Artificial Intelligence Powered X-Rays?

GE Healthcare, a world leader in diagnostic imaging agents and radiopharmaceuticals, is an American business that operates as a conglomerate (subsidiary of General Electric) with sites spanning the planet, with 6 in the United Kingdom alone. From CT imaging machines to MRI dyes, GE Healthcare is a giant manufacturer and distributor in the healthcare industry, employing over 54,000 personnel and racking up $18.3 billion in revenue in 2016. In recent times. the company has orientated its efforts towards research purposes, such as with its innovative movements in patient monitoring systems, drug discovery, biopharmaceutical manufacturing and disease research, with a slow extending input of AI to ensure faster and more effective care plans.

 
GE Singa series MRI Scanner, in action at the Narayana Multi-speciality Hospital in Jaipur, India. Credit: Wikipedia

GE Singa series MRI Scanner, in action at the Narayana Multi-speciality Hospital in Jaipur, India. Credit: Wikipedia

 

These research-based efforts were seen earlier last week, as a brand new ‘artificial intelligence powered X-ray device’ that GE Healthcare claimed could reduce the time interval in detecting a collapsed lung by a staggering factor of 32, from 8 hours to 15 minutes, was cleared by the Food and Drug Administration (FDA). GE Healthcare CEO Kieran Murphy, in an interview with CNBC, said, “The health-care industry is producing huge amounts of data from images to digital health records, we strongly believe that you have to turn that data into information and insight to improve outcomes.” The device, coined the ‘Critical Care Suite’, utilizes artificial intelligence-based algorithms to scan X-ray images and detect cases of collapsed lung. One can only assume that deep learning algorithms are put in use to identify trends and patterns in the data sets that Kieran Murphy mentions, and is assessed in its efficacy to spot recurrent themes in a new sample of data. When the AI system has reason to suspect any condition, the scan is sent off to a radiologist as a confirming measure, and as such this is another example of AI-assisted healthcare, not AI-led, as is the reasonable role of artificial intelligence given its youth in the healthcare industry. However, what is a collapsed lung? Otherwise known as pneumothorax, it occurs when air escapes from the its reservoirs in the lungs, filling the area between your chest wall and lung (i.e. pleural cavity) , causing a heightened external pressure on the lung, causing difficulties during inspiratory movements due to the external pressure being a brick-wall to the lung’s expansion. This can be fatal, unless the patient is treated immediately, especially in cases of tension pneumothorax.


 
a0217945_pneumothorax_x-ray_science_photo_library_high.jpg

Note the example of pneumothorax on the left, which is fairly distinctive shape and made more obviously by the clear disparity in the relative sizes of the left and right side of the lung. Credit: MSD Manual

As for artificial intelligence, the admission of this product by the FDA is a significant milestone for AI in the healthcare industry, and will hopefully be the start of a revolution that sees doctor’s being heavily assisted in providing a fast and accurate diagnosis, and helping out in a time where doctors, nurses, and hospitals are being pushed to their very maximum in treating a growing population with limited resources. With the device speculated to reach markets some time this year, and GE Healthcare stating that the platform could detect ‘nearly all large pneumothoraxes’, ‘75% of small pneumothoraxes’, and drastically cut down false alerts in their studies, and allow for PACS (picture archiving and communication systems - the mode of data for radiologists) prioritisation, as well as the system being able to check for errors in acquisition, radiologists will be aided in cutting down review time, especially for the most severe (STAT priority) cases that may be difficult to account for.

 
Kieran Murphy, CEO of GE Healthcare, appears in a CNBC televised interview. Credit: CNBC

Kieran Murphy, CEO of GE Healthcare, appears in a CNBC televised interview. Credit: CNBC

 

The following are quotations from key figures, within and outside of the company.

  • University of California San Francisco surgeon Dr. Rachael Callcut said, “When a patient X-ray is taken, the minutes and hours it takes to process and interpret the image can impact the outcome in either direction. AI gives us an opportunity to speed up diagnosis, and change the way we care for patients, which could ultimately save lives and improve outcomes."

  • Jie Xue, president of GE Healthcare’s X-ray division said, “Currently, 62 percent of exams are marked ‘STAT’ or for urgent reading, but they aren’t all critical. Not only does Critical Care Suite flag images with a suspected pneumothorax with impressive accuracy and enable radiologists to prioritize those cases immediately, but it also makes AI accessible. Our embedded AI algorithms offer hospitals an opportunity to try AI without making investments into additional IT infrastructure, security assessments or cybersecurity precautions for routing images offsite."


A great big thank you to GE Healthcare, Kieran Murphy, Rachael Callcut and Jie Xue for their contributions to this article, and the field of healthcare. Especially given the enormity of GE Healthcare, it is great to see larger distributors such as GE Healthcare taking the initiative to ensure AI technologies can be of benefit around the globe, as they have displayed by their innovation in fields such as gene therapy.

Article thumbnail credit: How Do X-Ray Machines Work?, TECH-FAQ [click here for page]. Note that this article has not contributed to any of the written content of my article, and that you click the above link at your own discretion as the page has not been checked by our team. Thank you to our readers for your continued support, and to anyone who contributes to our fantastic team to keep our services in orderly fashion.