Competitor Teardown — Viz.ai

Capstone neurology AI landscape teardown — Viz.ai Audience: capstone team building the Week 8 AI-in-healthcare landscape map. Date: July 2026. Framework: 4-step structured teardown. All quotes verbatim from primary sources.

1. Surface scan

1.1 Company summary (one paragraph)

Viz.ai is the San-Francisco-and-Tel-Aviv–based, AI-powered care coordination platform for time-sensitive acute conditions, founded in 2016 by neurosurgeon Dr. Chris Mansi and machine-learning postdoc David Golan. Starting with stroke (Viz LVO, the first FDA-cleared AI triage software for large vessel occlusion in 2018), Viz.ai has expanded its FDA-cleared algorithm library to “more than 50 FDA-cleared AI algorithms” spanning neurovascular (stroke, aneurysm, ICH, subdural), cardiovascular (HCM, amyloidosis, RV/LV ratio), vascular (AAA, aortic dissection), and pulmonary (PE) pathways — sold as modular suites (Viz Neuro, Cardio, Vascular, Trauma, Radiology, Pulmonary) bundled under the Viz.ai One® enterprise platform. The company is deployed in “nearly 2,000 hospitals” across the U.S. and Europe covering “more than 230 million lives” (as of January 2026), reports ~ $40 M r e v e n u e in 2023 * * g r o w in g t o w a r d * *$ 100M ARR, $291.5 M t o t a l r ai se d * * a t a * *$ 1.2B valuation (Series D April 2022), and reached profitability in its healthcare business in 2025 (per Sacra and Viz.ai’s own press release). Viz.ai holds a unique regulatory moat: Viz LVO was the first AI software ever granted a CMS New Technology Add-on Payment (NTAP) — up to $1,040 per eligible Medicare use — which anchors its enterprise pricing and ROI story.

Sources: Viz.ai homepage, Sacra company profile, Viz.ai 2025 close press release.

1.2 Headline features (verbatim)

From viz.ai homepage and Viz.ai One brochure page:

“Viz.ai closes the gaps between patients, clinicians, and life-saving treatments with its leading-edge, AI-powered care coordination platform — driven by over 50 advanced, FDA-cleared algorithms.”

“Viz.ai One® is the AI-powered care coordination solution designed to revolutionize the way patient care is delivered. Supported by robust real-world evidence, Viz.ai One significantly saves time and improves clinical and economic outcomes by autodetecting suspected diseases earlier and streamlining workflows. With tailored integration with your EHR, PACS, and worklist, you can implement AI at scale with minimal IT resources.”

Product suite (from viz.ai/vizai-one):

Viz Neuro — “AI-powered suite of solutions tailored to accelerate the detection and treatment of suspected neuro diseases.”
Viz Cardio — “AI-powered suite of solutions developed to meet cardiovascular needs.”
Viz Vascular — “AI-powered suite of solutions designed to transform patient outcomes in vascular medicine.”
Viz Trauma — “AI-powered capabilities tailored to meet today’s rigorous trauma center guidelines.”
Viz Radiology — “AI-powered suite of solutions designed to accelerate patient diagnosis and treatment.”
Viz Connect — “A proprietary communication and workflow tool that unites multidisciplinary specialists and care teams across therapeutic areas.”
Viz Pulmonary Suite (new May 2026) — “the first comprehensive AI-powered solution dedicated to pulmonary care delivery.”
Viz Oncology Suite (launched May 2025; partnered with NCCN).
Viz Assist (launched October 2025) — multimodal AI agent combining ambient listening, EHR data, and FDA-cleared imaging AI.
Viz Life Sciences — pharma/medical-device partnerships for patient finding.

1.3 Pricing model

No public pricing. Viz.ai publishes no /pricing page; every CTA is “Request a demo” or “Speak to an expert.” The model is:

Subscription SaaS, per-facility (per-hospital), annual, modular by disease suite, scaled by the number of FDA-cleared algorithms licensed and the population served.
Reimbursement offset: Viz LVO was the first AI software ever granted a CMS New Technology Add-on Payment (NTAP) — “up to $1,040 per eligible use” in suspected-stroke patients. (Source: Viz.ai NTAP press release, JNIS NTAP commentary.)
Third-party estimate (snippet only): Per Out-of-Pocket’s Nikhil, third-party analyses of CMS filings have cited *“~ $25 k / y r p er h os p i t a l " * * a s an e x am pl es t r o k e - m o d u l e f i g u r e — b u t : * " V i z . ai c ha r g es a s u b scr i pt i o n t o u se t h e i r m o d e l . T h ecos t i s n o tw ha tw a s in c l u d e d a s^{'} an e x am pl e^{'} in t h e C M S d oc u m e n t s ($ 25k/yr per hospital), and I [Luke Oakden-Rayner] suspect it’s higher.” (Sources: Out-of-Pocket article, Luke Oakden-Rayner blog.) Treat this as third-party-reported, NOT official.
Sacra estimates ~ $40 M r e v e n u e in 2023 o na p a t h t o$ 100M ARR; healthcare business hit profitability in 2025. (Source: Sacra.)

