Automatic AI-Based Diagnosis Framework Enables Detection of COVID-19 from Chest X-Ray Images
By HospiMedica International staff writers Posted on 13 Jan 2022 |
A novel machine learning framework could alleviate the work of radiologists by providing fast and accurate diagnosis of COVID-19 from chest X-ray images.
A team of scientists at Incheon National University (Incheon, Korea) has developed an automatic COVID-19 diagnosis framework that turns things up a notch by combining two powerful artificial intelligence (AI)-based techniques. Their system can be trained to accurately differentiate between chest X-ray images of COVID-19 patients from non-COVID-19 ones.
Several studies have reported that AI-based systems can be used to detect COVID-19 in chest X-ray images because the disease tends to produce areas with pus and water in the lungs, which show up as white spots in the X-ray scans. Although various diagnostic AI models based on this principle have been proposed, improving their accuracy, speed, and applicability remains a top priority.
The scientists developed the new COVID-19 detection system by combining the two algorithms Faster R-CNN and ResNet-101. The first one is a machine learning-based model that uses a region-proposal network, which can be trained to identify the relevant regions in an input image. The second one is a deep-learning neural network comprising 101 layers, which was used as a backbone. ResNet-101, when trained with enough input data, is a powerful model for image recognition.
The scientists believe that their strategy could prove useful for the early detection of COVID-19 in hospitals and public health centers. Using automatic diagnostic techniques based on AI technology could take some work and pressure off of radiologists and other medical experts, who have been facing huge workloads since the pandemic started. Moreover, as more modern medical devices become connected to the Internet, it will be possible to feed vast amounts of training data to the proposed model; this will result in even higher accuracies, and not just for COVID-19.
"To the best of our knowledge, our approach is the first to combine ResNet-101 and Faster R-CNN for COVID-19 detection," said Professor Gwanggil Jeon of Incheon National University who led the team. "After training our model with 8800 X-ray images, we obtained a remarkable accuracy of 98%."
"The deep learning approach used in our study are applicable to other types of medical images and could be used to diagnose different diseases," added Prof. Jeon.
Related Links:
Incheon National University
Latest COVID-19 News
- Low-Cost System Detects SARS-CoV-2 Virus in Hospital Air Using High-Tech Bubbles
- World's First Inhalable COVID-19 Vaccine Approved in China
- COVID-19 Vaccine Patch Fights SARS-CoV-2 Variants Better than Needles
- Blood Viscosity Testing Can Predict Risk of Death in Hospitalized COVID-19 Patients
- ‘Covid Computer’ Uses AI to Detect COVID-19 from Chest CT Scans
- MRI Lung-Imaging Technique Shows Cause of Long-COVID Symptoms
- Chest CT Scans of COVID-19 Patients Could Help Distinguish Between SARS-CoV-2 Variants
- Specialized MRI Detects Lung Abnormalities in Non-Hospitalized Long COVID Patients
- AI Algorithm Identifies Hospitalized Patients at Highest Risk of Dying From COVID-19
- Sweat Sensor Detects Key Biomarkers That Provide Early Warning of COVID-19 and Flu
- Study Assesses Impact of COVID-19 on Ventilation/Perfusion Scintigraphy
- CT Imaging Study Finds Vaccination Reduces Risk of COVID-19 Associated Pulmonary Embolism
- Third Day in Hospital a ‘Tipping Point’ in Severity of COVID-19 Pneumonia
- Longer Interval Between COVID-19 Vaccines Generates Up to Nine Times as Many Antibodies
- AI Model for Monitoring COVID-19 Predicts Mortality Within First 30 Days of Admission
- AI Predicts COVID Prognosis at Near-Expert Level Based Off CT Scans