid (CD36) markers on primary heart ECs and an upregulation of ICAM­1 and VCAM­1 on primary ECs of the partially irradiated lung lobe. An artificially induced heart infarction induces similar effects with respect to inflammatory markers, albeit in a shorter time period. CONCLUSION: The long-lasting upregulation of prominent inflammatory markers on primary heart and lung ECs suggests that local heart irradiation induces chronic inflammation in the microvasculature of the heart and partially irradiated lung that leads to cardiac injury which might be related to altered lipid metabolism in the heart.

Implementation of artificial intelligence in thoracic imaging-a what, how, and why guide from the European Society of Thoracic Imaging (ESTI).

This statement from the European Society of Thoracic imaging (ESTI) explains and summarises the essentials for understanding and implementing Artificial intelligence (AI) in clinical practice in thoracic radiology departments. This document discusses the current AI scientific evidence in thoracic imaging, its potential clinical utility, implementation and costs, training requirements and validation, its' effect on the training of new radiologists, post-implementation issues, and medico-legal and ethical issues. All these issues have to be addressed and overcome, for AI to become implemented clinically in thoracic radiology. KEY POINTS: • Assessing the datasets used for training and validation of the AI system is essential. • A departmental strategy and business plan which includes continuing quality assurance of AI system and a sustainable financial plan is important for successful implementation. • Awareness of the negative effect on training of new radiologists is vital.

Ultra-high resolution CT imaging of interstitial lung disease: impact of photon-counting CT in 112 patients.

OBJECTIVES: To compare lung parenchyma analysis on ultra-high resolution (UHR) images of a photon-counting CT (PCCT) scanner with that of high-resolution (HR) images of an energy-integrating detector CT (EID-CT). METHODS: A total of 112 patients with stable interstitial lung disease (ILD) were investigated (a) at T0 with HRCT on a 3rd-generation dual-source CT scanner; (b) at T1 with UHR on a PCCT scanner; (c) with a comparison of 1-mm-thick lung images. RESULTS: Despite a higher level of objective noise at T1 (74.1 ± 14.1 UH vs 38.1 ± 8.7 UH; p < 0.0001), higher qualitative scores were observed at T1 with (a) visualization of more distal bronchial divisions (median order; Q1-Q3) (T1: 10th division [9-10]; T0: 9th division [8-9]; p < 0.0001); (b) greater scores of sharpness of bronchial walls (p < 0.0001) and right major fissure (p < 0.0001). The scores of visualization of CT features of ILD were significantly superior at T1 (micronodules: p = 0.03; linear opacities, intralobular reticulation, bronchiectasis, bronchiolectasis, and honeycombing: p < 0.0001), leading to the reclassification of 4 patients with non-fibrotic ILD at T0, recognized with fibrotic ILD at T1. At T1, the mean (± SD) radiation dose (CTDI vol: 2.7 ± 0.5 mGy; DLP: 88.5 ± 21 mGy.cm) was significantly lower than that delivered at T0 (CTDI vol: 3.6 ± 0.9 mGy; DLP: 129.8 ± 31.7 mGy.cm) (p < 0.0001), corresponding to a mean reduction of 27% and 32% for the CTDIvol and DLP, respectively. CONCLUSIONS: The UHR scanning mode of PCCT allowed a more precise depiction of CT features of ILDs and reclassification of ILD patterns with significant radiation dose reduction. CLINICAL RELEVANCE STATEMENT: Evaluation of lung parenchymal structures with ultra-high-resolution makes subtle changes at the level of the secondary pulmonary lobules and lung microcirculation becoming visually accessible, opening new options for synergistic collaborations between highly-detailed morphology and artificial intelligence. KEY POINTS: • Photon-counting CT (PCCT) provides a more precise analysis of lung parenchymal structures and CT features of interstitial lung diseases (ILDs). • The UHR mode ensures a more precise delineation of fine fibrotic abnormalities with the potential of modifying the categorization of ILD patterns. • Better image quality at a lower radiation dose with PCCT opens new horizons for further dose reduction in noncontrast UHR examinations.

