AI in Effect: How AI is Improving Efficiency on the Front Lines of COVID-19

Katelyn Nye, General Manager, Global Mobile Radiography & Artificial Intelligence, speaks with a cardiothoracic radiologist* about the challenges of COVID-19, and how on-device AI for X-ray imaging in the ICU has streamlined imaging efficiency during the pandemic.

No one could have predicted what this year was going to bring and more so, the impact it would have on the healthcare industry. The world watched as healthcare providers worked heroically to serve an influx of critically ill patients with limited staff and resources, all while trying to provide the best care possible to patients.

Today there are more than 55 million COVID-19 cases worldwide, and in the United States, quickly ramping up as another wave of COVID-19 is sweeping across the country with a recent record of over 100,000 new cases in just one day.[1]

While we’ve seen the pandemic upheave many aspects of life and the healthcare industry, it has also served as an unprecedented catalyst for healthcare innovation and technology, accelerating the adoption of artificial intelligence (AI) and analytics. In fact, a recent survey[2] found that even X-ray patients themselves are more amenable to AI being used in their care.

These increased pressures on healthcare providers and adoption of advanced technology led to our own pivot as a team, rethinking how we could make a difference with the AI algorithms. Shifting our development roadmap for the second half of the year, we prioritized bringing technology to market to help our customers where and when they need it most.

Industry-first AI for endotracheal tube (ETT) measurement

This redirect has led to the recent introduction of an industry-first AI algorithm to help clinicians assess endotracheal tube (ETT) placements for critically ill COVID-19 patients. Critical Care Suite 2.0[3] uses AI to automatically detect ETTs in chest X-ray images and provides an accurate and automated measurement of ETT positioning to clinicians within seconds of image acquisition.

With anywhere from 5-15 percent of patients requiring intensive care surveillance and intubation for ventilatory support[4] and the severe complications that can derive from misplacement, this algorithm is designed to help ICU staff identify and correct misplacements quickly. It also showcases the value AI can bring, helping providers efficiently  adjust care for critically ill patients.

The AI solution is one of five included in Critical Care Suite 2.0, an industry-first collection of AI algorithms embedded on a mobile X-ray device for automated measurements, case prioritization and quality control. Already, we have seen the value of on-device AI for X-Ray and we are excited to give providers another tool in the fight against COVID-19.

To learn more about the value of on-device AI for X-Ray imaging in the ICU, I spoke with Dr. Amit Gupta, Cardiothoracic Radiologist at University Hospitals in Cleveland*:

How has AI on mobile X-ray impacted your practice during COVID-19?

Dr. Gupta: As we all know, in the present times, that AI is not a luxury, but a necessity for the high-quality patient care we’re providing in the hospital. We started using Critical Care Suite[5] for pneumothorax detection in early 2020, just before the pandemic hit and we’ve been able to have the firsthand experience of the role AI can play in our practice.

The on-device AI alert that is embedded on the mobile X-Ray system is designed to quickly identify and prioritize critical cases of pneumothorax so that clinicians can respond in a timely fashion.

Other benefits that we’ve noticed with our experience with the technology is that we were able to seamlessly integrate it with the PACS system we have.

My residents have been able to quickly adopt the technology as it’s available in the PACS and they’re able to alert the team in a timely fashion. The AI first preps the image and kind of performs quality control: rotating the X-ray, checking for the completeness, and then also checking if it’s a chest or thoracic radiograph. Then the larger AI algorithm comes into play and identifies and flags any suspected pneumothorax.

For example, in one case we had of a COVID-19 patient where there was a sizeable pneumothorax which was quickly flagged by Critical Care Suite.  The response to that was we were able to quickly address it and inform the clinical team to place a chest tube, and the chest tube was placed in 15 minutes from when we placed the call. You can imagine on a routine day when there are approximately 150 exams to read and it takes almost six hours to get to a typical exam with no flag or alert attached to it.

This is just one example of how radiologists and AI can work together to help in patient care.

How could a solution like the Critical Care Suite 2.0 ETT AI algorithm help treating ICU and COVID-19 patients? 

Dr. Gupta: What excites me and my team is what is on the horizon. We’re currently evaluating AI for endotracheal tube (ETT) measurements, which we hope will be equally valuable tool as we continue caring for critically ill COVID-19 patients.

This is another critical finding we always have to be very attentive too. For example, the tool can help not only highlight the position of the ET tube, but it measures its distance to the carina which is an important landmark when you place ET tubes. If the tube is where it should not be, the information is integrated into the PACS so we could make full use of this technology. This will help us take care of our critical patients more efficiently, especially COVID-19 patients.

What learnings would you share with other radiologists adopting AI in their practice given the current environment?

Dr. Gupta: The role of AI is not to replace radiologists but assist radiologists. Radiologists are a critical part to the success of AI, and we should continue to be part of the development of AI. Now, we can have a second line of defense where the AI tool can help tell us which patient requires immediate attention.

As we’re currently operating with limited staff, scarce resources, and we have residents helping at many times, the team must make split second decisions in emergency situations. This algorithm helps them identify these critical findings and it helps our residents prioritize these types of patients.

This is just one example of how AI and radiologists can work hand in hand and to help each other provide the best patient care.

To hear more from Dr. Gupta on the use of AI on mobile X-Ray, click here to register for an upcoming webinar and live Q&A during this year’s RSNA. 

 

*The following remarks represents the clinical practice and views of Dr. Amit Gupta. Factors that should be considered by clinicians include cleared and approved product labeling and guidelines provided by medically sourced organizations. Dr. Amit Gupta specializes in Cardiothoracic Radiology.

 

[1] Accessed on 11/18/20: https://covid19.who.int

[2] GE Healthcare data on file.

[3] Only available in the United States. Not cleared or approved by the FDA.  Distributed in accordance with FDA imaging guidance regarding COVID-19 public health emergency.

[4] Möhlenkamp S, Thiele H. “Ventilation of COVID-19 patients in intensive care units.” Nature Public Health Emergency Collection. 2020 Apr 20 :1–3.

[5] Not available in all geographies.


Previous
Previous

From Weeks to Hours: One-Stop Breast Care Model Comes to the US

Next
Next

Looking ahead with 20/20: Our vision for the future of breast health