Impact of pixel, intensity, & temporal resolution on automatic scoring of LUS from Coronavirus disease 2019 patients

With the outbreak of the COVID-19, remote diagnosis, patient monitoring, collection, and transmission of data from electronic devices is rapidly taking share in the health sector. These devices are however limited on resources like energy, memory and processing power. Consequently, it is highly relevant to investigate minimizing the data, keeping intact the information content. The objective of this study is to thus observe the impact of pixel, intensity, & temporal resolution on automated scoring of LUS data. First, 448 videos from 20 patients were normalized to a common pixel resolution, i.e., the largest found over the dataset (841 pixels/cm2). Next, pixel and intensity resolution were further reduced by down-sampling factor of 2,3, and 4, and by quantization factor of 2,4, and 8 respectively. Furthermore, number of frames were down-sampled as a function of time by factor of 1 to 10 with step-size of 1. Resampled, quantized, and temporally reduced videos were evaluated using the DL algorithm (doi: 10.1109/TMI.2020.2994459) and frame, video, and prognostic-level results were obtained. It was found that no significant change in the prognostic results is observed when the data is reduced by 32 times to its original size and by 10 times to the original number of frames. © 2022 Acoustical Society of America.

Neuro-symbolic interpretable AI for automatic COVID-19 patient-stratification based on standardised lung ultrasound data

In the current pandemic, being able to efficiently stratify patients depending on their probability to develop a severe form of COVID-19 can improve the outcome of treatments and optimize the use of the available resources. To this end, recent studies proposed to use deep-networks to perform automatic stratification of COVID-19 patients based on lung ultrasound (LUS) data. In this work, we present a novel neuro-symbolic approach able to provide video-level predictions by aggregating results from frame-level analysis made by deep-networks. Specifically, a decision tree was trained, which provides direct access to the decision process and a high-level explainability. This approach was tested on 1808 LUS videos acquired from 100 patients diagnosed as COVID-19 positive by a RT-PCR swab test. Each video was scored by LUS experts according to a 4-level scoring system specifically developed for COVID-19. This information was utilised for both the training and testing of the algorithms. A five-folds cross-validation process was utilised to assess the performance of the presented approach and compare it with results achieved by deep-learning models alone. Results show that this novel approach achieves better performance (82% of mean prognostic agreement) than a threshold-based ensemble of deep-learning models (78% of mean prognostic agreement). © 2022 Acoustical Society of America.

Automatically Scoring Lung Ultrasound Videos of COVID-19 and post-COVID-19 Patients

Lung ultrasound (LUS) imaging is playing an important role in the current pandemic, allowing the evaluation of patients affected by COVID-19 pneumonia. However, LUS is limited to the visual inspection of ultrasound data, which negatively affects the reproducibility and reliability of the findings. For these reasons, we were the first to propose a standardized imaging protocol and a scoring system, from which we developed the first artificial intelligence (AI) models able to evaluate LUS videos. Furthermore, we demonstrated prognostic value of our approach and its utility for patients' stratification. In this study, we report on the level of agreement between AI and LUS clinical experts (MD) on LUS data acquired from both COVID-19 patients and post-COVID-19 patients. © 2022 IEEE.

Automatic Scoring of COVID-19 LUS Videos Using Cross-correlation-Based Features Aggregated from Frame-Level Confidence Levels Obtained by a Pre-trained Deep Neural Network

The emergence of COVID-19 has encouraged researchers to seek a method to detect and monitor patients infected with SARS-CoV 2. The use of lung ultrasound (LUS) in this setting is rapidly spreading because of its portability, cost-effectiveness, real-time imaging, and safety. LUS has demonstrated the potential to be widely used to assess the condition of the lungs in COVID-19 patients. Given frame-level labels provided by a pre-trained deep neural network (DNN), our goal is to identify an aggregation strategy that allows to move from frame-level to video-level, which is the output required by physicians for clinical evaluation. To achieve this goal, we propose a novel aggregation method based on the cross-correlation coefficients. The logic behind this idea is that, based on the similarity between the score's variables (at frame level), the cross-correlation should be informative as to how to discriminate at video level. We applied our approach to the LUS data from a multi-center study comprising of 283, 231, and 448 LUS videos from Lodi General, Gemelli, and San Matteo Hospital, respectively. Results show that the video-level agreement with clinical experts is obtained in 87.6% of the cases, which represents a promising video-level accuracy. © 2022 IEEE.