Reproducibility of a combined artificial intelligence and optimal-surface graph-cut method to automate bronchial parameter extraction.

OBJECTIVES: Computed tomography (CT)-based bronchial parameters correlate with disease status. Segmentation and measurement of the bronchial lumen and walls usually require significant manpower. We evaluate the reproducibility of a deep learning and optimal-surface graph-cut method to automatically segment the airway lumen and wall, and calculate bronchial parameters. METHODS: A deep-learning airway segmentation model was newly trained on 24 Imaging in Lifelines (ImaLife) low-dose chest CT scans. This model was combined with an optimal-surface graph-cut for airway wall segmentation. These tools were used to calculate bronchial parameters in CT scans of 188 ImaLife participants with two scans an average of 3 months apart. Bronchial parameters were compared for reproducibility assessment, assuming no change between scans. RESULTS: Of 376 CT scans, 374 (99%) were successfully measured. Segmented airway trees contained a mean of 10 generations and 250 branches. The coefficient of determination (R2) for the luminal area (LA) ranged from 0.93 at the trachea to 0.68 at the 6th generation, decreasing to 0.51 at the 8th generation. Corresponding values for Wall Area Percentage (WAP) were 0.86, 0.67, and 0.42, respectively. Bland-Altman analysis of LA and WAP per generation demonstrated mean differences close to 0; limits of agreement (LoA) were narrow for WAP and Pi10 (± 3.7% of mean) and wider for LA (± 16.4-22.8% for 2-6th generations). From the 7th generation onwards, there was a sharp decrease in reproducibility and a widening LoA. CONCLUSION: The outlined approach for automatic bronchial parameter measurement on low-dose chest CT scans is a reliable way to assess the airway tree down to the 6th generation. STATEMENT ON CLINICAL RELEVANCE: This reliable and fully automatic pipeline for bronchial parameter measurement on low-dose CT scans has potential applications in screening for early disease and clinical tasks such as virtual bronchoscopy or surgical planning, while also enabling the exploration of bronchial parameters in large datasets. KEY POINTS: • Deep learning combined with optimal-surface graph-cut provides accurate airway lumen and wall segmentations on low-dose CT scans. • Analysis of repeat scans showed that the automated tools had moderate-to-good reproducibility of bronchial measurements down to the 6th generation airway. • Automated measurement of bronchial parameters enables the assessment of large datasets with less man-hours.

Development of image-based decision support systems utilizing information extracted from radiological free-text report databases with text-based transformers.

OBJECTIVES: To investigate the potential and limitations of utilizing transformer-based report annotation for on-site development of image-based diagnostic decision support systems (DDSS). METHODS: The study included 88,353 chest X-rays from 19,581 intensive care unit (ICU) patients. To label the presence of six typical findings in 17,041 images, the corresponding free-text reports of the attending radiologists were assessed by medical research assistants ("gold labels"). Automatically generated "silver" labels were extracted for all reports by transformer models trained on gold labels. To investigate the benefit of such silver labels, the image-based models were trained using three approaches: with gold labels only (MG), with silver labels first, then with gold labels (MS/G), and with silver and gold labels together (MS+G). To investigate the influence of invested annotation effort, the experiments were repeated with different numbers (N) of gold-annotated reports for training the transformer and image-based models and tested on 2099 gold-annotated images. Significant differences in macro-averaged area under the receiver operating characteristic curve (AUC) were assessed by non-overlapping 95% confidence intervals. RESULTS: Utilizing transformer-based silver labels showed significantly higher macro-averaged AUC than training solely with gold labels (N = 1000: MG 67.8 [66.0-69.6], MS/G 77.9 [76.2-79.6]; N = 14,580: MG 74.5 [72.8-76.2], MS/G 80.9 [79.4-82.4]). Training with silver and gold labels together was beneficial using only 500 gold labels (MS+G 76.4 [74.7-78.0], MS/G 75.3 [73.5-77.0]). CONCLUSIONS: Transformer-based annotation has potential for unlocking free-text report databases for the development of image-based DDSS. However, on-site development of image-based DDSS could benefit from more sophisticated annotation pipelines including further information than a single radiological report. CLINICAL RELEVANCE STATEMENT: Leveraging clinical databases for on-site development of artificial intelligence (AI)-based diagnostic decision support systems by text-based transformers could promote the application of AI in clinical practice by circumventing highly regulated data exchanges with third parties. KEY POINTS: • The amount of data from a database that can be used to develop AI-assisted diagnostic decision systems is often limited by the need for time-consuming identification of pathologies by radiologists. • The transformer-based structuring of free-text radiological reports shows potential to unlock corresponding image databases for on-site development of image-based diagnostic decision support systems. • However, the quality of image annotations generated solely on the content of a single radiology report may be limited by potential inaccuracies and incompleteness of this report.

