Refining diagnostic criteria for paediatric bronchiectasis using low-dose CT scan

Radiographic Outcomes in Pediatric Bronchiectasis and Factors Associated with Reversibility

Rationale: Conventionally considered irreversible, bronchiectasis has been demonstrated to be reversible in children in small studies. However, the factors associated with radiographic reversibility of bronchiectasis have yet to be defined. Objectives: In a large cohort of children with bronchiectasis, we aimed to determine: 1) if and to what extent bronchiectasis is reversible and 2) factors associated with radiographic chest high-resolution computed tomography (cHRCT) resolution. Methods: We identified children with bronchiectasis who had a repeat multidetector cHRCT scan between 2010 and 2021. We excluded those with cystic fibrosis, surgical pulmonary resection, traction bronchiectasis only, or lobar opacification. Measurements and Main Results: cHRCT scans were scored using the modified Reiff score (MRS) with a pediatric correction. Resolution was defined as an absence of abnormal bronchoarterial ratio (>0.8) on the second cHRCT scan. We included 142 children (median age, 5 years; IQR, 2.6-7.4). Inter- and intrarater agreement in MRSs was excellent (weighted κ = 0.83-0.86 and 0.95, respectively). There was radiographic resolution in 57 of 142 patients (40.1%), improvement in 56 of 142 (39.4%), and no change or worsening in 29 of 142 (20.4%). Pseudomonas aeruginosa (PsA) was absolutely associated with a lack of resolution. On multivariable regression, in those without PsA cultured, younger age at the time of diagnosis (risk ratio [RR], 0.94; 95% confidence interval [CI], 0.88-0.99), lower MRS (RR, 0.89; 95% CI, 0.82-0.97), and lower annual rate of exacerbations requiring intravenous antibiotic therapy (RR, 0.60; 95% CI, 0.37-0.98) increased the likelihood of radiographic resolution. Conclusions: This first large cohort confirms that bronchiectasis in children is often reversible with appropriate management. Younger children and those with lesser radiographic severity at diagnosis were most likely to exhibit radiographic reversibility, whereas those with PsA infection were least likely.

Improving the Diagnosis and Treatment of Paediatric Bronchiectasis Through Research and Translation

Bronchiectasis, particularly in children, is an increasingly recognised yet neglected chronic lung disorder affecting individuals in both low-to-middle and high-income countries. It has a high disease burden and there is substantial inequity within and between settings. Furthermore, compared with other chronic lung diseases, considerably fewer resources are available for children with bronchiectasis. The need to prevent bronchiectasis and to reduce its burden also synchronously aligns with its high prevalence and economic costs to health services and society. Like many chronic lung diseases, bronchiectasis often originates early in childhood, highlighting the importance of reducing the disease burden in children. Concerted efforts are therefore needed to improve disease detection, clinical management and equity of care. Modifiable factors in the causal pathways of bronchiectasis, such as preventing severe and recurrent lower respiratory infections should be addressed, whilst also acknowledging the role played by social determinants of health. Here, we highlight the importance of early recognition/detection and optimal management of bronchiectasis in children, and outline our research, which is attempting to address important clinical knowledge gaps discussed in a recent workshop. The research is grouped under three themes focussing upon primary prevention, improving diagnosis and disease characterisation, and providing better management. Our hope is that others in multiple settings will undertake additional studies in this neglected field to further improve the lives of people with bronchiectasis. We also provide a resource list with links to help inform consumers and healthcare professionals about bronchiectasis and its recognition and management.

An Automated Broncho-Arterial (BA) Pair Segmentation Process and Assessment of BA Ratios in Children with Bronchiectasis Using Lung HRCT Scans: A Pilot Study

Bronchiectasis in children can progress to a severe lung condition if not diagnosed and treated early. The radiological diagnostic criteria for the diagnosis of bronchiectasis is an increased broncho-arterial (BA) ratio. From high-resolution computed tomography (HRCT) scans, the BA pairs must be detected first to derive the BA ratio. This study aims to identify potential BA pairs from HRCT scans of children undertaken to evaluate suppurative lung disease through an automated approach. After segmenting the lung regions, the HRCT scans are cleaned using a histogram analysis-based approach followed by a potential arteries identification process comprising four conditions based on imaging features. Potential arteries and their connected components are extracted, and potential bronchi are identified. Finally, the coordinates of potential arteries and potential bronchi are matched as the last step of BA pairs extraction. A total of 8-50 BA pairs are detected for each patient. Additionally, the area and several diameters of the bronchi and arteries are measured, and BA ratios based on these are calculated. Through this approach, the BA pairs of a CT scan datasets are detected and utilizing a deep learning model, a high classification test accuracy of 98.53% is achieved, validating the robustness of the proposed BA detection approach. The results show that visible BA pairs can be identified and segmented automatically, and the BA ratio calculated may help diagnose bronchiectasis with less effort and time.