1.4 Funding & company basics

Datum	Value	Source
Founded	2016 (founders: Chris Mansi MD neurosurgeon + David Golan ML postdoc, met at Stanford)	Sacra
HQ	San Francisco, CA (+ Tel Aviv)	Sacra
Total raised	*$291.5 M a cr oss 10 r o u n d s * (S a cr a) / * $ 252M across 7 rounds* (Tracxn) — discrepancy is rounding + undisclosed growth rounds	Sacra, Tracxn
Series D	$100 M a t$ 1.2B valuation, April 2022, led by Tiger Global + Insight Partners (with Greenoaks, Kleiner Perkins, Threshold, CRV, Innovation Endeavors, Susa)	Cardiovascular News, LinkedIn announcement
Series C	$71M March 2021, led by Scale Venture Partners + Insight Partners	Viz.ai press release
Series A	July 2018	Sacra
Growth debt	$40M from CIBC Innovation Banking	Forge Global
Customer count	”1,700+ hospitals” (homepage tagline, Viz.ai); “nearly 2,000 hospitals” and “more than 230 million lives” (Jan 2026 press release, Viz.ai closes 2025); progressed from 1,400+ (2022–2023) to 1,500+ (Feb 2024 ICH press release) to 1,700 (homepage) to 1,800 (viz.ai/vizai-one) to ~2,000 (Jan 2026) — customer count climbs ~25–33% YoY
Lives covered	220M+ (2022–2024 press releases) → 230M+ (Jan 2026)	Viz.ai
Profitability	Healthcare business achieved profitability in 2025; Sacra updated Mar 10, 2026	Sacra, Viz.ai closes 2025
Headcount	approximate; LinkedIn estimates not pulled in this teardown
Annual publications	”more than 120 peer-reviewed publications and abstracts” (Jan 2026)	Viz.ai closes 2025

1.5 Customer quotes (verbatim)

5+ verbatim customer quotes (primary or near-primary sources):

Dr. Don Frei, MD — Director, Neurointerventional Surgery, Radiology Imaging Associates / RIA Neurovascular / Swedish Medical Center (Denver). On Viz.ai home page (https://www.viz.ai/):

“Viz.ai alerts my team to potential LVOs in our network and allows me to quickly view them on my phone. This is the new standard for stroke care.”
Dr. Raul Nogueira — Director, Marcus Stroke & Neuroscience Center, Grady Health. FeaturedCustomers (https://www.featuredcustomers.com/vendor/vizai):

“Viz CTP uses artificial intelligence to automatically correct for motion artifact and quickly deliver perfusion maps I can trust.”
Heidi Woessner — Managing Director, Neuroscience Piedmont Healthcare. FeaturedCustomers (https://www.featuredcustomers.com/vendor/vizai):

“This Viz.ai CS team is so wonderful to work with and helped us expedite our implementation. They continue to be a resource to us after our implementation has been completed.”
Dr. Thomas Devlin — Director, Marcus Stroke & Neuroscience Center, Grady Health (also CHI Memorial). FeaturedCustomers (https://www.featuredcustomers.com/vendor/vizai):

“What we’ve experienced with Viz is a fundamental change in how we do stroke care. It’s a change in the paradigm from push to pull. With Viz, images are automatically [pushed, analyzed, and routed].”
Dr. Adam Arthur — Professor of Neurosurgery, University of Tennessee Health Sciences Center / Semmes-Murphey Clinic. Viz.ai NTAP press release (https://www.viz.ai/news/viz-granted-medicare-ntap):

“Since using Viz at our center, we have seen a decrease in time to treatment and an improvement in patient outcomes. This software should be considered now the gold standard for all systems treating stroke.”
Dr. Alexander Heard — Chief Medical Officer, Adventist Health Rideout (UC Davis partner hospital). UC Davis Health press release (https://health.ucdavis.edu/news/headlines/ai-tech-helps-partner-hospital-reduce-stroke-transfer-time-by-half/2024/10):

