Automated lung ultrasound scoring for evaluation of coronavirus disease 2019 pneumonia using two-stage cascaded deep learning model

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.

Combining quantitative and qualitative analysis for scoring pleural line in lung ultrasound

Objective.Accurate assessment of pleural line is crucial for the application of lung ultrasound (LUS) in monitoring lung diseases, thereby aim of this study is to develop a quantitative and qualitative analysis method for pleural line.Approach.The novel cascaded deep learning model based on convolution and multilayer perceptron was proposed to locate and segment the pleural line in LUS images, whose results were applied for quantitative analysis of textural and morphological features, respectively. By using gray-level co-occurrence matrix and self-designed statistical methods, eight textural and three morphological features were generated to characterize the pleural lines. Furthermore, the machine learning-based classifiers were employed to qualitatively evaluate the lesion degree of pleural line in LUS images.Main results.We prospectively evaluated 3770 LUS images acquired from 31 pneumonia patients. Experimental results demonstrated that the proposed pleural line extraction and evaluation methods all have good performance, with dice and accuracy of 0.87 and 94.47%, respectively, and the comparison with previous methods found statistical significance (P< 0.001 for all). Meanwhile, the generalization verification proved the feasibility of the proposed method in multiple data scenarios.Significance.The proposed method has great application potential for assessment of pleural line in LUS images and aiding lung disease diagnosis and treatment.

The value of lung ultrasound score in neonatal respiratory distress syndrome: a prospective diagnostic cohort study

Rationale

The accurate diagnosis of critically ill patients with respiratory failure can be achieved through lung ultrasound (LUS) score. Considering its characteristics, it is speculated that this technique might also be useful for patients with neonatal respiratory distress syndrome (NRDS). Thus, there is a need for precise imaging tools to monitor such patients.

Objectives

This double-blind randomized cohort study aims to investigate the impact of LUS and related scores on the severity of NRDS patients.

Methods

This study was conducted as a prospective double-blind randomized study. Bivariate correlation analysis was conducted to investigate the relationship between LUS score and Oxygenation Index (OI), Respiratory Index (RI), and Sequential Organ Failure Assessment (SOFA) score. Spearman's correlation coefficient was used to generate correlation heat maps, elucidating the associations between LUS and respective parameters in different cohorts. Receiver Operating Characteristic (ROC) curves were employed to calculate the predictive values, sensitivity, and specificity of different scores in determining the severity of NRDS.

Results

This study ultimately included 134 patients admitted to the intensive care unit (ICU) between December 2020 and June 2022. Among these patients, 72 were included in the NRDS cohort, while 62 were included in the Non-NRDS (N-NRDS) cohort. There were significant differences in the mean LUS scores between NRDS and N-NRDS patients (p < 0.01). The LUS score was significantly negatively correlated with the OI (p < 0.01), while it was significantly positively correlated with the RI and SOFA scores (p < 0.01). The correlation heatmap revealed the highest positive correlation coefficient between LUS and RI (0.82), while the highest negative correlation coefficient was observed between LUS and OI (-0.8). ROC curves for different scores demonstrated that LUS score had the highest area under the curve (0.91, 95% CI: 0.84-0.98) in predicting the severity of patients' conditions. The combination of LUS and other scores can more accurately predict the severity of NRDS patients, with the highest AUC value of 0.93, significantly higher than using a single indicator alone (p < 0.01).

Conclusion

Our double-blind randomized cohort study demonstrates that LUS, RI, OI, and SOFA scores can effectively monitor the lung ventilation and function in NRDS. Moreover, these parameters and their combination have significant predictive value in evaluating the severity and prognosis of NRDS patients. Therefore, these results provide crucial insights for future research endeavors.

Diagnosis and Prognosis Evaluation of Severe Pneumonia by Lung Ultrasound Score Combined with Serum Inflammatory Markers

Introduction

To analyze the significance of lung ultrasound score (LUS) combined with serum inflammatory indexes in different severities of severe pneumonia and its clinical value on prognosis.