Fully automatic coronary calcium scoring in non-ECG-gated low-dose chest CT: comparison with ECG-gated cardiac CT.

OBJECTIVES: To validate an artificial intelligence (AI)-based fully automatic coronary artery calcium (CAC) scoring system on non-electrocardiogram (ECG)-gated low-dose chest computed tomography (LDCT) using multi-institutional datasets with manual CAC scoring as the reference standard. METHODS: This retrospective study included 452 subjects from three academic institutions, who underwent both ECG-gated calcium scoring computed tomography (CSCT) and LDCT scans. For all CSCT and LDCT scans, automatic CAC scoring (CAC_auto) was performed using AI-based software, and manual CAC scoring (CAC_man) was set as the reference standard. The reliability and agreement of CAC_auto was evaluated and compared with that of CAC_man using intraclass correlation coefficients (ICCs) and Bland-Altman plots. The reliability between CAC_auto and CAC_man for CAC severity categories was analyzed using weighted kappa (κ) statistics. RESULTS: CAC_auto on CSCT and LDCT yielded a high ICC (0.998, 95% confidence interval (CI) 0.998-0.999 and 0.989, 95% CI 0.987-0.991, respectively) and a mean difference with 95% limits of agreement of 1.3 ± 37.1 and 0.8 ± 75.7, respectively. CAC_auto achieved excellent reliability for CAC severity (κ = 0.918-0.972) on CSCT and good to excellent but heterogenous reliability among datasets (κ = 0.748-0.924) on LDCT. CONCLUSIONS: The application of an AI-based automatic CAC scoring software to LDCT shows good to excellent reliability in CAC score and CAC severity categorization in multi-institutional datasets; however, the reliability varies among institutions. KEY POINTS: • AI-based automatic CAC scoring on LDCT shows excellent reliability with manual CAC scoring in multi-institutional datasets. • The reliability for CAC score-based severity categorization varies among datasets. • Automatic scoring for LDCT shows a higher false-positive rate than automatic scoring for CSCT, and most common causes of a false-positive are image noise and artifacts for both CSCT and LDCT.

Transformer-based structuring of free-text radiology report databases.

