Oct 28, 2025
By Christopher J. Love, Ph.D., Senior Clinical Scientist, Viz.ai
Are the benefits of AI-enhanced clinical workflows a vision or reality?
At Viz.ai, our clinical and medical affairs team works hard to bring visions of AI enhancements in healthcare into reality. Moreover, Viz.ai’s benefit to workflows has already been demonstrated in over 100 published, peer-reviewed abstracts and manuscripts. We’re pioneering AI implementation in clinical settings, and its impact on patients, providers, and healthcare systems is clear. The data-driven evidence from our research, whether from studies published independently or in collaboration with Good Clinical Practice, demonstrates how we live our core values.
As the clinical science leader at Viz.ai over the last three-and-a-half years, I’d like to highlight how we live two of our core values: Time is Brain and Patients First.
“Time is Brain” (and heart, lung, …)
Time to treatment is a key metric impacting patient outcomes. This understanding is readily apparent in acute conditions like stroke for which the concept “time is brain” is well known and multiple studies have been published.
One example is the VALIDATE study, which found a 40-minute time savings in arrival to neurointerventionalist notification involving telehealth consults in two contemporaneous cohorts: 5,559 patients without the use of AI technology, and 8,557 patients with Viz LVO. A recent meta-analysis confirms across-the-board time savings in CT scan to endovascular thrombectomy time, door to groin time, and CT to recanalization time with Viz LVO implementation. The published time savings of AI-enhanced clinical workflows makes AI-supported neuroimaging a recommended part of the ideal foundational requirements for all stroke centers (2023 AHA scientific statement).
The improvement in stroke care enabled by AI technology is personal. I became aware of the debilitating impact of the disease firsthand through a close family member of mine. This experience motivated me to work on a new treatment for stroke for my doctoral thesis and subsequent postdoctoral research. Viz captured my attention for the technology’s enablement of immediate, near-term benefits to stroke care, and I decided to join that mission through clinical research.
But why stop at stroke? Non-acute diseases require proper identification, referral, and specialist coordination post-diagnosis, and AI-enhanced care coordination could assist these efforts. As the company expands with new solutions for cardiology, pulmonology, osteoporosis, and oncology, I’m applying the lessons learned from our prior work in the design of studies that quantify the impact of the new and forthcoming products.
In the recent publication of our multicenter pilot for hypertrophic cardiomyopathy (HCM), a non-acute disease, we found that an AI-enhanced electrocardiogram (ECG) can be successfully integrated into diverse clinical workflows to help identify undiagnosed patients with HCM with promising initial indications for rapid time to diagnostic imaging and diverse patient screening. While contributing real-world data on AI-ECG device implementation, the research enabled us to learn valuable insights that have guided product development and accelerated the realization of new features for clinical decision-making including echocardiogram results and a generative AI summary of data in the electronic health record.
I’ll conclude the topic of external time savings in clinical workflows with an example of how our internal team is keen on rapid dissemination of research results. Within a year of FDA approval for Viz HCM, three abstracts on the technology were presented at ACC; within two years of FDA approval, four abstracts were presented at ACC and manuscripts were published in NEJM AI, JACC EP, Circulation: Heart Failure, and BMJ Heart!
Patients First
Our work is guided by its impact on patients. In fact, patient impact is quantified and tracked internally as a quarterly goal. (Currently, a patient is served by Viz every 5 seconds.) Some of our most edifying and motivating internal updates, called “Vizstories”, are the personal accounts of how our product directly impacted patient care and facilitated good outcomes.
While internal case reports are motivating, a proper assessment of impact to patient outcomes needs rigorous study across many patients. In addition to the numerous studies on workflow time savings, including the ones mentioned in the previous section, two recent examples in acute and non-acute disease highlight and quantify how Viz puts patients first in other ways:
- Recognizing an underlying disease (hypertrophic cardiomyopathy) that is often challenging to detect.
- Over an eight-month period, Viz HCM detected 63 new HCM cases that were previously undetected (Desai et al., 2025).
- Furthermore, a retrospective study revealed the potential for earlier diagnosis (AI-ECG flagged suspected disease in 27 patients over 1 year before clinical diagnosis and up to 16.3 years early) and use of AI-ECG to address racial differences in HCM identification.
- Optimizing stroke evaluation to reduce unnecessary patient transfers from spoke to hub hospitals.
- These transfers are costly and displace patients from their families. A recent study found that the odds of a transfer undergoing an endovascular thrombectomy procedure at the hub after implementing Viz LVO was 1.9 times higher than in the period prior to Viz LVO implementation with a large estimated potential economic benefit to spoke hospitals and payors.
An AI-enhanced clinical workflow that assists disease identification and coordinates proper, timely follow-up has the potential to increase a patient’s access to life-saving therapies and improve outcomes.
Concluding Remarks
We’re grateful for the physicians who share our mission, whether through independent validation or direct collaboration, to study the impact of AI in healthcare. As Viz.ai expands into new disease states, we look forward to continuing to pioneer the study of AI in clinical workflows with the patient at the center. How are you seeing AI impact clinical workflows in your own practice?