Methods

100 patients with severe pneumonia treated in the Gansu Provincial Hospital from June 2017 to June 2021 were selected as the research objects. According to the acute physiology and chronic health (APACHE II) score, they were divided into a low-risk group (28 cases), a medium-risk group (39 cases) and a high-risk group (33 cases). The general clinical data of the patients (age, gender, smoking history, and underlying diseases) were collected, the lung ultrasound score (LUS) of the patients was measured, and the serum inflammatory indicators (IL-6, IL-10, TNF-α, CRP and NLR) levels; Pearson correlation analysis to evaluate the correlation between LUS score, serum inflammatory index levels and disease severity; receiver operating characteristic (ROC) curve analysis to evaluate the prognostic value of the combined diagnosis of LUS score and serum inflammatory index for the severity of severe pneumonia.

Results

With the increase in the severity of severe pneumonia, the LUS score and the level of inflammation in the body continued to increase, and LUS combined with serum inflammatory indexes could distinguish the severity of low-risk, medium-risk and high-risk severe pneumonia and had high diagnostic value. In addition, the combined diagnosis of LUS and serum inflammatory markers is also closely related to the prognosis of patients with severe pneumonia, which can distinguish the prognosis.

Conclusion

LUS combined with serum inflammatory indicators (IL-6, IL-10, TNF-α, CRP and NLR) can differentiate the severity and prognosis of severe pneumonia, which may be a new direction for diagnosing severe pneumonia and guide early clinical intervention.

The role of ultrasound in predicting non-invasive ventilation outcomes: a systematic review

Purpose

To systematically review and compare ultrasonographic methods and their utility in predicting non-invasive ventilation (NIV) outcomes.

Methods

A systematic review was performed using the PubMed, Medline, Embase, and Cochrane databases from January 2015 to March 2023. The search terms included the following: ultrasound, diaphragm, lung, prediction, non-invasive, ventilation, and outcomes. The inclusion criteria were prospective cohort studies on adult patients requiring non-invasive ventilation in the emergency department or inpatient setting.

Results

Fifteen studies were analyzed, which comprised of 1,307 patients (n = 942 for lung ultrasound score studies; n = 365 patients for diaphragm dysfunction studies). Lung ultrasound scores (LUS) greater than 18 were associated with NIV failure with a sensitivity 62-90.5% and specificity 60-91.9%. Similarly, a diaphragm thickening fraction (DTF) of less than 20% was also associated with NIV failure with a sensitivity 80-84.6% and specificity 76.3-91.5%.

Conclusion

Predicting NIV failure can be difficult by routine initial clinical impression and diagnostic work up. This systematic review emphasizes the importance of using lung and diaphragm ultrasound, in particular the lung ultrasound score and diaphragm thickening fraction respectively, to accurately predict NIV failure, including the need for ICU-level of care, requiring invasive mechanical ventilation, and resulting in higher rates of mortality.

A new classification method for diagnosing COVID-19 pneumonia based on joint CNN features of chest X-ray images and parallel pyramid MLP-mixer module

During the past three years, the coronavirus disease 2019 (COVID-19) has swept the world. The rapid and accurate recognition of covid-19 pneumonia are ,therefore, of great importance. To handle this problem, we propose a new pipeline of deep learning framework for diagnosing COVID-19 pneumonia via chest X-ray images from normal, COVID-19, and other pneumonia patients. In detail, the self-trained YOLO-v4 network was first used to locate and segment the thoracic region, and the output images were scaled to the same size. Subsequently, the pre-trained convolutional neural network was adopted to extract the features of X-ray images from 13 convolutional layers, which were fused with the original image to form a 14-dimensional image matrix. It was then put into three parallel pyramid multi-layer perceptron (MLP)-Mixer modules for comprehensive feature extraction through spatial fusion and channel fusion based on different scales so as to grasp more extensive feature correlation. Finally, by combining all image features from the 14-channel output, the classification task was achieved using two fully connected layers as well as Softmax classifier for classification. Extensive simulations based on a total of 4099 chest X-ray images were conducted to verify the effectiveness of the proposed method. Experimental results indicated that our proposed method can achieve the best performance in almost all cases, which is good for auxiliary diagnosis of COVID-19 and has great clinical application potential.

