State of the Art in Lung Ultrasound, Shifting from Qualitative to Quantitative Analyses

FLUEnT: Transformer for detecting lung consolidations in videos using fused lung ultrasound encodings

Pneumonia is the leading cause of death among children around the world. According to WHO, a total of 740,180 lives under the age of five were lost due to pneumonia in 2019. Lung ultrasound (LUS) has been shown to be particularly useful for supporting the diagnosis of pneumonia in children and reducing mortality in resource-limited settings. The wide application of point-of-care ultrasound at the bedside is limited mainly due to a lack of training for data acquisition and interpretation. Artificial Intelligence can serve as a potential tool to automate and improve the LUS data interpretation process, which mainly involves analysis of hyper-echoic horizontal and vertical artifacts, and hypo-echoic small to large consolidations. This paper presents, Fused Lung Ultrasound Encoding-based Transformer (FLUEnT), a novel pediatric LUS video scoring framework for detecting lung consolidations using fused LUS encodings. Frame-level embeddings from a variational autoencoder, features from a spatially attentive ResNet-18, and encoded patient information as metadata combiningly form the fused encodings. These encodings are then passed on to the transformer for binary classification of the presence or absence of consolidations in the video. The video-level analysis using fused encodings resulted in a mean balanced accuracy of 89.3 %, giving an average improvement of 4.7 % points in comparison to when using these encodings individually. In conclusion, outperforming the state-of-the-art models by an average margin of 8 % points, our proposed FLUEnT framework serves as a benchmark for detecting lung consolidations in LUS videos from pediatric pneumonia patients.

Knowledge fused latent representation from lung ultrasound examination for COVID-19 pneumonia severity assessment

COVID-19 pneumonia severity assessment is of great clinical importance, and lung ultrasound (LUS) plays a crucial role in aiding the severity assessment of COVID-19 pneumonia due to its safety and portability. However, its reliance on qualitative and subjective observations by clinicians is a limitation. Moreover, LUS images often exhibit significant heterogeneity, emphasizing the need for more quantitative assessment methods. In this paper, we propose a knowledge fused latent representation framework tailored for COVID-19 pneumonia severity assessment using LUS examinations. The framework transforms the LUS examination into latent representation and extracts knowledge from regions labeled by clinicians to improve accuracy. To fuse the knowledge into the latent representation, we employ a knowledge fusion with latent representation (KFLR) model. This model significantly reduces errors compared to approaches that lack prior knowledge integration. Experimental results demonstrate the effectiveness of our method, achieving high accuracy of 96.4 % and 87.4 % for binary-level and four-level COVID-19 pneumonia severity assessments, respectively. It is worth noting that only a limited number of studies have reported accuracy for clinically valuable exam level assessments, and our method surpass existing methods in this context. These findings highlight the potential of the proposed framework for monitoring disease progression and patient stratification in COVID-19 pneumonia cases.

Tracking Ovine Pulmonary Adenocarcinoma Development Using an Experimental Jaagsiekte Sheep Retrovirus Infection Model

Ovine pulmonary adenocarcinoma (OPA) is an infectious, neoplastic lung disease of sheep that causes significant animal welfare and economic issues throughout the world. Understanding OPA pathogenesis is key to developing tools to control its impact. Central to this need is the availability of model systems that can monitor and track events after Jaagsiekte sheep retrovirus (JSRV) infection. Here, we report the development of an experimentally induced OPA model intended for this purpose. Using three different viral dose groups (low, intermediate and high), localised OPA tumour development was induced by bronchoscopic JSRV instillation into the segmental bronchus of the right cardiac lung lobe. Pre-clinical OPA diagnosis and tumour progression were monitored by monthly computed tomography (CT) imaging and trans-thoracic ultrasound scanning. Post mortem examination and immunohistochemistry confirmed OPA development in 89% of the JSRV-instilled animals. All three viral doses produced a range of OPA lesion types, including microscopic disease and gross tumours; however, larger lesions were more frequently identified in the low and intermediate viral groups. Overall, 31% of JSRV-infected sheep developed localised advanced lesions. Of the sheep that developed localised advanced lesions, tumour volume doubling times (calculated using thoracic CT 3D reconstructions) were 14.8 ± 2.1 days. The ability of ultrasound to track tumour development was compared against CT; the results indicated a strong significant association between paired CT and ultrasound measurements at each time point (R2 = 0.799, p < 0.0001). We believe that the range of OPA lesion types induced by this model replicates aspects of naturally occurring disease and will improve OPA research by providing novel insights into JSRV infectivity and OPA disease progression.

