Written by 
Dr. Ron Boucher serves as the Chief Medical Officer of Teleradiology at Experity, a software and services company focused on the U.S. urgent care market.
Experity offers an integrated operating system that includes electronic medical records, practice management, patient engagement, billing, teleradiology, business intelligence, and consulting solutions. Nearly half of the urgent care clinics in the U.S. use Experity’s platform. Experity’s teleradiology overread services address the shortage of radiologists by providing clinics with extended support. These services are recognized for their industry-leading turnaround times, 99.97% accuracy, and real-time access to radiologists. The integration of AI into scan reads aims to further enhance both efficiency and accuracy in care delivery.
For readers who are unfamiliar with this term, what is Teleradiology?
Teleradiology is a medical service that enables radiologists to provide clinical interpretation services on X-rays, Ultrasounds, and other diagnostic imaging without needing to be physically present with the patient. In the case of urgent care, the teleradiologist functions as an extension of a clinic, offering faster turnaround times, real-time consultation, and even sharpened accuracy.
With teleradiology, patients receive faster and more precise care, clinic staff save time by receiving timely responses, and clinic providers can confidently rely on diagnoses reviewed by board-certified radiologists. Additionally, clinics that produce a small volume of radiology exams can save a significant amount of money by not having a dedicated radiologist onsite and only pay for the exams performed. This is particularly important whenever a subspecialist radiologist is needed, typically only available at larger institutions and academic centers.
Could you elaborate on the main challenges you’ve encountered with AI integration in teleradiology, and how have you addressed these challenges?
The challenges we’ve faced so far have been primarily clinical, with the largest being a small group of radiologists that are not ready to incorporate AI in their workflows. This is mostly due to clinicians wanting to understand the technology and maintain their independence as providers and utilizing traditional practices. As the technology experts behind the AI integration, we understand that AI is meant to facilitate and improve the standard workflow, not replace the role of radiologists. With the continued advancements being made to AI and other technologies that enable providers to improve their practices, we urge providers to maintain an open mindset toward the tools that can help make their jobs easier and, in tandem, deliver more efficient and better care.
Another challenge is trying to understand the strengths and weaknesses of the fracture detection software with which we have integrated. Once those are identified, the radiologist, as they gain more confidence in the software, can adjust the workflow to improve the overall accuracy and care delivery process. It’s our job at Experity to show and advocate for the true value that AI brings to radiologists’ workflows once those initial adoption challenges are overcome.
Why do you believe that adopting AI in healthcare settings, particularly in radiology, is more beneficial than avoiding it?
Most hesitancy surrounding AI stems from concerns of replacing humans, but in the case of teleradiology, radiologists are still required to interpret results. AI augments the radiologist’s tasks, but board-certified clinicians are still required to oversee the process. Both speed and quality of care are drastically increased with AI’s integration into radiology overread services.
One key advantage of AI in this capacity is the significant improvement in the efficiency and accuracy of imaging interpretation. For instance, our AI software assists radiologists by identifying fractures in adults and pinpointing potential injury locations – both of which are particularly useful in teleradiology where patient histories may be incomplete or when the study is sub-optimally performed or positioned
AI helps reduce the time radiologists spend searching for abnormalities, which leads to a 15-20% increase in speed. This efficiency allows for faster patient care without compromising quality. In fact, the quality of reads with this integration has improved by about 40%, as AI helps prevent missed diagnoses, ensuring more accurate and reliable results. Patient expectations for quality and efficiency will only increase in the future, especially for urgent care, so choosing to embrace AI and maximize the support it offers will help to best meet those needs.
How has AI integration in teleradiology specifically contributed to better patient outcomes?
AI not only increases speed on workflow, but also improves patient care by enhancing the detection and diagnosis of fractures. These fractures might otherwise be missed, so AI is significantly increasing the possibility of better outcomes for patients. Systems that utilize AI can identify additional fractures that radiologists might overlook due to their subtlety or because they occur alongside more obvious injuries. This capability is crucial for comprehensive patient care and seeing the full picture, regardless of medical records being available.
