Positional contrastive learning for improved thigh muscle segmentation in MR images.

The accurate segmentation of individual muscles is essential for quantitative MRI analysis of thigh images. Deep learning methods have achieved state-of-the-art results in segmentation, but they require large numbers of labeled data to perform well. However, labeling individual thigh muscles slice by slice for numerous volumes is a laborious and time-consuming task, which limits the availability of annotated datasets. To address this challenge, self-supervised learning (SSL) emerges as a promising technique to enhance model performance by pretraining the model on unlabeled data. A recent approach, called positional contrastive learning, exploits the information given by the axial position of the slices to learn features transferable on the segmentation task. The aim of this work was to propose positional contrastive SSL for the segmentation of individual thigh muscles from MRI acquisitions in a population of elderly healthy subjects and to evaluate it on different levels of limited annotated data. An unlabeled dataset of 72 T1w MRI thigh acquisitions was available for SSL pretraining, while a labeled dataset of 52 volumes was employed for the final segmentation task, split into training and test sets. The effectiveness of SSL pretraining to fine-tune a U-Net architecture for thigh muscle segmentation was compared with that of a randomly initialized model (RND), considering an increasing number of annotated volumes (S = 1, 2, 5, 10, 20, 30, 40). Our results demonstrated that SSL yields substantial improvements in Dice similarity coefficient (DSC) when using a very limited number of labeled volumes (e.g., for S $$ S $$ = 1, DSC 0.631 versus 0.530 for SSL and RND, respectively). Moreover, enhancements are achievable even when utilizing the full number of labeled subjects, with DSC = 0.927 for SSL and 0.924 for RND. In conclusion, positional contrastive SSL was effective in obtaining more accurate thigh muscle segmentation, even with a very low number of labeled data, with a potential impact of speeding up the annotation process in clinics.

The impact of muscle mass and myosteatosis on mortality in critically ill patients with Sars-Cov2-related pneumonia.

BACKGROUND & AIMS: Sars-Cov-2 pneumonia can lead to severe complications, requiring invasive mechanical ventilation and admission to the intensive care unit (ICU). Low muscle quantity and quality (the latter evaluated by the amount of ectopic fat infiltration in the muscle [myosteatosis]) at ICU admission are associated with worse outcomes in critically ill patients. The purpose of the present study is to assess muscle mass and myosteatosis of paravertebral skeletal muscle, in critically ill patients with Sars-Cov2 pneumonia and its association with mortality. METHODS: We conducted a retrospective observational study in 110 critically ill patients with severe Sars-Cov-2 pneumonia that had a high - resolution chest Computerized Tomography (HR-CT) at ICU admission. We acquired CT images at the level of the thoracic 12 (T12) vertebral body and measured skeletal muscle area (SMA), intermuscular adipose tissue (IMAT), and low attenuation muscle area (LAMA). Patients were followed until ICU mortality or discharge. RESULTS: Patients were 59.8 ± 8.1 years old, 77% were male. Seventy-nine percent of patients were considered at nutritional risk, and 22% were obese. Average Acute Physiology and Chronic Health Evaluation II (APACHE II) score was 17 ± 5.4, and the overall ICU mortality was 48,2% (53/110). At ICU admission, both parameters of myosteatosis were associated with higher mortality (IMAT [per 10% increase] HR: 2.01 (95% Confidence Interval [CI] 1.27 to 3.17), P = 0.003; LAMA HR [per 10% increase]: 1.53 (95% CI 1.10 to 2.13), P = 0.012). CONCLUSION: Myosteatosis as assessed by CT scans plays a relevant role as a prognostic marker in critically ill patients with Sars-Cov2 severe pneumonia.

Structural changes in lung morphology detected by MRI after modulating therapy with elexacaftor/tezacaftor/ivacaftor in adolescent and adult patients with cystic fibrosis.