An overview of deep learning techniques for COVID-19 detection: methods, challenges, and future works

The World Health Organization (WHO) declared a pandemic in response to the coronavirus COVID-19 in 2020, which resulted in numerous deaths worldwide. Although the disease appears to have lost its impact, millions of people have been affected by this virus, and new infections still occur. Identifying COVID-19 requires a reverse transcription-polymerase chain reaction test (RT-PCR) or analysis of medical data. Due to the high cost and time required to scan and analyze medical data, researchers are focusing on using automated computer-aided methods. This review examines the applications of deep learning (DL) and machine learning (ML) in detecting COVID-19 using medical data such as CT scans, X-rays, cough sounds, MRIs, ultrasound, and clinical markers. First, the data preprocessing, the features used, and the current COVID-19 detection methods are divided into two subsections, and the studies are discussed. Second, the reported publicly available datasets, their characteristics, and the potential comparison materials mentioned in the literature are presented. Third, a comprehensive comparison is made by contrasting the similar and different aspects of the studies. Finally, the results, gaps, and limitations are summarized to stimulate the improvement of COVID-19 detection methods, and the study concludes by listing some future research directions for COVID-19 classification.

Multiclass Classification for Detection of COVID-19 Infection in Chest X-Rays Using CNN

Coronavirus took the world by surprise and caused a lot of trouble in all the important fields in life. The complexity of dealing with coronavirus lies in the fact that it is highly infectious and is a novel virus which is hard to detect with exact precision. The typical detection method for COVID-19 infection is the RT-PCR but it is a rather expensive method which is also invasive and has a high margin of error. Radiographies are a good alternative for COVID-19 detection given the experience of the radiologist and his learning capabilities. To make an accurate detection from chest X-Rays, deep learning technologies can be involved to analyze the radiographs, learn distinctive patterns of coronavirus' presence, find these patterns in the tested radiograph, and determine whether the sample is actually COVID-19 positive or negative. In this study, we propose a model based on deep learning technology using Convolutional Neural Networks and training it on a dataset containing a total of over 35,000 chest X-Ray images, nearly 16,000 for COVID-19 positive images, 15,000 for normal images, and 5,000 for pneumonia-positive images. The model's performance was assessed in terms of accuracy, precision, recall, and F1-score, and it achieved 99% accuracy, 0.98 precision, 1.02 recall, and 99.0% F1-score, thus outperforming other deep learning models from other studies.

CM-SegNet: A deep learning-based automatic segmentation approach for medical images by combining convolution and multilayer perceptron

Accurate segmentation of lesions in medical images is of great significance for clinical diagnosis and evaluation. The low contrast between lesions and surrounding tissues increases the difficulty of automatic segmentation, while the efficiency of manual segmentation is low. In order to increase the generalization performance of segmentation model, we proposed a deep learning-based automatic segmentation model called CM-SegNet for segmenting medical images of different modalities. It was designed according to the multiscale input and encoding-decoding thoughts, and composed of multilayer perceptron and convolution modules. This model achieved communication of different channels and different spatial locations of each patch, and considers the edge related feature information between adjacent patches. Thus, it could fully extract global and local image information for the segmentation task. Meanwhile, this model met the effective segmentation of different structural lesion regions in different slices of three-dimensional medical images. In this experiment, the proposed CM-SegNet was trained, validated, and tested using six medical image datasets of different modalities and 5-fold cross validation method. The results showed that the CM-SegNet model had better segmentation performance and shorter training time for different medical images than the previous methods, suggesting it is faster and more accurate in automatic segmentation and has great potential application in clinic.