“Viz.ai has decreased our door-in-door-out times by nearly 50%. This not only enhances the speed at which patients receive care but also significantly improves their outcomes. The impact on our stroke patients has been truly transformative.”
Dr. Kwan Ng — Director, UC Davis Health Comprehensive Stroke Center. UC Davis Health press release (https://health.ucdavis.edu/news/headlines/ai-tech-helps-partner-hospital-reduce-stroke-transfer-time-by-half/2024/10):

“Being able to connect stroke experts at UC Davis Health with ED providers at Rideout rapidly improves access to life-changing treatments and outcomes for stroke patients. Time to treatment is brain saved.”
Dr. Theresa Sevilis, DO — Director of Clinical Research, TeleSpecialists (VALIDATE study co-author). Viz.ai VALIDATE press release (https://www.viz.ai/news/large-real-world-multi-center-study-demonstrates-viz-ai-platform-saves-critical-minutes-in-stroke-care):

“In the world of stroke treatment, the saying ‘time is brain’ comes from the fact that when brain tissue is deprived of oxygenated blood, approximately two million neurons die every minute… VALIDATE data show that use of the Viz platform resulted in a clinically important decrease in the time it took for large blood vessel occlusion recognition and contact with the interventional team, which in turn may translate to fewer neurons dying and, ultimately, better patient outcomes.”
Caezar G. Jara, BSN, CEN, SCRN, ASC-BC — Stroke Program Manager, Adventist Health + Rideout (ISC 2026 study). Viz.ai ISC 2026 press release (https://www.viz.ai/news/viz-ai-study-demonstrates-44-reduction-in-interfacility-stroke-transfer-times):

“Our initiative was driven by the need to eliminate manual bottlenecks that delay stroke care in regional settings. By integrating Viz.ai, we replaced a complex transfer process with an automated, real-time workflow, helping patients reach life-saving intervention sooner.”
Dr. Jayme Strauss — Chief Clinical Officer, Viz.ai (internal; useful for talking-points language). Viz.ai VALIDATE press release:

“The first multi-center study involving a large number of patients, VALIDATE compared results from the same time period between hospitals that used the Viz.ai platform and those that did not have AI software, thus eliminating the potential confounding effect of changes in workflow that occur in hospitals over time. This tightly controlled scientific study clearly illustrates the tremendous impact of our technology.”
Dr. Jason Davies, MD, PhD — Associate Professor of Neurosurgery and Biomedical Informatics, SUNY Buffalo. Viz.ai Subdural page (https://www.viz.ai/subdural):

“Viz Subdural allows us to detect both acute and chronic subdural hemorrhages to better identify early treatment pathways. Having an algorithm that identifies both can allow us to take better care of our patients.”

2. What they compute

2.1 Architecture diagram (text)

                              HOSPITAL PACS / CT SCANNER
                              (Siemens, GE, Philips, Canon, etc.)
                              │
                              │ DICOM push / HL7
                              ▼
   ┌──────────────────────────────────────────────────────────────┐
   │              VIZ.AI CLOUD (HIPAA-compliant)                  │
   │  ┌─────────────────────────────────────────────────────────┐  │
   │  │  PRE-PROCESSING: DICOM receive, motion correction,        │ │
   │  │  anonymization, slice selection, CTA reconstruction      │ │
   │  └─────────────────────────────────────────────────────────┘  │
   │                            │                                   │
   │                            ▼                                   │
   │  ┌─────────────────────────────────────────────────────────┐  │
   │  │  AI INFERENCE (parallel, by suite):                      │  │
   │  │  • Viz LVO     — CTA → ICA/MCA-M1 binary LVO flag      │  │
   │  │  • Viz ICH     — NCCT → ICH volume + midline shift      │  │
   │  │  • Viz CTP     — CTP → rCBF/rCBV/MTT/Tmax maps         │  │
   │  │  • Viz ANX     — CTA → cerebral aneurysm ≥4mm           │  │
   │  │  • Viz SDH     — NCCT → subdural collection quant.      │  │
   │  │  • Viz RV/LV   — CTPA → RV:LV diameter ratio           │  │
   │  │  • Viz AAA/AD  — CT/A → aortic aneurysm / dissection    │  │
   │  │  • Viz HCM     — 12-lead ECG → hypertrophic cardiomyop. │  │
   │  └─────────────────────────────────────────────────────────┘  │
   │                            │                                   │
   │                            ▼                                   │
   │  ┌─────────────────────────────────────────────────────────┐  │
   │  │  CARE COORDINATION LAYER                                │  │
   │  │  • Mobile push (specialist iOS/Android app)              │  │
   │  │  • Desktop viewer (PACS integration)                    │  │
   │  │  • HIPAA-compliant group chat                          │  │
   │  │  • Worklist triage (radiology workflow)                │  │
   │  │  • EHR + on-call scheduler hooks                        │  │
   │  │  • Viz Connect (cross-specialty communication)          │  │
   │  │  • Viz Assist (ambient listening + LLM summary)         │  │
   │  └─────────────────────────────────────────────────────────┘  │
   └──────────────────────────────────────────────────────────────┘
                              │
                              ▼
                  ON-CALL SPECIALIST'S PHONE
                  (text alert + compressed image preview →
                   confirm on diagnostic viewer → engage team)

Source: synthesized from Viz.ai Indications-for-Use page, Sacra, and individual product pages.