Visual lung ultrasound protocol (VLUP) in acute respiratory failure: description and application in clinical cases

Lung ultrasound (LUS) is widely used as a diagnostic and monitoring tool in critically ill patients. Lung ultrasound score (LUSS) based on the examination of twelve thoracic regions has been extensively validated for pulmonary assessment. However, it has revealed significant limitations: when applied to heterogeneous lung diseases with intermediate LUSS pattern (LUSS 1 and 2), for instance, intra-observer consistency is relatively low. In addition, LUSS is time-consuming and a more rapid overview of the extent of lung pathology and residual lung aeration is often required, especially in emergency setting. We propose a Visual Lung Ultrasound Protocol (VLUP) as a rapid monitoring tool for patients with acute respiratory failure. It consists of a probe sliding along the mid-clavicular, mid-axillary and scapular lines in transversal scan. VLUP allows a visualization of a large portion of the antero-lateral and/or posterior pleural surface. Serial assessments of two clinical cases are recorded and visually compared, enabling rapid understanding of lung damage and its evolution over time. VLUP allows a semi-quantitative and qualitative point-of-care assessment of lung injury. Through this standardized approach it is possible to accurately compare subsequent scans and to monitor the evolution of regional parenchymal damage. VLUP enables a quick estimation of the quantitative-LUSS (qLUSS) as the percentage of pleura occupied by artifacts, more suitable than LUSS in inhomogeneous diseases. VLUP is designed as a standardized, point-of-care lung aeration assessment and monitoring tool. The purpose of the paper is to illustrate this new technique and to describe its applications.

A machine learning-based lung ultrasound algorithm for the diagnosis of acute heart failure

Lung ultrasound (LUS) is an effective tool for diagnosing acute heart failure (AHF). However, several imaging protocols currently exist and how to best use LUS remains undefined. We aimed at developing a lung ultrasound-based model for AHF diagnosis using machine learning. Random forest and decision trees were generated using the LUS data (via an 8-zone scanning protocol) in patients with acute dyspnea admitted to the Emergency Department (PLUME study, N = 117) and subsequently validated in an external dataset (80 controls from the REMI study, 50 cases from the Nancy AHF cohort). Using the random forest model, total B-line sum (i.e., in both hemithoraces) was the most significant variable for identifying AHF, followed by the difference in B-line sum between the superior and inferior lung areas. The decision tree algorithm had a good diagnostic accuracy [area under the curve (AUC) = 0.865] and identified three risk groups (i.e., low 24%, high 70%, and very high-risk 96%) for AHF. The very high-risk group was defined by the presence of 14 or more B-lines in both hemithoraces while the high-risk group was described as having either B-lines mostly localized in superior points or in the right hemithorax. Accuracy in the validation cohort was excellent (AUC = 0.906). Importantly, adding the algorithm on top of a validated clinical score and classical definition of positive LUS scanning for AHF resulted in a significant improvement in diagnostic accuracy (continuous net reclassification improvement = 1.21, P < 0.001). Our simple lung ultrasound-based machine learning algorithm features an excellent performance and may constitute a validated strategy to diagnose AHF.

