In this AXIS Imaging News exclusive, Aidoc CEO and Cofounder Elad Walach discusses how the company’s artificial intelligence (AI) technology is helping combat physician burnout and labor shortages—issues compounded by the pandemic. 

AXIS Imaging News: What does Aidoc’s AI do?

EIad Walach: Aidoc’s comprehensive AI platform addresses the enterprise’s overall medical AI needs. Aidoc’s AI solutions analyze medical images directly after patients are scanned, suggesting prioritization of time-sensitive pathologies, as well as notifying and activating cross-specialty teams to reduce turnaround time, shorten length of stay, and improve overall patient outcomes.

AXIS: What differentiates Aidoc from other AI for medical imaging companies? 

Walach: Aidoc offers a complete, enterprise-grade AI platform, as opposed to other AI vendors merely offering point solutions. At the core of Aidoc’s platform is its AI OS (operating system), which is automated and always running in the background. This eliminates the need for manual activation or training of new personnel. 

Aidoc’s platform includes the most FDA-cleared triage and notification solutions, alongside a full suite of care coordination solutions. Without an always-on AI OS, other vendors lack the ability to orchestrate multiple algorithms seamlessly into existing workflows, resulting in integration complexity, management overhead, and additional costs.

AXIS: What common healthcare/radiology problems is AI addressing?

Walach: AI solutions address multiple issues in healthcare, including medical imaging, patient management, and resource planning. For medical imaging, AI can help in a variety of ways, but we’ll focus on two: labor shortage and healthcare disparity. 

On the labor shortage front, radiologists today are facing larger case volumes, which often leads to burnout. Radiologists typically work chronologically down a worklist, meaning that scans are read on a first-come-first-serve basis. Without any tool notifying them, they might be delayed in getting to a positive case that’s further down the list. So when faced with a larger case load because of shortages, the chronological style of case-reading can exacerbate care delays. But AI can immediately flag and triage any positive cases, so the radiologist can address them first, getting the patient access to life-saving treatment sooner.

And on healthcare disparity, hospitals which serve smaller communities tend to experience more labor shortages and fewer amenities than larger, urban facilities. In medical imaging, AI can expedite turnaround time for cases with positive findings by prioritizing them in the worklist, ensuring a certain standard of care with the medical imaging experience across the health system. 

AI can also provide stronger health system coordination, organizing and mobilizing a health system’s response teams and facilitating instant, clear communication between them for patient treatment pathways. This can reduce patient transfer delays, reducing the disparity between treatment times that smaller and larger medical centers provide to patients.

AXIS: What is the trajectory of healthcare AI after radiology?

Walach: Radiology is the natural entry point into healthcare because it’s often the first step in evaluating and treating patients. But after a patient is diagnosed, how can hospital departments—which are often siloed and not always synchronized—consult on the best care pathway for that patient? Broken communication is a huge challenge in the care coordination process—how can we bring hospital departments together, not just within a single facility, but across an entire health system? The answer is AI-driven care coordination workflows that encompass more hospital service lines.

Let’s frame this in the context of a hub-and-spoke health system. In the event that a patient with a positive finding for an acute condition requires urgent care, he/she may need to be transferred to a larger facility. This is often guided by manual processes of coordination and can delay treatment, which can be costly to the patient’s health. 

By using AI to coordinate cross-specialty care for a patient, relevant team members are instantly notified and activated, transferring patient information for evaluation and consultation directly on members’ mobile devices to ensure timely and quality treatment. 

AXIS: What is the economic value of AI in radiology?

Walach: Better patient outcomes delivered by AI means enhanced patient satisfaction and higher leapfrog scores, leading to increased revenues from more patients, who feel they can trust their hospital. Finding additional true positive cases leads to earlier hospital interventions for the patient, reducing length of stay. The estimated national average in 2019 for a hospital bed per day totaled $2,500. 

In a study conducted by Yale New Haven Health, AI has been clinically proven to help reduce length of stay. The team found that AI for flagging intracranial hemorrhages reduced the inpatient length of stay by 2.6 days. In addition, medical imaging AI is eligible for NTAP reimbursements for hospitals working with CMS patients.

AXIS: Why are hospitals hesitant to integrate AI?

Walach: Current AI point solutions available in the market require integrating multiple AI tools from multiple vendors. In fact, with point solutions, each pathology requires a new integration of an additional algorithm. As a result, IT is rightfully concerned about implementation, security, and maintenance. On top of that, more and more hospitals are realizing that point solutions don’t enable the adoption of AI at scale, which is what they want to drive as an innovative organization. 

A full-suit platform offering, eliminates these issues, simplifying the process by working with one vendor for any number of AI-powered tools, and seamlessly automating physicians’ existing workflows. This empowers physicians to leverage AI and offer immediate and long-term quality of care, which is in line with the hospital of the future, as well as with the expectations of the patient of the future.