Deep learning introduces vast improvements to automated abdominal CT interpretation

Orange County, CA - July 13th 2016 -  In 1959, Robert Ledley and Lee B. Lusted published “Reasoning Foundations of Medical Diagnosis,” a seminal work urging the adoption of computers to aid in diagnostic processes to biomedical researchers and physicians. Nearly 60 years later, their hopes are becoming fully realized with artificial intelligence (AI) systems like IBM Watson, which can aid in producing precise diagnoses and research from data collection.

In an article for the American Journal of Roentgenology, Ronald Summers, MD, PhD, from the National Institute of Health’s laboratory, calls on radiologists to recognize that the medical industry landscape will be intrinsically altered by this technology, having already been integrated in cardiac, lung, brain, and breast imaging.

In analyzing the two prevalent ideologies adopted by software engineers when creating AI technology, Summers found that whereas one side concentrates on “handcrafted” methods to distinguish between pathologies and normal tissue, the other utilizes deep learning algorithms that “teach” AI systems to interpret medical images.

Deep learning introduces vast improvements to automated abdominal CT interpretation

Ledley and Lusted claim that in every diagnosis, three mathematical disciplines—symbolic logic, probability, and value theory—inherently exist. Computers are ideal mechanisms to facilitate these characteristics, as they can be programmed to intertwine logical scenarios with probabilistic analysis from previous data collection.

In addition, the duo insists that there are three factors that characterize medical diagnoses: medical knowledge of the relationship between diseases and their symptoms, how symptoms are exhibited by the patient, and the key identifiers that contribute to diagnosis.

With the latest advancements in AI technology, the guidelines for computer-aided diagnosis are in the early stages of effectuation. Deep learning machines have the ability to learn from a host of medical images, use that knowledge to interpret new scans, and cross-examine those images with their corresponding reports. Furthermore, AI systems can analyze bulks of unstructured data to produce extensive knowledge for radiologists.

To perform fully automated abdominal CT image interpretation at the level of a radiologist, the computer must assess all the organs and [survey for] abnormalities present in the images,” Summers writes. “Although [a] seemingly daunting task for the software developer, the numbers of organs and potential abnormalities are finite and can be addressed methodically.”

Despite the common fear that technology will replace humans, the AI, according to Summers, Ledley, and Lusted, will not function as a substitute, but rather, an assistant. All three stress that computer-aided diagnostic tools are necessary and will make the role of the radiologist more intricate in the medical environment. The outlook for the adoption of this technology is positive, as smart diagnostic methods will usher a shift away from repetitive, minute tasks and move towards utilization of the insight and experience these specialists have to offer.

Contact Ampronix:

Increasing Breastfeeding Rates

Email: info@ampronix.com 

International Sales: +1 949-273-8000

Domestic Sales: 1800-400-7972 for US and Canada

Follow Us:

FacebookTwitterLinkedIn

Share This Article:

TwitterFacebookLinkedIn

View our Product Catalog Online Here

 

About Ampronix

Ampronix is a renowned authorized master distributor of the medical industry's top brands as well as a world-class manufacturer of innovative technology. Since 1982, Ampronix has been dedicated to meeting the growing needs of the medical community with its extensive product knowledge, outstanding service, and state-of-the-art repair facility. Ampronix prides itself on its ability to offer tailored, one-stop solutions at a faster and more cost-effective rate than other manufacturers. Ampronix is an ISO & ANSI/ESD certified facility. To learn more go here.

Deep learning introduces vast improvements to automated abdominal CT interpretation Orange County, CA – July 13th 2016 –  In 1959, Robert Ledley and Lee B. Lusted published “Reasoning Foundations of Medical Diagnosis,” a seminal work urging the adoption of computers to aid in diagnostic processes to biomedical researchers and physicians. Nearly 60 years later, their hopes […]