Improving the Generalizability and Performance of an Ultrasound Deep Learning Model Using Limited Multicenter Data for Lung Sliding Artifact Identification

Deep learning (DL) models for medical image classification frequently struggle to generalize to data from outside institutions. Additional clinical data are also rarely collected to comprehensively assess and understand model performance amongst subgroups. Following the development of a single-center model to identify the lung sliding artifact on lung ultrasound (LUS), we pursued a validation strategy using external LUS data. As annotated LUS data are relatively scarce-compared to other medical imaging data-we adopted a novel technique to optimize the use of limited external data to improve model generalizability. Externally acquired LUS data from three tertiary care centers, totaling 641 clips from 238 patients, were used to assess the baseline generalizability of our lung sliding model. We then employed our novel Threshold-Aware Accumulative Fine-Tuning (TAAFT) method to fine-tune the baseline model and determine the minimum amount of data required to achieve predefined performance goals. A subgroup analysis was also performed and Grad-CAM++ explanations were examined. The final model was fine-tuned on one-third of the external dataset to achieve 0.917 sensitivity, 0.817 specificity, and 0.920 area under the receiver operator characteristic curve (AUC) on the external validation dataset, exceeding our predefined performance goals. Subgroup analyses identified LUS characteristics that most greatly challenged the model's performance. Grad-CAM++ saliency maps highlighted clinically relevant regions on M-mode images. We report a multicenter study that exploits limited available external data to improve the generalizability and performance of our lung sliding model while identifying poorly performing subgroups to inform future iterative improvements. This approach may contribute to efficiencies for DL researchers working with smaller quantities of external validation data.

Deep learning for real-time multi-class segmentation of artefacts in lung ultrasound

Lung ultrasound (LUS) has emerged as a safe and cost-effective modality for assessing lung health, particularly during the COVID-19 pandemic. However, interpreting LUS images remains challenging due to its reliance on artefacts, leading to operator variability and limiting its practical uptake. To address this, we propose a deep learning pipeline for multi-class segmentation of objects (ribs, pleural line) and artefacts (A-lines, B-lines, B-line confluence) in ultrasound images of a lung training phantom. Lightweight models achieved a mean Dice Similarity Coefficient (DSC) of 0.74, requiring fewer than 500 training images. Applying this method in real-time, at up to 33.4 frames per second in inference, allows enhanced visualisation of these features in LUS images. This could be useful in providing LUS training and helping to address the skill gap. Moreover, the segmentation masks obtained from this model enable the development of explainable measures of disease severity, which have the potential to assist in the triage and management of patients. We suggest one such semi-quantitative measure called the B-line Artefact Score, which is related to the percentage of an intercostal space occupied by B-lines and in turn may be associated with the severity of a number of lung conditions. Moreover, we show how transfer learning could be used to train models for small datasets of clinical LUS images, identifying pathologies such as simple pleural effusions and lung consolidation with DSC values of 0.48 and 0.32 respectively. Finally, we demonstrate how such DL models could be translated into clinical practice, implementing the phantom model alongside a portable point-of-care ultrasound system, facilitating bedside assessment and improving the accessibility of LUS.

Lung Ultrasound Signs: The Beginning. Part 3-An Accademia di Ecografia Toracica Comprehensive Review on Ultrasonographic Signs and Real Needs

Over the last 20 years, scientific literature and interest on chest/lung ultrasound (LUS) have exponentially increased. Interpreting mixed-anatomical and artifactual-pictures determined the need of a proposal of a new nomenclature of artifacts and signs to simplify learning, spread, and implementation of this technique. The aim of this review is to collect and analyze different signs and artifacts reported in the history of chest ultrasound regarding normal lung, pleural pathologies, and lung consolidations. By reviewing the possible physical and anatomical interpretation of these artifacts and signs reported in the literature, this work aims to present the AdET (Accademia di Ecografia Toracica) proposal of nomenclature and to bring order between published studies.