2.2 Per-feature I/O table (verbatim from FDA indications-for-use text where possible)

All text in quotes is direct from https://www.viz.ai/indications-for-use unless otherwise marked.

Feature	Input (data)	Compute (algorithm)	Output (clinician sees)	Latency	FDA 510(k) cleared
Viz LVO (flagship, 2018)	CT angiogram (CTA) of the brain, terminal ICA + MCA-M1 vessels	”AI algorithm to analyze images for findings suggestive of a pre-specified clinical condition”	Mobile push to neurovascular specialist: “a suspected large vessel occlusion has been identified and recommends review of those images”; compressed preview in mobile app for informational purposes (NOT for diagnostic use)	Real-time, ~5–6 min median CTA-to-notification (cited: “6 min average from patient scan to provider alert” — UC Davis press release)	De Novo DEN180041, Feb 2018 + multiple subsequent 510(k)s (e.g., K223042 for Viz LVO/ContaCT algorithm update — FDA)
Viz ICH	Non-contrast CT (NCCT) of the brain, acute setting	AI image classifier for ICH	”Notifications to a neurovascular or neurosurgical specialist that a suspected intracranial hemorrhage has been identified”	Real-time (~minutes)	510(k) cleared (date not in fetched 510(k) text; marketing claim)
Viz ICH Plus (Feb 2024)	NCCT brain	”to automate the process of identifying, labeling, and quantifying the volume of segmentable brain structures on non-contrast computed tomography (NCCT) images… analyzing intracranial hyperdensities, lateral ventricles and midline shift”	Volume measurements of brain bleeds for neurosurgeon decision-making	Real-time	FDA 510(k) cleared Feb 8, 2024 (press release)
Viz CTP	CT Perfusion (CTP), DICOM-compliant	”image processing software… runs on a standard ‘off-the-shelf’ computer or a virtual platform, such as VMware… image processing, analysis, and communication of computed tomography (CT) perfusion scans of the brain"	"CT perfusion parameter maps from a raw CTP scan are exported in standard DICOM format and may be viewed on existing radiological imaging viewers” (rCBF, rCBV, MTT, Tmax)	Minutes	510(k) cleared
Viz ANEURYSM (ANX)	CTA of head	”AI to analyze images and highlight studies with suspected aneurysms in a standalone application for study list prioritization or triage in parallel to ongoing standard of care”	Worklist prioritization flag; “compressed preview images that are meant for informational purposes only and not intended for diagnostic use”; “limited to detecting aneurysms at least 4mm in diameter”	Real-time	FDA 510(k) cleared Feb 24, 2022 (press release) — algorithm accuracy cited: 94% on 528 CTAs / 674 aneurysms, U Toronto (per same press release)
Viz SDH (Subdural) (Viz Subdural Plus FDA-cleared June 2025)	NCCT head	”AI algorithm to analyze images for findings suggestive of a… subdural hemorrhage”	Notification to neurovascular/neurosurgical specialist; “automated volume, thickness, and midline shift measurements” per Sacra	Real-time	FDA 510(k) cleared (original Viz SDH + Viz Subdural Plus June 2025); CNN-based per Viz AI: “Viz Subdural CNN performed exceptionally well at identifying and quantifying key features of subdural hemorrhages in an independent validation imaging data set” (https://www.viz.ai/subdural)
Viz RV/LV	Volumetric CTPA (CT pulmonary angiogram)	“designed to measure the maximal diameters of the right and left ventricles of the heart from a volumetric CTPA acquisition and report the ratio of those measurements”	Annotated images with RV:LV ratio — used for PE severity assessment	Real-time	FDA 510(k) cleared (CINA CHEST lineage)
Viz PE & Viz Aortic Dissection	Chest CT angiography (PE) + Chest/thoraco-abdominal CTA (AD)	CINA CHEST: “radiological computer aided triage and notification software… AI algorithm to analyze images and highlight cases with detected PE and AD on a standalone Web application in parallel to the ongoing standard of care image interpretation”	Web flag for PE or AD	Real-time	FDA 510(k) cleared
Viz AAA (Abdominal Aortic Aneurysm)	CT angiography, abdomen	”AI to automatically search for the presence of an abdominal aortic aneurysm from computed tomography angiography (CTA) from any scanner in a hospital network”	Mobile/email alert to vascular surgery team; automated care routing	Real-time	First-of-its-kind FDA 510(k) clearance March 21, 2023 (press release)
Viz HCM (Sept 2024)	12-lead ECG recordings (from compatible ECG devices, patients ≥18 y)	“to analyze recordings of 12-lead ECGs… detecting signs associated with hypertrophic cardiomyopathy (HCM), and allowing the user to view the ECG and analysis results”	ECG analysis card in mobile/desktop	Real-time	510(k) cleared
Viz Oncology Suite (May 2025)	EHR + imaging + pathology (multimodal)	“end-to-end AI-powered oncology care coordination that surfaces high-risk cancer patients earlier and supports guideline-directed interventions”; embedded NCCN guidelines for breast/prostate/bladder cancer	Risk stratification + guideline-prompts to oncology team	Near-real-time (workflow integrated)	Suite-level clinical decision support; FDA SaMD pathway expected
Viz Assist (Oct 2025)	Ambient audio + EHR data + imaging AI outputs	”multimodal AI agent platform, combines ambient listening, EHR data, and FDA-cleared imaging AI to generate clinical summaries and support documentation”	AI-generated clinical notes/documentation	Real-time (during encounter)	TBD (per Sacra: “plans to expand toward coding recommendations”)
Viz Subdural ROI (“Reduced Hospital Costs”)	—	—	“+ $33 k H os p i t a l C os t, + 132$ 4.8B Annual Payer Impact” (per viz.ai/subdural)	—	—

