Worth a Thousand Words: Automated Diagnosis of COVID-19 from Chest CTs

In a little over 18 months, the novel coronavirus (Sars-CoV-2) has infected over 18 million people and caused more than 690,000 deaths. The current standard for diagnosis through reverse transcription polymerase chain reaction is limited owing to its low sensitivity, high rate of false positives, and long testing times. This makes it difficult to identify infected patients quickly and provide them with treatment. Furthermore, there is a risk that patients will still spread the disease while waiting for the results of their diagnostic test.

Chest CT scans have emerged as a quick and effective way to diagnose the disease, but they require radiologist expertise to interpret, and sometimes the scans look similar to other kinds of lung infections, like bacterial pneumonia. Now, a new paper in Medical Imaging Science by a team of scientists, including those from Daegu Gyeongbuk Institute of Science (DGIST), South Korea, details a technique for the automated and accurate interpretation of chest CT scans. “As academics who were equally affected by the COVID pandemic, we were keen to use our expertise in medical image analysis to aid in faster diagnosis and improve clinical workflows,” says Prof. Sang Hyun Park and Philip Chikontwe from DGIST, who led the study.

To build their diagnostic framework, the research team used a Machine Learning technique called “Multiple Instance Learning” (MIL). In MIL, the machine learning algorithm is “trained” using sets, or “bags,” of multiple examples called “instances.” The MIL algorithm then uses these bags to learn to label individual examples or inputs. The research team trained their new framework, called dual attention contrastive based MIL (DA-CMIL), to differentiate between COVID and bacterial pneumonia, and found that its performance was on par to other state-of-the-art automated image analysis methods. Moreover, the DA-CMIL algorithm can leverage limited or incomplete information to efficiently train its AI system.

"Our study can be viewed from both a technical and clinical perspective. First, the algorithms introduced here can be extended to similar settings with other types of medical images. Second, the 'dual attention,' particularly the 'spatial attention,' used in the model improves the interpretability of the algorithm, which will help clinicians understand how automated solutions make decisions," explain Prof. Park and Mr. Chikontwe.

This research extends far beyond the COVID pandemic, laying the foundation for the development of more robust and cheap diagnostic systems, which will be of particular benefit to under-developed countries or countries with otherwise limited medical and human resources.

Philip Chikontwe, Miguel Luna, Myeongkyun Kang, Kyung Soo Hong, June Hong Ahn, Sang Hyun Park.
Dual attention multiple instance learning with unsupervised complementary loss for COVID-19 screening.
Medical Image Analysis, 2021. doi: 10.1016/j.media.2021.102105

Most Popular Now

West Midlands to Digitally Transform Can…

NHS patients throughout the West Midlands are to benefit from a digital pathology programme, designed to help reduce cancer backlogs, transform services, and improve the speed and accuracy of cancer...

AI Transforms Smartwatch ECG Signals int…

A study published in Nature Medicine reports the ability of a smartwatch ECG to accurately detect heart failure in nonclinical environments. Researchers at Mayo Clinic applied artificial intelligence (AI) to...

Siemens Healthineers Splits Fast-Growing…

Siemens Healthineers is splitting its Asia Pacific operations into two to allow both China and the rest of the region to achieve their full potential. China, now its own region...

Siemens Healthineers Presents Two Revolu…

7 Tesla (T) Magnetom Terra.X(1) will offer excellent imaging of even the smallest structures 3T Magnetom Cima.X(2) more than doubles the gradient amplitude(3) AI algorithms which can reduce scanning...

3D Protein Structure Predictions Made by…

In a living being, proteins make up roughly everything: from the molecular machines running every cell's metabolism, to the tip of your hair. Encoded in the DNA, a protein may...

New Group to Advance Digital Twins in He…

EDITH (Ecosystem for Digital Twins in Healthcare) Coordination and Support Action (CSA) - a group made up of numerous internationally renowned research institutions, professional associations, companies, and hospitals of excellence...

Willingness to Use Video Telehealth Incr…

Americans' use and willingness to use video telehealth has increased since the beginning of the COVID-19 pandemic, rising most sharply among Black Americans and people with less education, according to...

Evaluating Use of New AI Technology in D…

Published in the Journal of the American Medical Informatics Association, University of Minnesota researchers led a study evaluating federated learning variations for COVID-19 diagnosis in chest x-rays. Federated learning is...

DMEA Call for Papers: Supporting Digital…

25 - 27 April 2023, Berlin, Germany. Health meets digitalisation: from 25 to 27 April 2023 at DMEA - Connecting Digital Health, all actors aiming to promote health IT will be...

Machine Learning can Help Predict Patien…

Predicting which patients will respond well to treatment is a quandary that has plagued the field of cancer immunotherapy for more than four decades. Now, researchers at the Johns Hopkins...

MEDICA 2022 and COMPAMED 2022: Internati…

14 - 17 November 2022, Düsseldorf, Germany. Next week sees the return of the date marked in extra thick outline in many yearly calendars of the international health and medical technology...

MEDICA and COMPAMED Present Themselves a…

14 - 17 November 2022, Düsseldorf, Germany. MEDICA and COMPAMED continue to develop in an extremely vital manner. The world's leading medical trade fair and the international No. 1 for the...