Low-complexity lung ultrasound video scoring by means of intensity projection-based video compression

Since the outbreak of COVID-19, efforts have been made towards semi-quantitative analysis of lung ultrasound (LUS) data to assess the patient's condition. Several methods have been proposed in this regard, with a focus on frame-level analysis, which was then used to assess the condition at the video and prognostic levels. However, no extensive work has been done to analyze lung conditions directly at the video level. This study proposes a novel method for video-level scoring based on compression of LUS video data into a single image and automatic classification to assess patient's condition. The method utilizes maximum, mean, and minimum intensity projection-based compression of LUS video data over time. This enables to preserve hyper- and hypo-echoic data regions, while compressing the video down to a maximum of three images. The resulting images are then classified using a convolutional neural network (CNN). Finally, the worst predicted score given among the images is assigned to the corresponding video. The results show that this compression technique can achieve a promising agreement at the prognostic level (81.62%), while the video-level agreement remains comparable with the state-of-the-art (46.19%). Conclusively, the suggested method lays down the foundation for LUS video compression, shifting from frame-level to direct video-level analysis of LUS data.

Can Artificial Intelligence Aid Diagnosis by Teleguided Point-of-Care Ultrasound? A Pilot Study for Evaluating a Novel Computer Algorithm for COVID-19 Diagnosis Using Lung Ultrasound

With the 2019 coronavirus disease (COVID-19) pandemic, there is an increasing demand for remote monitoring technologies to reduce patient and provider exposure. One field that has an increasing potential is teleguided ultrasound, where telemedicine and point-of-care ultrasound (POCUS) merge to create this new scope. Teleguided POCUS can minimize staff exposure while preserving patient safety and oversight during bedside procedures. In this paper, we propose the use of teleguided POCUS supported by AI technologies for the remote monitoring of COVID-19 patients by non-experienced personnel including self-monitoring by the patients themselves. Our hypothesis is that AI technologies can facilitate the remote monitoring of COVID-19 patients through the utilization of POCUS devices, even when operated by individuals without formal medical training. In pursuit of this goal, we performed a pilot analysis to evaluate the performance of users with different clinical backgrounds using a computer-based system for COVID-19 detection using lung ultrasound. The purpose of the analysis was to emphasize the potential of the proposed AI technology for improving diagnostic performance, especially for users with less experience.

Advancing healthcare through thoracic ultrasound research in older patients

This paper reports the proceedings of a meeting convened by the Research Group on Thoracic Ultrasound in Older People of the Italian Society of Gerontology and Geriatrics, to discuss the current state-of-the-art of clinical research in the field of geriatric thoracic ultrasound and identify unmet research needs and potential areas of development. In the last decade, point-of-care thoracic ultrasound has entered clinical practice for diagnosis and management of several respiratory illnesses, such as bacterial and viral pneumonia, pleural effusion, acute heart failure, and pneumothorax, especially in the emergency-urgency setting. Very few studies, however, have been specifically focused on older patients with frailty and multi-morbidity, who frequently exhibit complex clinical pictures needing multidimensional evaluation. At the present state of knowledge, there is still uncertainty on the best requirements of ultrasound equipment, methodology of examination, and reporting needed to optimize the advantages of thoracic ultrasound implementation in the care of geriatric patients. Other issues regard differential diagnosis between bacterial and aspiration pneumonia, objective grading of interstitial syndrome severity, quantification and monitoring of pleural effusions and solid pleural lesions, significance of ultrasonographic assessment of post-COVID-19 sequelae, and prognostic value of assessment of diaphragmatic thickness and motility. Finally, application of remote ultrasound diagnostics in the community and nursing home setting is still poorly investigated by the current literature. Overall, the presence of several open questions on geriatric applications of thoracic ultrasound represents a strong call to implement clinical research in this field.