2.3 Published evaluations (key real-world studies, verbatim numbers where possible)

Study / Citation	Population / Design	Headline result
VALIDATE-ED study (Viz.ai × TeleSpecialists), presented ISC 2023, press release Feb 8 2023 — https://www.viz.ai/news/large-real-world-multi-center-study-demonstrates-viz-ai-platform-saves-critical-minutes-in-stroke-care	Retrospective comparison of stroke consults at 166 facilities in 17 states, n = 14,116 patients (Viz hospitals n=8,557 vs non-AI n=5,559)	Median door-to-neurointerventionalist notification: 50 min with Viz.ai vs 89.5 min without (p<0.001) — a 39.5-minute reduction. Independent of thrombectomy-capability of the center.
Adventist Health + Rideout (UC Davis partner), ISC 2026 — https://www.viz.ai/news/viz-ai-study-demonstrates-44-reduction-in-interfacility-stroke-transfer-times	Regional primary stroke center, pre/post Viz.ai deployment	Average DIDO time decreased from 202 min to 113 min (44% reduction) — now exceeds Joint Commission 120-min national benchmark by ~6%. CTA-to-detection down 84%. Care-team notification: 45 min → 7 min.
Hassan et al. 2020 (cited in JNIS NTAP commentary as ref 11): Viz LVO “saves 66 min on average” across full stroke workflow	—	Per JNIS: “Implementation of Viz LVO has been demonstrated to save 66 min on average, suggesting a significant return on investment for CMS.” (JNIS)
Sacra-reported (Mar 2026 update): Real-world March 2026 data showing “Viz.ai reduced door-in-door-out transfer time for LVO stroke patients by 44%, from 202 minutes to 113 minutes, with CTA-to-detection time falling 84% and care-team notification time dropping from 45 minutes to 7 minutes”	—	Confirms ISC 2026 numbers; adds that Viz.ai “increasing surgery volumes 50-60% at large hubs”
Sarhan et al. systematic review + meta-analysis (Translational Stroke Research, May 2025) — https://pmc.ncbi.nlm.nih.gov/articles/PMC12596299/	12 studies, 15,595 patients; Viz.ai vs pre-AI control	Viz.ai associated with lesser CT-to-EVT time (SMD −0.71), lesser door-to-groin-puncture time (SMD −0.50), lesser CT-to-recanalization time (SMD −0.55), lesser door-in door-out time (SMD −0.49) — all p<0.001. Patient clinical outcomes (mortality, mRS, ICH) not statistically significantly different — p>0.05.
Scoping Review of Brainomix / Aidoc / RapidAI / Viz.ai (Medicina, Mar 2026) — https://www.mdpi.com/1648-9144/62/3/582	PRISMA-ScR mapping, 29 studies (2019–2025)	Viz.ai associated with door-to-puncture time reductions (11–25 min); platforms show proximal LVO sensitivity 78–97%; “RapidAI is frequently mapped using historical perfusion trial parameters; however, volumetric discrepancies with platforms like Viz.ai indicate outputs are not interchangeable.” “Brainomix shows extensive validation for automated NCCT ASPECTS in triage. Aidoc demonstrates operational advantages via worklist prioritization.”
Viz LVO sensitivity/specificity — historical Viz LVO multi-center study (cited in Series C press release, n=2,544 patients across 139 hospitals, multi-vendor scanners)	Largest LVO AI validation at the time	96% sensitivity, 94% specificity for LVO detection; median time-to-notification 5 min 45 sec across all sites
Viz.ai Connect / pre-AI workflow Sacra citation	—	“Viz.ai’s stroke module cut time-to-specialist-notification from ~58.7 minutes (standard of care) to ~7.3 minutes — and at large stroke hubs reportedly lifted thrombectomy/surgery volume 50–60%” (UsagePricing)
Viz.ai click-through rate	Internal product metric	”90% click-through rate on clinical alerts and workflows” (Viz.ai closes 2025)
Viz.ai notification click-through (UCDavis 2024)	—	“the average time from patient scan to provider alert is six minutes, across nationwide users” (UC Davis)