INTRODUCTION: Elexacaftor/tezacaftor/ivacaftor (ELX/TEZ/IVA) improves CFTR function in cystic fibrosis (CF) patients homozygous or heterozygous for F508del mutation. The aim of the study was to evaluate the response to ELX/TEZ/IVA treatment both clinically and morphologically in terms of bronchiectasis, bronchial wall thickening, mucus plugging, abscess and consolidations. METHODS: We retrospectively collected data from CF patients followed at Parma CF Centre (Italy) treated by ELX/TEZ/IVA between March and November 2021. Post-treatment changes in respiratory function, quality of life, sweat chloride concentration, body mass index, pulmonary exacerbations and lung structure by chest magnetic resonance imaging (MRI) were assessed. T2-and T1-weighted sequences were acquired with a 20 min-long scanning protocol on a 1.5T MRI scanner (Philips Ingenia) without administration of intravenous contrast media. RESULTS: 19 patients (32.5 ± 10.2 years) were included in the study. After 6 months of treatment with ELX/TEZ/IVA, MRI showed significant improvements in the morphological score (p < 0.001), with a reduction in bronchial wall thickening (p < 0.001) and mucus plugging (p 0.01). Respiratory function showed significant improvement in predicted FEV1% (58.5 ± 17.5 vs 71.4 ± 20.1, p < 0.001), FVC% (79.0 ± 11.1 vs 88.3 ± 14.4, p < 0.001), FEV1/FVC (0.61 ± 0.16 vs 0.67 ± 0.15, <0.001) and LCI2.5% (17.8 ± 4.3 vs 15.8 ± 4.1 p < 0.005). Significant improvement was found in body mass index (20.6 ± 2.7 vs 21.9 ± 2.4, p < 0.001), pulmonary exacerbations (2.3 ± 1.3 vs 1.4 ± 1.3 p 0.018) and sweat chloride concentration (96.5 ± 36.6 vs 41.1 ± 16.9, p < 0.001). CONCLUSIONS: Our study confirms the efficacy of ELX/TEZ/IVA in CF patients not only from a clinical point of view but also in terms of morphological changes of the lungs.

Clinical Impact of Aspergillus fumigatus in Children with Cystic Fibrosis.

Background. The clinical relevance of Aspergillus fumigatus (Af) in cystic fibrosis (CF) is controversial. The aims of the study were to assess the prevalence of Af disease in our cohort of CF patients and evaluate whether allergic bronchopulmonary aspergillosis (ABPA) and sensitization to Af affected lung function, body mass index (BMI) and exacerbations. Methods. Clinical data and lung function of CF patients aged 6−18 years followed at the CF Centre of Parma (Italy) were recorded. Patients were classified as: patients with no signs of Af, patients sensitized or colonized by Af, patients with ABPA or patients with Aspergillus bronchitis (Ab). Results. Of 38 CF patients (14.2 years (6.2−18.8) M 23), 8 (21%) showed Af sensitization, 7 (18.4%) showed ABPA, 1 (2.6%) showed Af colonization and 1 (2.6%) showed Ab. Compared to non-ABPA, patients with ABPA had lower BMI (15.9 ± 1.6 vs. 19.7 ± 3.4, p < 0.005), lower lung function (FEV1 61.5 ± 25.9% vs. 92.3 ± 19.3%, p < 0.001) and more exacerbations/year (4.43 ± 2.44 vs. 1.74 ± 2.33, p < 0.005). Patients with Af sensitization showed more exacerbations/year than non-Af patients (3.5 ± 3.2 vs. 0.9 ± 1.2, p < 0.005). ABPA and sensitized patients had more abnormalities on chest CT scans. Conclusion. This study showed the relevant clinical impact of ABPA and Af sensitization in terms of exacerbations and lung structural damage.

Detection and Classification of Bronchiectasis Through Convolutional Neural Networks.