OBJECTIVES: To provide insights for on-site development of transformer-based structuring of free-text report databases by investigating different labeling and pre-training strategies. METHODS: A total of 93,368 German chest X-ray reports from 20,912 intensive care unit (ICU) patients were included. Two labeling strategies were investigated to tag six findings of the attending radiologist. First, a system based on human-defined rules was applied for annotation of all reports (termed "silver labels"). Second, 18,000 reports were manually annotated in 197 h (termed "gold labels") of which 10% were used for testing. An on-site pre-trained model (Tmlm) using masked-language modeling (MLM) was compared to a public, medically pre-trained model (Tmed). Both models were fine-tuned on silver labels only, gold labels only, and first with silver and then gold labels (hybrid training) for text classification, using varying numbers (N: 500, 1000, 2000, 3500, 7000, 14,580) of gold labels. Macro-averaged F1-scores (MAF1) in percent were calculated with 95% confidence intervals (CI). RESULTS: Tmlm,gold (95.5 [94.5-96.3]) showed significantly higher MAF1 than Tmed,silver (75.0 [73.4-76.5]) and Tmlm,silver (75.2 [73.6-76.7]), but not significantly higher MAF1 than Tmed,gold (94.7 [93.6-95.6]), Tmed,hybrid (94.9 [93.9-95.8]), and Tmlm,hybrid (95.2 [94.3-96.0]). When using 7000 or less gold-labeled reports, Tmlm,gold (N: 7000, 94.7 [93.5-95.7]) showed significantly higher MAF1 than Tmed,gold (N: 7000, 91.5 [90.0-92.8]). With at least 2000 gold-labeled reports, utilizing silver labels did not lead to significant improvement of Tmlm,hybrid (N: 2000, 91.8 [90.4-93.2]) over Tmlm,gold (N: 2000, 91.4 [89.9-92.8]). CONCLUSIONS: Custom pre-training of transformers and fine-tuning on manual annotations promises to be an efficient strategy to unlock report databases for data-driven medicine. KEY POINTS: • On-site development of natural language processing methods that retrospectively unlock free-text databases of radiology clinics for data-driven medicine is of great interest. • For clinics seeking to develop methods on-site for retrospective structuring of a report database of a certain department, it remains unclear which of previously proposed strategies for labeling reports and pre-training models is the most appropriate in context of, e.g., available annotator time. • Using a custom pre-trained transformer model, along with a little annotation effort, promises to be an efficient way to retrospectively structure radiological databases, even if not millions of reports are available for pre-training.

The yield of chest X-ray or ultra-low-dose chest-CT in emergency department patients suspected of pulmonary infection without respiratory symptoms or signs.

OBJECTIVE: The yield of pulmonary imaging in patients with suspected infection but no respiratory symptoms or signs is probably limited, ultra-low-dose CT (ULDCT) is known to have a higher sensitivity than Chest X-ray (CXR). Our objective was to describe the yield of ULDCT and CXR in patients clinically suspected of infection, but without respiratory symptoms or signs, and to compare the diagnostic accuracy of ULDCT and CXR. METHODS: In the OPTIMACT trial, patients suspected of non-traumatic pulmonary disease at the emergency department (ED) were randomly allocated to undergo CXR (1210 patients) or ULDCT (1208 patients). We identified 227 patients in the study group with fever, hypothermia, and/or elevated C-reactive protein (CRP) but no respiratory symptoms or signs, and estimated ULDCT and CXR sensitivity and specificity in detecting pneumonia. The final day-28 diagnosis served as the clinical reference standard. RESULTS: In the ULDCT group, 14/116 (12%) received a final diagnosis of pneumonia, versus 8/111 (7%) in the CXR group. ULDCT sensitivity was significantly higher than that of CXR: 13/14 (93%) versus 4/8 (50%), a difference of 43% (95% CI: 6 to 80%). ULDCT specificity was 91/102 (89%) versus 97/103 (94%) for CXR, a difference of - 5% (95% CI: - 12 to 3%). PPV was 54% (13/24) for ULDCT versus 40% (4/10) for CXR, NPV 99% (91/92) versus 96% (97/101). CONCLUSION: Pneumonia can be present in ED patients without respiratory symptoms or signs who have a fever, hypothermia, and/or elevated CRP. ULDCT's sensitivity is a significant advantage over CXR when pneumonia has to be excluded. CLINICAL RELEVANCE STATEMENT: Pulmonary imaging in patients with suspected infection but no respiratory symptoms or signs can result in the detection of clinically significant pneumonia. The increased sensitivity of ultra-low-dose chest CT compared to CXR is of added value in vulnerable and immunocompromised patients. KEY POINTS: • Clinical significant pneumonia does occur in patients who have a fever, low core body temperature, or elevated CRP without respiratory symptoms or signs. • Pulmonary imaging should be considered in patients with unexplained symptoms or signs of infections. • To exclude pneumonia in this patient group, ULDCT's improved sensitivity is a significant advantage over CXR.