A Multicenter Study Assessing Artificial Intelligence Capability in Scoring Lung Ultrasound Videos of COVID-19 Patients

Lung ultrasound (LUS) imaging is playing an important role in the current pandemic, allowing the evaluation of patients affected by COVID-19 pneumonia. However, LUS is limited to the visual inspection of ultrasound data, which negatively affects the reproducibility and reliability of the findings. For these reasons, we were the first to propose a standardized imaging protocol and a scoring system, from which we developed the first artificial intelligence (AI) models able to evaluate LUS videos. Furthermore, we demonstrated the prognostic value of our approach and its utility for patients' stratification. In this study, we report on the level of agreement between the AI and LUS clinical experts (MD) when evaluating LUS data. Specifically, in the stratification between patients at high risk of clinical worsening and patients at low risk, the agreement between MDs and AI reached 82%. These encouraging results open to the possibility of exploiting AI for fast and accurate stratification of COVID-19 patients. © 2021 IEEE.

Point of care image analysis for COVID-19

Early detection of COVID-19 is key in containing the pandemic. Disease detection and evaluation based on imaging is fast and cheap and therefore plays an important role in COVID-19 handling. COVID-19 is easier to detect in chest CT, however, it is expensive, non-portable, and difficult to disinfect, making it unfit as a point-of-care (POC) modality. On the other hand, chest X-ray (CXR) and lung ultrasound (LUS) are widely used, yet, COVID-19 findings in these modalities are not always very clear. Here we train deep neural networks to significantly enhance the capability to detect, grade and monitor COVID-19 patients using CXRs and LUS. Collaborating with several hospitals in Israel we collect a large dataset of CXRs and use this dataset to train a neural network obtaining above 90% detection rate for COVID-19. In addition, in collaboration with ULTRa (Ultrasound Laboratory Trento, Italy) and hospitals in Italy we obtained POC ultrasound data with annotations of the severity of disease and trained a deep network for automatic severity grading. © 2021 IEEE

How to perform lung ultrasound in pregnant women with suspected COVID-19.

Under certain circumstances, such as during the current COVID-19 outbreak, pregnant women can be a target for respiratory infection, and lung examination may be required as part of their clinical evaluation, ideally while avoiding exposure to radiation. We propose a practical approach for obstetricians/gynecologists to perform lung ultrasound examination, discussing potential applications, semiology and practical aspects, which could be of particular importance in emergency situations, such as the current pandemic infection of COVID-19. Copyright © 2020 ISUOG. Published by John Wiley & Sons Ltd.

Clinical role of lung ultrasound for diagnosis and monitoring of COVID-19 pneumonia in pregnant women.

Lung ultrasound has been suggested recently by the Chinese Critical Care Ultrasound Study Group and Italian Academy of Thoracic Ultrasound as an accurate tool to detect lung involvement in COVID-19. Although chest computed tomography (CT) represents the gold standard to assess lung involvement, with a specificity superior even to that of the nasopharyngeal swab for diagnosis, lung ultrasound examination can be a valid alternative to CT scan, with certain advantages, particularly for pregnant women. Ultrasound can be performed directly at the bed-side by a single operator, reducing the risk of spreading the disease among health professionals. Furthermore, it is a radiation-free exam, making it safer and easier to monitor those patients who require a series of exams. We report on four cases of pregnant women affected by COVID-19 who were monitored with lung ultrasound examination. All patients showed sonographic features indicative of COVID-19 pneumonia at admission: irregular pleural lines and vertical artifacts (B-lines) were observed in all four cases, and patchy areas of white lung were observed in two. Lung ultrasound was more sensitive than was chest X-ray in detecting COVID-19. In three patients, we observed almost complete resolution of lung pathology on ultrasound within 96 h of admission. Two pregnancies were ongoing at the time of writing, and two had undergone Cesarean delivery with no fetal complications. Reverse transcription polymerase chain reaction analysis of cord blood and newborn swabs was negative in both of these cases. Copyright © 2020 ISUOG. Published by John Wiley & Sons Ltd.