Revolutionizing AI-enabled Medical Imaging – Asian Scientist Magazine

Asian Scientist Magazine (Jun. 24, 2022) – Medical imaging is an important part of modern healthcare, enhancing both the precision, reliability and development of treatment for various diseases. Over the years, artificial intelligence has further enhanced the process.

However, conventional medical image diagnosis employing AI algorithms requires large amounts of annotations as supervision signals for model training. To acquire accurate labels for the AI ​​algorithms, radiologists prepare radiology reports for each of their patients, followed by annotation staff extracting and confirming structured labels from those reports using human-defined rules and existing natural language processing (NLP) tools. The ultimate accuracy of extracted labels hinges on the quality of human work and various NLP tools. The method comes at a heavy price, being both labor intensive and time consuming.

To get around that challenge, a team of researchers at the University of Hong Kong (HKU) has developed a new approach “REFERS” (Reviewing Free-text Reports for Supervision), which can cut human cost down by 90 percent, by enabling the automatic acquisition of supervision signals from hundreds of thousands of radiology reports at the same time. Its predictions are highly accurate, surpassing its counterpart of conventional medical image diagnosis employing AI algorithms. The breakthrough was published in Nature Machine Intelligence.

“AI-enabled medical image diagnosis has the potential to support medical specialists in reducing their workload and improving the diagnostic efficiency and accuracy, including but not limited to reducing the diagnosis time and detecting subtle disease patterns,” said Professor Yu Yizhou, leader of the team from HKU’s Department of Computer Science under the Faculty of Engineering.

“We believe abstract and complex logical reasoning sentences in radiology reports provide sufficient information for learning easily transferable visual features. With appropriate training, REFERS directly learns radiograph representations from free-text reports without the need to involve manpower in labeling, ”said Professor Yu.

For training REFERS, the research team uses a public database with 370,000 X-Ray images, and associated radiology reports, on 14 common chest diseases including atelectasis, cardiomegaly, pleural effusion, pneumonia and pneumothorax.

REFERS achieves the goal by accomplishing two report-related tasks, i.e., report generation and radiograph – report matching.

“Compared to conventional methods that heavily rely on human annotations, REFERS has the ability to acquire supervision from each word in the radiology reports. We can substantially reduce the amount of data annotation by 90 percent and the cost to build medical artificial intelligence. It marks a significant step towards realizing generalized medical artificial intelligence, ”said the paper’s first author Dr. ZHOU Hong-Yu.


Source: The University of Hong Kong; Photo: Unsplash

The article can be found at Generalized radiograph representation learning via cross-supervision between images and free-text radiology reports.

fbq(‘init’, ‘2004394943144714’, {}, {
“agent”: “wordpress-5.5.9-1.7.5”

fbq(‘track’, ‘PageView’, {
“source”: “wordpress”,
“version”: “5.5.9”,
“pluginVersion”: “1.7.5”

if(typeof jQuery != ‘undefined’) {
jQuery(‘body’).on(‘added_to_cart’, function(event) {

// Ajax action.
$.get(‘?wc-ajax=fb_inject_add_to_cart_event’, function(data) {


Source link


Fresh protests as new US abortion reality takes shape

Thousands of abortion rights defenders have rallied across America in protest against the Supreme Court’s ruling. The ruling will now allow states to deny the right to abortion. Many fear that the Supreme Court, with a clear conservative majority, might next set its sights on rights like same-sex marriage and contraception. Abortion rights defenders fanned […]

Read More

How Chelsea could line up next season with Cristiano Ronaldo in attack as ‘Man Utd star’s agent holds transfer talks’

CHELSEA could remarkably begin the 2022/23 season with Cristiano Ronaldo leading the line at Stamford Bridge. Shocking reports emerged on Saturday that the Portuguese superstar’s agent Jorge Mendes met with new Blues owner Todd Boehly last week and discussed a possible move for his client. 3 Ronaldo’s agent has reportedly held talks with Chelsea about […]

Read More

Kate Middleton Dons Full Military Uniform In Honor Of UK’s Armed Forces Day. See Pics

<!– –> Kate Middleton shared photos of recent visit to England’s Army Training Center Pirbright. On United Kingdom’s Armed Forces Day, the Duchess of Cambridge Kate Middleton shared pictures of herself dressed in full military gear. The images went live on the Duke and Duchess of Cambridge’s official Instagram account. The first slide showed Kate […]

Read More