Staging Chest CT in Patients With Early-Stage Colon Cancer: Analysis of Impact on Survival Using Inverse Probability Weighting and Causal Diagram.

BACKGROUND. Staging chest CT has been shown to have negligible diagnostic yield for detecting lung metastases in patients with early-stage colon cancer. Nonetheless, staging chest CT may have potential survival benefits, including opportunistic screening of comorbidity and provision of a baseline examination for future comparisons. Evidence is lacking regarding the impact of staging chest CT on survival in patients with early-stage colon cancer. OBJECTIVE. The purpose of this study was to determine whether the performance of staging chest CT affects survival in patients with early-stage colon cancer. METHODS. This retrospective study included patients with early-stage colon cancer (defined as clinical stage 0 or I on staging abdominal CT) at a single tertiary hospital between January 2009 and December 2015. Patients were divided into two groups according to the presence of a staging chest CT examination. To ensure comparability between the two groups, inverse probability weighting was applied to adjust for the confounders derived from a causal diagram. The between-group differences in adjusted restricted mean survival time at 5 years were measured for overall survival, relapse-free survival, and thoracic metastasis-free survival. Sensitivity analyses were performed. RESULTS. A total of 991 patients (618 men and 373 women; median age, 64 years [IQR, 55-71 years]) were included: 606 patients (61.2%) had staging chest CT. For overall survival, the difference between groups in restricted mean survival time at 5 years was not significant (0.4 months [95% CI, -0.8 to 2.1 months]). The differences between groups in restricted mean survival at 5 years were also not significant for relapse-free survival (0.4 months [95% CI, -1.1 to 2.3 months]) and for thoracic metastasis-free survival (0.6 months [95% CI, -0.8 to 2.4 months]). Similar results were observed in sensitivity analyses that tested 3- and 10-year RMST differences, excluded patients who underwent FDG PET/CT during staging workup, and added treatment decision (surgery vs no surgery) to the causal diagram. CONCLUSION. The use of staging chest CT did not affect survival in patients with early-stage colon cancer. CLINICAL IMPACT. Staging chest CT may be omitted from the staging workup for patients with colon cancer of clinical stage 0 or I.

CT after emergency surgery in penetrating trauma: a seven-year experience in a level I Nordic trauma center.

BACKGROUND: Patients with severe penetrating trauma may require emergency surgery on arrival, and postoperative computed tomography (CT) can reveal significant additional injuries. PURPOSE: To determine the utility of postoperative CT performed within 48 h of emergency surgery after penetrating trauma. MATERIAL AND METHODS: Trauma registry data were retrieved over a seven-year period at a single level 1 trauma center. All patients aged ≥17 years, admitted with penetrating injury, who underwent urgent surgery and postoperative CT imaging within 48 h, were included. Pre- and intraoperative medical records were compared to CT findings. Age, sex, Injury Severity Score (ISS), New Injury Severity Score (NISS), 30-day mortality, injury mechanism, surgical intervention, and intensive care unit length of stay were extracted. RESULTS: Out of 1262 patients, 38 fulfilled the study criteria (36 men [94.7%], 2 women [5.3%]; mean age = 31.5 years. Stab wound (SW) was the most common injury mechanism (26/38, 68.4%) followed by gunshot wound (GSW; 10/38, 26.3%). Patients with GSWs were more severely injured than SW victims (median NISS = 34 [range = 3-75]; for GSWs = 34; for SWs = 26; P = 0.045). Out of 38 patients, 20 (52.6%) had additional findings at postoperative CT. Six patients (15.8%) had unidentified or underestimated findings at CT that were severe enough to warrant additional surgery or angiography. CONCLUSION: Postoperative CT imaging after emergency surgery in penetrating trauma is an important tool in evaluating the injury panorama. Out of 38 patients, 6 (15.8%) had findings at postoperative CT that warranted additional surgical or angiographic intervention.