PURPOSE: Bronchiectasis is a chronic disease characterized by an irreversible dilatation of bronchi leading to chronic infection, airway inflammation, and progressive lung damage. Three specific patterns of bronchiectasis are distinguished in clinical practice: cylindrical, varicose, and cystic. The predominance and the extension of the type of bronchiectasis provide important clinical information. However, characterization is often challenging and is subject to high interobserver variability. The aim of this study is to provide an automatic tool for the detection and classification of bronchiectasis through convolutional neural networks. MATERIALS AND METHODS: Two distinct approaches were adopted: (i) direct network performing a multilabel classification of 32×32 regions of interest (ROIs) into 4 classes: healthy, cylindrical, cystic, and varicose and (ii) a 2-network serial approach, where the first network performed a binary classification between normal tissue and bronchiectasis and the second one classified the ROIs containing abnormal bronchi into one of the 3 bronchiectasis typologies. Performances of the networks were compared with other architectures presented in the literature. RESULTS: Computed tomography from healthy individuals (n=9, age=47±6, FEV1%pred=109±17, FVC%pred=116±17) and bronchiectasis patients (n=21, age=59±15, FEV1%pred=74±25, FVC%pred=91±22) were collected. A total of 19,059 manually selected ROIs were used for training and testing. The serial approach provided the best results with an accuracy and F1 score average of 0.84, respectively. Slightly lower performances were observed for the direct network (accuracy=0.81 and F1 score average=0.82). On the test set, cylindrical bronchiectasis was the subtype classified with highest accuracy, while most of the misclassifications were related to the varicose pattern, mainly to the cylindrical class. CONCLUSION: The developed networks accurately detect and classify bronchiectasis disease, allowing to collect quantitative information regarding the radiologic severity and the topographical distribution of bronchiectasis subtype.

COVID-19 in Children: Update on Diagnosis and Management.

In December 2019, a new infectious disease called coronavirus disease 2019 (COVID-19) attributed to the new virus named severe scute respiratory syndrome coronavirus 2 (SARS-CoV-2) was detected. The gold standard for the diagnosis of SARS-CoV-2 infection is the viral identification in nasopharyngeal swab by real-time polymerase chain reaction. Few data on the role of imaging are available in the pediatric population. Similarly, considering that symptomatic therapy is adequate in most of the pediatric patients with COVID-19, few pediatric pharmacological studies are available. The main aim of this review is to describe and discuss the scientific literature on various imaging approaches and therapeutic management in children and adolescents affected by COVID-19. Clinical manifestations of COVID-19 are less severe in children than in adults and as a consequence the radiologic findings are less marked. If imaging is needed, chest radiography is the first imaging modality of choice in the presence of moderate-to-severe symptoms. Regarding therapy, acetaminophen or ibuprofen are appropriate for the vast majority of pediatric patients. Other drugs should be prescribed following an appropriate individualized approach. Due to the characteristics of COVID-19 in pediatric age, the importance of strengthening the network between hospital and territorial pediatrics for an appropriate diagnosis and therapeutic management represents a priority.

Imaging in non-cystic fibrosis bronchiectasis and current limitations.

Non-cystic fibrosis bronchiectasis represents a heterogenous spectrum of disorders characterised by an abnormal and permanent dilatation of the bronchial tree associated with respiratory symptoms. To date, diagnosis relies on computed tomography (CT) evidence of dilated airways. Nevertheless, definite radiological criteria and standardised CT protocols are still to be defined. Although largely used, current radiological scoring systems have shown substantial drawbacks, mostly failing to correlate morphological abnormalities with clinical and prognostic data. In limited cases, bronchiectasis morphology and distribution, along with associated CT features, enable radiologists to confidently suggest an underlying cause. Quantitative imaging analyses have shown a potential to overcome the limitations of the current radiological criteria, but their application is still limited to a research setting. In the present review, we discuss the role of imaging and its current limitations in non-cystic fibrosis bronchiectasis. The potential of automatic quantitative approaches and artificial intelligence in such a context will be also mentioned.

Chest radiography or computed tomography for COVID-19 pneumonia? Comparative study in a simulated triage setting.