Ultrasound multifrequency strategy to estimate the lung surface roughness, in silico and in vitro results

Lung ultrasound (LUS) is an important imaging modality to assess the state of the lung surface. Nevertheless, LUS is limited to the visual evaluation of imaging artifacts, especially the vertical ones. These artifacts are observed in pathologies characterized by a reduction of dimensions of air-spaces (alveoli). In contrast, there exist pathologies, such as chronic obstructive pulmonary disease (COPD), in which an enlargement of air-spaces can occur, which causes the lung surface to behave essentially as a perfect reflector, thus not allowing ultrasound penetration. This characteristic high reflectivity could be exploited to characterize the lung surface. Specifically, air-spaces of different sizes could cause the lung surface to have a different roughness, whose estimation could provide a way to assess the state of the lung surface. In this study, we present a quantitative multifrequency approach aiming at estimating the lung surface's roughness by measuring image intensity variations along the lung surface as a function of frequency. This approach was tested both in silico and in vitro, and it showed promising results. For the in vitro experiments, radiofrequency (RF) data were acquired from a novel experimental model. The results showed consistency between in silico and in vitro experiments.

An Interpretable Neonatal Lung Ultrasound Feature Extraction and Lung Sliding Detection System Using Object Detectors

The objective of this study was to develop an interpretable system that could detect specific lung features in neonates. A challenging aspect of this work was that normal lungs showed the same visual features (as that of Pneumothorax (PTX)). M-mode is typically necessary to differentiate between the two cases, but its generation in clinics is time-consuming and requires expertise for interpretation, which remains limited. Therefore, our system automates M-mode generation by extracting Regions of Interest (ROIs) without human in the loop. Object detection models such as faster Region Based Convolutional Neural Network (fRCNN) and RetinaNet models were employed to detect seven common Lung Ultrasound (LUS) features. fRCNN predictions were then stored and further used to generate M-modes. Beyond static feature extraction, we used a Hough transform based statistical method to detect "lung sliding" in these M-modes. Results showed that fRCNN achieved a greater mean Average Precision (mAP) of 86.57% (Intersection-over-Union (IoU) = 0.2) than RetinaNet, which only displayed a mAP of 61.15%. The calculated accuracy for the generated RoIs was 97.59% for Normal videos and 96.37% for PTX videos. Using this system, we successfully classified 5 PTX and 6 Normal video cases with 100% accuracy. Automating the process of detecting seven prominent LUS features addresses the time-consuming manual evaluation of Lung ultrasound in a fast paced environment. Clinical impact: Our research work provides a significant clinical impact as it provides a more accurate and efficient method for diagnosing lung diseases in neonates.

Ultrasound imaging of lung disease and its relationship to histopathology: An experimentally validated simulation approach

Lung ultrasound (LUS) is a widely used technique in clinical lung assessment, yet the relationship between LUS images and the underlying disease remains poorly understood due in part to the complexity of the wave propagation physics in complex tissue/air structures. Establishing a clear link between visual patterns in ultrasound images and underlying lung anatomy could improve the diagnostic accuracy and clinical deployment of LUS. Reverberation that occurs at the lung interface is complex, resulting in images that require interpretation of the artifacts deep in the lungs. These images are not accurate spatial representations of the anatomy due to the almost total reflectivity and high impedance mismatch between aerated lung and chest wall. Here, we develop an approach based on the first principles of wave propagation physics in highly realistic maps of the human chest wall and lung to unveil a relationship between lung disease, tissue structure, and its resulting effects on ultrasound images. It is shown that Fullwave numerical simulations of ultrasound propagation and histology-derived acoustical maps model the multiple scattering physics at the lung interface and reproduce LUS B-mode images that are comparable to clinical images. However, unlike clinical imaging, the underlying tissue structure model is known and controllable. The amount of fluid and connective tissue components in the lung were gradually modified to model disease progression, and the resulting changes in B-mode images and non-imaging reverberation measures were analyzed to explain the relationship between pathological modifications of lung tissue and observed LUS.