Hip Imaging in Children With Cerebral Palsy: Estimation and Intrapatient Comparison of Patient-Specific Radiation Doses of Low-Dose CT and Radiography

OBJECTIVES

Hip displacement is the second most common orthopedic problem affecting children with cerebral palsy (CP). Routine radiographic hip surveillance typically involves an anteroposterior (AP) pelvis radiograph. Unfortunately, this imaging protocol is limited by its projectional technique and the positioning challenges in children with CP. Alternatively, hip low-dose computed tomography (LDCT) has been advocated as a more accurate strategy for imaging surveillance as it provides biofidelic details of the hip that is independent of patient positioning. However, the tradeoff is the (presumed) higher radiation dose to the patient. The goal of this study is to estimate patient-specific radiation doses of hip LDCTs and AP pelvis radiographs in CP patients, and perform an intrapatient dose comparison.

MATERIALS AND METHODS

A search of our imaging database was performed to identify children with CP who underwent hip LDCT and AP pelvis radiograph within 6 months of each other. The LDCTs were performed using weight-adjusted kVp and tube current modulation, whereas the radiographs were obtained with age-/size-adjusted kVp/mAs. The patient-specific organ and effective doses for LDCT were estimated by matching the patients to a nonreference pediatric phantom library from the National Cancer Institute Dosimetry System for Computed Tomography database with Monte Carlo-based dosimetry. The patient-specific organ and effective doses for radiograph were estimated using the National Cancer Institute Dosimetry System for Radiography and Fluoroscopy with Monte Carlo-based dose calculation. Dose conversion k-factors of dose area product for radiography and dose length product for LDCT were adapted, and the estimation results were compared with patient-specific dosimetry.

RESULTS

Our study cohort consisted of 70 paired imaging studies from 67 children (age, 9.1 ± 3.3 years). The patient-specific and dose length product-based effective doses for LDCT were 0.42 ± 0.21 mSv and 0.59 ± 0.28 mSv, respectively. The patient-specific and dose area product-based effective doses for radiography were 0.14 ± 0.09 mSv and 0.08 ± 0.06 mSv, respectively.

CONCLUSIONS

The radiation dose for a hip LDCT is ~4 times higher than pelvis radiograph, but it is still very low and poses minimal risk to the patient.

Automated Detection of Broncho-Arterial Pairs Using CT Scans Employing Different Approaches to Classify Lung Diseases

Current research indicates that for the identification of lung disorders, comprising pneumonia and COVID-19, structural distortions of bronchi and arteries (BA) should be taken into account. CT scans are an effective modality to detect lung anomalies. However, anomalies in bronchi and arteries can be difficult to detect. Therefore, in this study, alterations of bronchi and arteries are considered in the classification of lung diseases. Four approaches to highlight these are introduced: (a) a Hessian-based approach, (b) a region-growing algorithm, (c) a clustering-based approach, and (d) a color-coding-based approach. Prior to this, the lungs are segmented, employing several image preprocessing algorithms. The utilized COVID-19 Lung CT scan dataset contains three classes named Non-COVID, COVID, and community-acquired pneumonia, having 6983, 7593, and 2618 samples, respectively. To classify the CT scans into three classes, two deep learning architectures, (a) a convolutional neural network (CNN) and (b) a CNN with long short-term memory (LSTM) and an attention mechanism, are considered. Both these models are trained with the four datasets achieved from the four approaches. Results show that the CNN model achieved test accuracies of 88.52%, 87.14%, 92.36%, and 95.84% for the Hessian, the region-growing, the color-coding, and the clustering-based approaches, respectively. The CNN with LSTM and an attention mechanism model results in an increase in overall accuracy for all approaches with an 89.61%, 88.28%, 94.61%, and 97.12% test accuracy for the Hessian, region-growing, color-coding, and clustering-based approaches, respectively. To assess overfitting, the accuracy and loss curves and k-fold cross-validation technique are employed. The Hessian-based and region-growing algorithm-based approaches produced nearly equivalent outcomes. Our proposed method outperforms state-of-the-art studies, indicating that it may be worthwhile to pay more attention to BA features in lung disease classification based on CT images.

The effect of a training webinar on decreasing inter-observer variability in making a radiologic diagnosis of bronchiectasis

Non-cystic fibrosis bronchiectasis is a clinically important disease with an estimated 340,000–522,000 persons living with the disease and 70,000 being diagnosed annually. The radiographic diagnosis remains a pivotal part of recognizing the disease due to its protean clinical manifestations. As physicians are sensitized to this disease, a greater proportion of patients are being diagnosed with mild to moderate bronchiectasis. Despite the established use of CT chest as the main tool for making a radiologic diagnosis of bronchiectasis, the literature supporting the process of making that diagnosis is somewhat sparse. Concurrently, there has been an increased trend to have Web-based radiologic tutorials due to its convenience, the ability of the learner to set the pace of learning and the reduced cost compared to in-person learning. The COVID-19 pandemic has accelerated this trend. We wanted to look carefully at the effect of a Web-based training session on interrater reliability. Agreement was calculated as percentages and kappa and prevalence adjusted kappa calculated. We found that a single Web-based training session had little effect on the variability and accuracy of diagnosis of bronchiectasis. Larger studies are needed in this area with multiple training sessions.