AI in teleradiology has also contributed to faster turnaround times. This speed is particularly beneficial in urgent care settings where quick diagnosis and treatment are essential. Physicians benefit from the rapid availability of accurate diagnostic information, enabling them to treat patients more efficiently and discharge them quicker, thus improving overall patient satisfaction and clinic success.
In what ways has AI technology improved operational efficiencies and accuracy in radiology readings?
Prior to AI, clinics and practices would work to treat and release patients as efficiently as possible, but the quality of care was jeopardized with this rushed approach. Now with a national shortage of radiologists, finding ways to streamline operations while maintaining quality of care is crucial to the success of a practice. By improving turnaround times and maintaining high-quality standards, AI is helping the teleradiology industry thrive by meeting its high demand for quick and precise diagnoses.
Patients will ultimately seek care from those who can deliver a satisfactory balance of quality and efficiency – both innate qualities of urgent care that are only amplified with the use of AI. At Experity, our teleradiology overread services have an industry-leading turnaround time with 99.94% accuracy rates. Our AI technology helps radiologists identify equivocal and obscure abnormalities that otherwise may not be indicated by the patient’s history, exam, or records, expanding the accuracy of reads with an additional component of timeliness.
What do you see as the future role of AI in healthcare and how can healthcare providers prepare for these changes?
When I attended the Radiology Society of North America’s conference this year, AI took up about 30% of the floor space. AI is the direction we’re headed in, and it can impact every aspect of our workflows as radiologists. For those who choose to carry on and ignore AI, many practices will eventually become obsolete. The physicians and practices who choose to embrace technology will be the survivors of the transition. For instance, when teleradiology services became mainstream, this process will be heavily reliant on leveraging advanced technology. Radiologists will need to adapt to the changing landscape of AI integration. AI will not replace radiologists, but instead will enhance their roles as a clinical provider by improving patient care and quality while reading more efficiently and accurately. Radiologists who do not adopt AI in their workflows in some manner will be obsolete.
How do you balance the benefits of AI automation with the need for human oversight in radiological assessments?
Our goal with integrating AI into our teleradiology services is for it to be supplemental and help our urgent care partners deliver the best care possible. AI does not involve emotions or understanding a patient’s history, so those components need to be manually integrated with the history and knowledge provided by a clinician. One Danger of AI is a clinician or patient taking the AI result at face value without the professional insight of a radiologist or clinical expert to ensure the output is accurate and verified.
Mistakes can happen, which is why maintaining human oversight is essential for the solution’s integration. The algorithm can mark false negatives or positives, but its ability to point out areas of interest in the Radiology exam reduces the human error rates more effectively and outweighs reading exams without AI involved.
Can you discuss any regulatory hurdles related to the use of AI in healthcare and how Experity is navigating these?
I’m very optimistic about AI and the role it will play in Radiology. However, it will take time to understand the legal implications. Regulations surrounding AI are going to drastically change over the next few years, and this drives meaningful resistance among radiologists. If an AI product identifies an abnormality and the physician disagrees with it, how does it impact a lawsuit if something were to go wrong in the care delivery process?
Without regulations, the default leads to tort law, which is not optimal to ensure patient safety. Physicians are ultimately responsible for the diagnosis and image reporting. There are not any set legal ramifications currently, which can lead to uncertainty from both patients and providers as cases occur. Radiologists are the licensed physicians delivering care to patients, so there are gray areas that need to be explored and addressed as AI becomes more prominent across the industry.
Can you discuss how AI in teleradiology has impacted access to healthcare services, particularly in underserved or rural areas?
As I previously mentioned, the specialty of Radiology is an area of healthcare that is feeling more severe effects of the national physician labor shortages. Teleradiology alone provides new opportunities for patients to receive care in rural areas with a lack of medical resources and care available. Partnering with a third party to provide the professional imaging interpretation process vastly expands a clinic’s capabilities and increases the type and quality of care they deliver. It brings subspecialty care to their patients.
With AI being integrated into these more rural practices, the quality and efficiency of care can be prioritized more and even standardized to the care a patient would receive in a more urban setting. Not only is the care available more extensive, but the accuracy and efficiency can also be improved.
Thank you for the great interview, readers who wish to learn more should visit Experity.