Machine Learning Algorithm Detection of Confluent B-Lines

OBJECTIVE

B-lines are a ring-down artifact of lung ultrasound that arise with increased alveolar water in conditions such as pulmonary edema and infectious pneumonitis. Confluent B-line presence may signify a different level of pathology compared with single B-lines. Existing algorithms aimed at B-line counting do not distinguish between single and confluent B-lines. The objective of this study was to test a machine learning algorithm for confluent B-line identification.

METHODS

This study used a subset of 416 clips from 157 subjects, previously acquired in a prospective study enrolling adults with shortness of breath at two academic medical centers, using a hand-held tablet and a 14-zone protocol. After exclusions, random sampling generated a total of 416 clips (146 curvilinear, 150 sector and 120 linear) for review. A group of five experts in point-of-care ultrasound blindly evaluated the clips for presence/absence of confluent B-lines. Ground truth was defined as majority agreement among the experts and used for comparison with the algorithm.

RESULTS

Confluent B-lines were present in 206 of 416 clips (49.5%). Sensitivity and specificity of confluent B-line detection by algorithm compared with expert determination were 83% (95% confidence interval [CI]: 0.77-0.88) and 92% (95% CI: 0.88-0.96). Sensitivity and specificity did not statistically differ between transducers. Agreement between algorithm and expert for confluent B-lines measured by unweighted κ was 0.75 (95% CI: 0.69-0.81) for the overall set.

CONCLUSION

The confluent B-line detection algorithm had high sensitivity and specificity for detection of confluent B-lines in lung ultrasound point-of-care clips, compared with expert determination.

COVID-19 Lung Ultrasound Scores and Lessons from the Pandemic: A Narrative Review

The WHO recently declared that COVID-19 no longer constitutes a public health emergency of international concern; however, lessons learned through the pandemic should not be left behind. Lung ultrasound was largely utilized as a diagnostic tool thanks to its feasibility, easy application, and the possibility to reduce the source of infection for health personnel. Lung ultrasound scores consist of grading systems used to guide diagnosis and medical decisions, owning a good prognostic value. In the emergency context of the pandemic, several lung ultrasound scores emerged either as new scores or as modifications of pre-existing ones. Our aim is to clarify the key aspects of lung ultrasound and lung ultrasound scores to standardize their clinical use in a non-pandemic context. The authors searched on PubMed for articles related to "COVID-19", "ultrasound", and "Score" until 5 May 2023; other keywords were "thoracic", "lung", "echography", and "diaphragm". A narrative summary of the results was made. Lung ultrasound scores are demonstrated to be an important tool for triage, prediction of severity, and aid in medical decisions. Ultimately, the existence of numerous scores leads to a lack of clarity, confusion, and an absence of standardization.

The Role of Ultrasound in the Diagnosis of Pulmonary Infection Caused by Intracellular, Fungal Pathogens and Mycobacteria: A Systematic Review

BACKGROUND

Lung ultrasound (LUS) is a widely available technique allowing rapid bedside detection of different respiratory disorders. Its reliability in the diagnosis of community-acquired lung infection has been confirmed. However, its usefulness in identifying infections caused by specific and less common pathogens (e.g., in immunocompromised patients) is still uncertain.

METHODS

This systematic review aimed to explore the most common LUS patterns in infections caused by intracellular, fungal pathogens or mycobacteria.

RESULTS

We included 17 studies, reporting a total of 274 patients with M. pneumoniae, 30 with fungal infection and 213 with pulmonary tuberculosis (TB). Most of the studies on M. pneumoniae in children found a specific LUS pattern, mainly consolidated areas associated with diffuse B lines. The typical LUS pattern in TB consisted of consolidation and small subpleural nodes. Only one study on fungal disease reported LUS specific patterns (e.g., indicating "halo sign" or "reverse halo sign").

CONCLUSIONS

Considering the preliminary data, LUS appears to be a promising point-of-care tool, showing patterns of atypical pneumonia and TB which seem different from patterns characterizing common bacterial infection. The role of LUS in the diagnosis of fungal disease is still at an early stage of exploration. Large trials to investigate sonography in these lung infections are granted.