Society is receptive to artificial intelligence (AI) especially that it has been disrupting different industries over the years—the field of healthcare included. Technology-powered machines not only provide beneficial help, but also valuable insight on how to make better the condition of people suffering from various ailments.
Now, AI has once again proven its capabilities as it is venturing into radiology, one of the most vital specializations in medicine. The experimentation done by the researchers from Boston University’s Department of Bioengineering and Bioinformatics and Icahn School of Medicine at Mount Sinai in New York could create a bold impact to the future endeavors of the medical sector.
In the study, the team was able to create a machine learning technology that has deciphered a total number of 96, 303 reports from the radiologists, and 91% of its interpretation was said to be accurate.
The researchers worked with the newly-developed machine to perfect the interpretation of the following medical procedures:
- Computed Tomography (CT) Scans
- Magnetic Resonance Imaging (MRI)
They were also successful in teaching the AI words like ‘colonoscopy’, ‘heartburn’, and ‘phospholipid’.
According to the first author, John Zech, who is also a medical student at Icahn, “The ultimate goal is to create algorithms that help doctors accurately diagnose patients. Deep learning has many potential applications in radiology – triaging to identify studies that require immediate evaluation, flagging abnormal parts of cross-sectional imaging for further review, characterizing masses concerning for malignancy – and those applications will require many labeled training examples.”
Companies That Have Actually Introduced AI To Radiology
Companies such as Nuance Communications Incorporated, the number one supplier of natural language and voice recognition processing solutions; and Royal Philips, a leading healthcare technology company, have already worked together in order to present AI-based reporting capabilities and image interpretation to radiologists.
Both companies already have AI-powered systems before they joined forces. Nuance Communications Incorporated has its very own PowerScribe 360. The system is a well-developed radiology reporting and communications platform.
On the other hand, Royal Philips has been innovating the work of radiologists through its Illumeo system. Illumeo is an informatics and imaging technology that uses adaptive intelligence in order to enhance and redefine medical images.
Yair Briman, Business Leader, Healthcare Informatics at Royal Philips, said, “Through this collaboration, Philips and Nuance demonstrate our commitment to deliver AI-based technology to help improve radiologists’ daily workflow and bring focus back to the patient.”
He added, “Pressured by time constraints and increased volume of studies in modern practice, radiologists will have access to AI-driven solutions and practical applications, including the integration of the American College of Radiology (ACR) guidelines, to help save time and effort while improving report quality.”
The following are other companies that continue to innovate radiology:
Zebra Medical Vision – The startup company has a medical imaging research platform. Their machine learning technology has the ability to let researchers and scientists create algorithms and imaging knowledge with huge datasets. Elad Benjamin is the Co-Founder and CEO of Zebra Medical Vision. It has an estimated revenue of $1.3 million.
Butterfly Network Incorporated – It was established in 2011 and Jonathan Rothberg is the Chairman and CEO of the company. The startup has a projected revenue of $5 million. The Connecticut-based company transforms non-invasive surgery and medical imaging using learning, semiconductors, and cloud computing technologies.
Radlogics – The startup company headed by Moshe Becker is one of the leading providers when it comes to machine learning solutions and medical imaging analysis. Radlogics has an estimated revenue of $5 million.
The machine learning technology that was able to interpret reports from radiologists still needs a lot of improvement, however, its creation is also a major step forward for the future of the medical industry. Because of the continuous effort made by the researchers, there is no doubt that one day AI systems will become commonplace.