2.4 FDA clearances (chronological, with 510(k) numbers where confirmed)

Year	Module	FDA pathway	Source / Evidence
Feb 2018	Viz ContaCT / Viz LVO (original)	De Novo DEN180041 — first FDA-cleared AI triage software (stroke)	PR Newswire; historic
2022	Viz ANEURYSM (ANX)	510(k) clearance Feb 24, 2022	press release
2022	K223042 (Viz LVO algorithm update)	510(k)	FDA CDRH PDF
Mar 21, 2023	Viz AAA (abdominal aortic aneurysm) — first FDA-cleared AI for AAA	510(k) clearance	press release
Feb 8, 2024	Viz ICH Plus (ICH quantification w/ volume, lateral ventricles, midline shift)	510(k) clearance	press release
2024 (Sept)	Viz HCM (12-lead ECG → hypertrophic cardiomyopathy)	510(k) cleared	Sacra + marketing materials
June 2025	Viz Subdural Plus (subdural hematoma — automated volume, thickness, midline shift)	510(k) cleared	Sacra
Cumulatively	”more than 50 FDA-cleared AI algorithms” / 12 products stated across neuro / cardio / vascular / PE / oncology (per homepage and Sacra)	Mostly 510(k); original LVO De Novo	Viz.ai marketing

Note on competitive accuracy (verbatim, important for capstone): Per RapidAI press release May 21, 2025 (1,591 code strokes at a comprehensive stroke center): RapidAI detected 93% (109) of MeVOs vs 70% (82) by Viz.ai on CT Perfusion — “33% more MeVOs than Viz.ai.” And per Diagnostic Imaging article on the DUEL study, ISC 2025: Rapid LVO detected 146 LVOs vs 110 by Viz LVO (in 1,525 code strokes); “Rapid LVO also ruled out 94% of LVO-negative cases in comparison to 91% for Viz LVO.” Authors: “The substantial number of LVOs missed by the VIZ software could lead to delay in LVO diagnosis and treatment times” — attributed 23% of missed cases to “failure of the software to process cases.” These are competitor-published figures, but they are peer-reviewed at ISC and constrain Viz.ai’s claimed sensitivity leadership in the MeVO / distal occlusion category.

2.5 Indications-for-Use verbatim guardrails (every cleared module’s disclaimer)

From viz.ai/indications-for-use, every module repeats boilerplate limitations like:

“Viz LVO is limited to analysis of imaging data and should not be used in-lieu of full patient evaluation or relied upon to make or confirm diagnosis.”

“Images that are previewed through the mobile application are compressed and are for informational purposes only and not intended for diagnostic use beyond notification. Notified clinicians are responsible for viewing non-compressed images on a diagnostic viewer and engaging in appropriate patient evaluation and relevant discussion with a treating physician before making care-related decisions or requests.”

This framing — “notification, not diagnosis” — is the FDA-friendly posture that lets Viz.ai clear every module as Class II SaMD under the radiology-triage category. It also defines the clinical specialty boundary.

3. Pricing & go-to-market

3.1 Pricing model (no public rates)

Direct pricing: None. No public price list. No /pricing page. Every CTA is “Request a demo” or “Speak to an expert.”

