Artificial Intelligence

Artificial intelligence (AI) is becoming a crucial component of healthcare to help augment physicians and make them more efficient. In medical imaging, it is helping radiologists more efficiently manage PACS worklists, enable structured reporting, auto detect injuries and diseases, and to pull in relevant prior exams and patient data. In cardiology, AI is helping automate tasks and measurements on imaging and in reporting systems, guides novice echo users to improve imaging and accuracy, and can risk stratify patients. AI includes deep learning algorithms, machine learning, computer-aided detection (CAD) systems, and convolutional neural networks.

AI tool helps identify 'invisible' head injuries on MRIs of college athletes
The tool uses a machine learning technique to identify changes on brain MRIs that would otherwise be overlooked by radiologists due to the subtlety of alterations.
Ultrasound AI could help standardize trauma care
Trauma patients who present with poor blood flow and suspected abdominal hemorrhage are well served by emergency physicians using AI-augmented FAST imaging.
Interpretive AI for medical imaging: 5 points of skepticism, idealism
Surveying the landscape of interpretive AI in radiology, two researchers note a yawning gap between great expectations set in the recent past and actual clinical implementations as of spring 2023.
Standalone AI excels at reading digital mammograms, but how does it hold up with DBT exams?
Standalone AI can significantly outperform radiologists' sensitivity in reading digital mammograms and has shown potential in DBT exams as well, but experts are not yet ready to hand over the reins.
CAD software is especially beneficial for radiologists in rural settings
Judging the software’s utility in rural settings with less experienced readers is important for expanding access to and improving care for patients, authors of a new AJR paper noted.

An example of artificial intelligence (AI) automated detection of a intracranial hemorrhage (ICH) in. a CT scan used to send alerts to the stroke acute care team before a radiologist even sees the exam. Example shown by TeraRecon at RSNA 2022.

FDA has now cleared more than 500 healthcare AI algorithms

More than 500 clinical AI algorithms have now been cleared by the FDA, with the majority just in the past couple years.

February 6, 2023

An example of an FDA cleared radiology AI algorithm to automatically take a cardiac CT scan and identify, contour and quantify soft plaque in the coronary arteries. The Cleerly software then generates an automated report with images, measurements and a risk assessment for the patient. This type of quantification is too time consuming and complex for human readers to bother with, but AI assisted reports like this may become a new normal over the next decade. Example from Cleerly Imaging at SCCT 2022.

Legal considerations for artificial intelligence in radiology and cardiology

There are now more than 520 FDA-cleared AI algorithms and the majority are for radiology and cardiology, raising the question of who is liable if the AI gets something wrong.

February 3, 2023

AI program ChatGPT now has a published article in Radiology—is it any good?

The human author reviewing the article wrote about the benefits and inherent risks of utilizing AI in a medical publication setting, concluding that, overall, it could be “a powerful tool” used in the future of medical publishing—when used with caution.

February 2, 2023

VIDEO: Who gets sued when radiology AI fails?

Brent Savoie, MD, JD, vice chair for radiology informatics, section chief of cardiovascular imaging, Vanderbilt University, explains who will get sued when there is a misdiagnosis due to artificial intelligence (AI).

February 2, 2023

#CTA #acuteischemicstroke #AIS #radiomics

CTA-based radiomics can reliably estimate time since stroke onset

Radiomics features could be especially useful for determining TSS in scenarios where clinical data is lacking or unreliable, such as when patients have altered mental status.

January 27, 2023

The key to AI integration? Keeping it straightforward, says GE HealthCare CMO

“To drive adoption, it is important that the technology designed to help productivity doesn’t add more work and complexity,” Dr. Mathias Goyen, Chief Medical Officer for Europe, the Middle East and Africa (EMEA) at GE HealthCare, told Health Imaging.

January 26, 2023

As artificial intelligence (AI) adoption expands in radiology, there is growing concern that AI algorithms needs to undergo quality assurance (QA) reviews. How to validate radiology AI? How can you validate medical imaging AI?

Cardiologists use video-based AI model to ID coronary artery disease

A team of specialists out of Cedars-Sinai developed the deep learning model using TTEs from nearly 3,000 patients.

January 26, 2023

Most radiology residents want more AI training, but few are getting it

The majority of radiology trainees have expressed interest in furthering their knowledge of artificial intelligence applications in the field, yet few are offered the opportunity to do so during the course of their education, according to new survey data.

January 20, 2023

Artificial Intelligence

AI tool helps identify 'invisible' head injuries on MRIs of college athletes

Ultrasound AI could help standardize trauma care

Interpretive AI for medical imaging: 5 points of skepticism, idealism

Standalone AI excels at reading digital mammograms, but how does it hold up with DBT exams?

CAD software is especially beneficial for radiologists in rural settings