INTRODUCTION: For the management of patients referred to respiratory triage during the early stages of the severe acute respiratory syndrome coronavirus type 2 (SARS-CoV-2) pandemic, either chest radiography or computed tomography (CT) were used as first-line diagnostic tools. The aim of this study was to compare the impact on the triage, diagnosis and prognosis of patients with suspected COVID-19 when clinical decisions are derived from reconstructed chest radiography or from CT. METHODS: We reconstructed chest radiographs from high-resolution CT (HRCT) scans. Five clinical observers independently reviewed clinical charts of 300 subjects with suspected COVID-19 pneumonia, integrated with either a reconstructed chest radiography or HRCT report in two consecutive blinded and randomised sessions: clinical decisions were recorded for each session. Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and prognostic value were compared between reconstructed chest radiography and HRCT. The best radiological integration was also examined to develop an optimised respiratory triage algorithm. RESULTS: Interobserver agreement was fair (Kendall's W=0.365, p<0.001) by the reconstructed chest radiography-based protocol and good (Kendall's W=0.654, p<0.001) by the CT-based protocol. NPV assisted by reconstructed chest radiography (31.4%) was lower than that of HRCT (77.9%). In case of indeterminate or typical radiological appearance for COVID-19 pneumonia, extent of disease on reconstructed chest radiography or HRCT were the only two imaging variables that were similarly linked to mortality by adjusted multivariable models CONCLUSIONS: The present findings suggest that clinical triage is safely assisted by chest radiography. An integrated algorithm using first-line chest radiography and contingent use of HRCT can help optimise management and prognostication of COVID-19.

Renal artery embolization of non-functioning graft: an effective treatment for graft intolerance syndrome.

BACKGROUND: Percutaneous renal artery embolization is a valid non-invasive technique alternative to nephrectomy for patients with symptomatic non-functioning allograft (graft intolerance syndrome-GIS). The purpose of this article is to report the experience of our centre. METHODS: We analysed retrospectively 15 patients with symptomatic non-functioning renal allograft treated with percutaneous embolization from 2003 to 2017. Occlusion was obtained with the injection of calibrated microspheres of increasing size (from 100 to 900 µm) and completed with 5 to 8 mm metal coils placement in the renal artery. RESULTS: Technical success was achieved in all cases at the end of the procedure. Clinical success was obtained in 11 patients (73%). In four cases, nephrectomy was necessary: in one case because of septic fever and in three cases because of GIS persistence. In one case, it was possible to perform another procedure to embolize a perirenal collateral from a lumbar artery. Four patients (27%) reported minor complications which spontaneously resolved during the hospital stay. CONCLUSIONS: According to the scientific literature, we believe that, in selected patients, percutaneous renal artery embolization is a valid treatment option for GIS thanks to its efficacy, repeatability, minimal invasiveness and the absence of severe complications.

Validation by CT scan of quadriceps muscle thickness measurement by ultrasound in acute kidney injury.

BACKGROUND: Accelerated muscle wasting still represents a major issue in critically ill patients. However, a key problem in the intensive care unit is the lack of adequate tools for bedside evaluation of muscle mass. Moreover, when acute kidney injury (AKI) coexists, fluid overload and/or rapid fluid shifts due to renal replacement therapies that frequently occur and may interfere with muscle mass assessment. The purpose of this study is to validate muscle ultrasound (US) by a gold standard (muscle CT scan) for the assessment of quadriceps muscle thickness in critically ill patients with AKI. METHODS: Quadriceps rectus femoris thickness and quadriceps vastus intermedius thickness of critically ill patients with AKI were blindly assessed at the same leg sites by both US and computed tomography (CT) scan. Using bivariate mixed-model linear regression analysis, we estimated, average difference in thickness between measurement sites, agreement (differential and proportional bias) of US compared to CT, and precision of the two methods, and eventually performed Bland-Altman analysis for repeated measurements on pooled results. RESULTS: We analyzed 233 couples of measurements (30 patients). Average muscle thickness ranged between 1.0 and 1.6, depending on the measurement site. When comparing US to CT, both the observed differential bias (between + 0.04 and + 0.26 cm depending on the muscle site) and the proportional bias (between 82 and 98% of the reference values, depending on the muscle site) were not statistically significant. However, precision analysis showed that US scan tended to be slightly less precise in comparison to CT. Bland-Altman analysis on pooled results showed that the 95% limits of agreement between the US and CT were narrow, ranging from - 0.34 to + 0.36 cm. CONCLUSION: In critically ill patients with AKI, quadriceps muscle thickness assessment based on US is unbiased, although it occurs with a minor loss of precision compared to CT.