From UsagePricing Viz.ai blueprint:

“Viz.ai is fully sales-led: there is no published price list, no self-serve signup, and no /pricing page — every button on the site says ‘Request a demo’ or ‘Speak to an expert.’ Pricing is a quoted per-facility (per-hospital) annual subscription, sold modularly by disease suite (Viz Neuro, Cardio, Vascular, Pulmonary and more). Total contract value scales with how many of the 50+ FDA-cleared algorithms a hospital licenses and the size of the population it serves.”

Third-party indicative estimate (NOT OFFICIAL): ~$25k/year per hospital for the stroke module. From Out-of-Pocket analysis:

“They’re charging a subscription, and while they reference $25K per year [as an example] in the CMS documents, that’s not what they’re actually charging — Viz.ai charges a subscription to use their model. The cost is not what was included as ‘an example’ in the CMS documents.”

(NOTE: This $25k figure originates from an Out-of-Pocket 2020 example — not Viz.ai’s actual price. Treat as illustrative / snippet only.)

The ROI counterweight — Medicare NTAP (up to $1,040 per eligible use):

From JNIS NTAP commentary, Hassan 2021:

“The Centers for Medicare and Medicaid Services (CMS) recently granted a New Technology Add-on Payment (NTAP) for Viz ContaCT (Viz LVO)… This is the first time CMS has reimbursed an artificial intelligence (AI)-based software using this designation. It applies to Viz.ai’s acute ischemic stroke product, Viz LVO… under which the ICD-10 Procedure Coding System (ICD-10-PCS) procedure code 4A03×5D was established.”

“For the Viz.ai NTAP code, the additional payment is capped at $1,040. To qualify, a patient must be a Medicare patient with a suspected stroke, and the estimated cost must exceed the Medicare reimbursement.”

“The median loss in net monetary benefit of thrombectomy per minute was calculated to be $1, 059 * *, an d * * s a v in g 10 min o na v er a g e a cr oss t h e U S A w o u l d s a v e$ 249 million annually. Implementation of Viz LVO has been demonstrated to save 66 min on average, suggesting a significant return on investment for CMS.”

Net effect (per UsagePricing): “For a hospital seeing enough Medicare stroke patients, that per-use payment effectively offsets the annual subscription, turning a software line item into something closer to cost-neutral.” — with the same contract netting out very differently for primary vs comprehensive stroke centers (primary stroke centers rely on NTAP to break even; comprehensive stroke centers profit from increased thrombectomy volume but see less NTAP).

3.2 Sales motion

Sales-led enterprise B2B. Top-down: hospital C-suite (CIO, CMIO, CFO), stroke program director, ED chair, neurointerventionalist service line. (“Equipped with dedicated 24/7 on-call clinical specialists, implementation experts, and customer success team” — homepage.)
Land-and-expand: Pilot at high-acuity use case (stroke), then cross-sell Cardio/Vascular/Trauma/PE/Oncology suites. Per Sacra: “Viz.ai’s land-and-expand strategy involves initially targeting high-impact areas like stroke care, then leveraging its proven value to secure broader adoption across the enterprise.” Plus 6 new life-sciences partnerships in past 18 months — pharma distribution.
KOL-anchored clinical selling: Heavily cited testimonials from neurosurgery, neurointerventional, vascular surgery, and ED chairs (Don Frei, Raul Nogueira, Adam Arthur, Jayme Strauss named as executive-director of neuroscience at Piedmont). Quote from Dr. Adam Arthur at the NTAP announcement explicitly endorsed the reimbursement model (“This software should be considered now the gold standard for all systems treating stroke”).
Regulatory as marketing wedge: NTAP and “first to clear” FDA milestones are repeatedly headline-anchored in press releases.

3.3 Onboarding time

Per Viz.ai’s own Indiana State Vendor Intake Form (https://www.in.gov/grow-rural-health/files/RHTP-Vendor-Form-Intake-Responses.xlsx, public spreadsheet): “Viz.ai provides onboarding support, live training sessions, and ongoing customer success resources and 24/7 support. 6-8 weeks.” — also per Ekipa: “Typical implementation spans 4–8 weeks with dedicated onboarding.”
Per Us2.ai integration note: Us2.ai’s integration with the Viz.ai marketplace took 6-9 weeks at NorthShore at Endeavor Health.
Customer testimonial quote (Heidi Woessner, Neuroscience Piedmont, FeaturedCustomers) — Viz.ai’s CS team “helped us expedite our implementation.”

