Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Ultrasound-Based AI for COVID-19 Detection: A Comprehensive Review of Public and Private Lung Ultrasound Datasets and Studies

Abrar Morshed, Abdulla Al Shihab, Md Abrar Jahin, Md Jaber Al Nahian, Md Murad Hossain Sarker, Md Sharjis Ibne Wadud, Mohammad Istiaq Uddin, Muntequa Imtiaz Siraji, Nafisa Anjum, Sumiya Rajjab Shristy, Tanvin Rahman, Mahmuda Khatun, Md Rubel Dewan, Mosaddeq Hossain, Razia Sultana, Ripel Chakma, Sonet Barua Emon, Towhidul Islam, Mohammad Arafat Hussain

TL;DR

It is suggested that ultrasound-based AI studies for COVID-19 detection have great potential for clinical use, especially for children and pregnant women.

Abstract

The COVID-19 pandemic has affected millions of people globally, with respiratory organs being strongly affected in individuals with comorbidities. Medical imaging-based diagnosis and prognosis have become increasingly popular in clinical settings for detecting COVID-19 lung infections. Among various medical imaging modalities, ultrasound stands out as a low-cost, mobile, and radiation-safe imaging technology. In this comprehensive review, we focus on AI-driven studies utilizing lung ultrasound (LUS) for COVID-19 detection and analysis. We provide a detailed overview of both publicly available and private LUS datasets and categorize the AI studies according to the dataset they used. Additionally, we systematically analyzed and tabulated the studies across various dimensions, including data preprocessing methods, AI models, cross-validation techniques, and evaluation metrics. In total, we reviewed 60 articles, 41 of which utilized public datasets, while the remaining employed private data. Our findings suggest that ultrasound-based AI studies for COVID-19 detection have great potential for clinical use, especially for children and pregnant women. Our review also provides a useful summary for future researchers and clinicians who may be interested in the field.

Ultrasound-Based AI for COVID-19 Detection: A Comprehensive Review of Public and Private Lung Ultrasound Datasets and Studies

TL;DR

It is suggested that ultrasound-based AI studies for COVID-19 detection have great potential for clinical use, especially for children and pregnant women.

Abstract

The COVID-19 pandemic has affected millions of people globally, with respiratory organs being strongly affected in individuals with comorbidities. Medical imaging-based diagnosis and prognosis have become increasingly popular in clinical settings for detecting COVID-19 lung infections. Among various medical imaging modalities, ultrasound stands out as a low-cost, mobile, and radiation-safe imaging technology. In this comprehensive review, we focus on AI-driven studies utilizing lung ultrasound (LUS) for COVID-19 detection and analysis. We provide a detailed overview of both publicly available and private LUS datasets and categorize the AI studies according to the dataset they used. Additionally, we systematically analyzed and tabulated the studies across various dimensions, including data preprocessing methods, AI models, cross-validation techniques, and evaluation metrics. In total, we reviewed 60 articles, 41 of which utilized public datasets, while the remaining employed private data. Our findings suggest that ultrasound-based AI studies for COVID-19 detection have great potential for clinical use, especially for children and pregnant women. Our review also provides a useful summary for future researchers and clinicians who may be interested in the field.

Paper Structure

This paper contains 62 sections, 15 equations, 4 figures, 7 tables.

Table of Contents

  1. Introduction
  2. Lung Ultrasound COVID-19 Datasets
  3. Publicly Accessible LUS COVID-19 Datasets
  4. Non-Accessible LUS COVID-19 Private Datasets
  5. Data Pre-processing and Augmentation
  6. Curve-to-linear/Linear-to-curve Conversion
  7. Image Resizing
  8. Intensity Normalization
  9. Image Augmentation
  10. Other Image Processing Techniques
  11. AI in LUS COVID-19 Studies
  12. AI Models
  13. Convolutional Neural Networks (CNN)
  14. Recurrent Neural Networks (RNN)
  15. COVID-Net
  16. ...and 47 more sections

Figures (4)

  • Figure 1: Demonstration of different types of lines that may appear in LUS images. A-lines are marked with blue, B-lines are marked with yellow, and the pleural line is marked with green born2020pocovid.
  • Figure 2: Example ultrasound images of a healthy lung (left), community-acquired pneumonia (CAP)-infected lung (middle), and COVID-19-infected lung (right) born2020pocovid.
  • Figure 3: A pie-chart showing the percentage of reviewed articles in this study per LUS datasets.
  • Figure 4: Organization of reviewed articles in terms of the AI model types and configurations. Hybrid models represent those studies that used two or more types of machine learning strategies together as part of a common predictive model. Four studies used both CML and DL strategies, but they were not part of a common predictive model. Those articles are shown twice under different categories and indicated with common superscripts, 1-4.