3.4 Switching costs — very high (this is the moat)

PACS integration (Siemens, GE, Philips, Canon, “any scanner in a hospital network” — Viz AAA press release). Each integration is per-scanner-vendor and per-protocol.
EHR integration with Epic (most US hospitals), Cerner/Oracle Health, Meditech. Bidirectional hooks into the orders/Imaging modality worklist.
On-call scheduling integration — mobile push notifications routed through Viz, not through the EHR.
Culture/workflow dependency — clinical staff must adopt the parallel workflow. Heidi Woessner references Viz’s CS team staying engaged after go-live. Per Sacra: “This level of trust reflects a highly loyal and expanding install base, as health systems continue to adopt additional functionality over time.” 90% click-through rate on clinical alerts is the behavioral lock-in.
Reimbursement / coding dependencies — ICD-10-PCS procedure code 4A03×5D for Viz ContaCT is established in CMS billing; ripping out Viz means losing the NTAP billing workflow built into the hospital’s revenue cycle.
Data and publications — Viz’s “more than 120 peer-reviewed publications and abstracts” are built on top of Viz-enabled sites; sites are essentially clinical-research infrastructure.

3.5 Recent press (last 12 months) — verbatim headlines

May 14, 2026: Viz.ai Launches Viz Pulmonary Suite — “the First Comprehensive AI-Powered Solution Dedicated to Pulmonary Care Delivery”
Mar 5, 2026: Viz.ai Study Demonstrates 44% Reduction in Interfacility Stroke Transfer Times, presented ISC 2026
Jan 12, 2026: Viz.ai Closes 2025 with Record Scale and Patient Impact — “achieving profitability in its healthcare business… adopted in nearly 2,000 hospitals… life sciences business doubles… 13 partnerships.”
2025: Viz.ai + Microsoft partnership for Cloud for Healthcare — Viz.ai runs on Azure / integrates with Microsoft Precision Imaging Network (Viz.ai & Microsoft)
2025: Viz.ai + Salesforce partnership (announced 2025)
2025: Viz.ai + NCCN partnership — embedded oncology guidelines
2025: Black Book Research #1 Healthcare AI Platform ranking
2025: Third consecutive Edison Award (Machine Learning Innovation)
Oct 2025: Viz Assist launched (ambient + imaging AI agent)
May 2025: Viz Oncology Suite launched
May 2025: Rival RapidAI publishes clinical data showing “RapidAI detects 33% more MeVOs than Viz.ai” — peer-reviewed at ESOC 2025 (RapidAI press release)
Feb 2025: Rival RapidAI publishes clinical data showing Rapid LVO detects 146 LVOs vs 110 by Viz LVO in 1,525 code strokes — ISC 2025 (Diagnostic Imaging)
Feb 2024: Viz ICH Plus FDA 510(k) clearance
Mar 2023: Viz AAA — first FDA-cleared AI for AAA

3.6 Competitive moat summary

From Sacra and Viz.ai 2025 close:

“Viz.ai… established as a leader in the application of AI to improve patient outcomes and streamline care delivery for time-critical conditions.”

“Medicare created a new reimbursement pathway for Viz.ai’s software, establishing it as a new therapeutic class.”

“Viz.ai was the first company to be awarded CMS reimbursement for AI and is ranked the #1 Healthcare AI Platform by hospitals and health systems in the Black Book Research survey.”

4. Gap analysis (8 patient pain points, same rubric as Abridge teardown)

This is the heart of the capstone analysis. Viz.ai is fundamentally a clinician-side workflow tool, not a patient-side tool — it does not see the patient, only imaging + EHR data on the patient. We score each of the 8 patient pain points on a 0–3 scale: 0 = no coverage; 1 = indirect/incidental; 2 = partial; 3 = direct coverage.

#	Pain point (patient-side)	Coverage	How Viz.ai addresses it (or doesn’t)	What’s missing
1	Doctors attribute symptoms to anxiety / psych	0 — none	Viz.ai sees only CT/CTP/CTA/ECG and never sees a free-text chief complaint or history. It does not push back on a physician’s attribution.	No clinical-reasoning layer; no patient-history integration; no “second opinion” framing for the patient. Clinical reasoning and dismissal happen before the CT is ordered.
2	13-year diagnostic journeys (rare-disease, autoimmune, functional)	0 — none	Viz.ai only fires on imaging studies already ordered in acute-care context. It is not an outpatient workup tool. It cannot help a patient with a 13-year journey to a connective-tissue disease diagnosis.	No longitudinal outpatient reasoning; no multi-specialty data aggregation across years; no “what else could this be” capability.
3	10-minute appointments (outpatient, primary care)	0 — none	Viz.ai is exclusively ED / inpatient / IR-suite. No outpatient footprint, no primary-care workflow integration.	Viz Assist (Oct 2025 ambient) is the closest outreach but is still inpatient/ED and aimed at clinician documentation, not patient conversations.
4	Dismissive comments on ambiguous test results	0 — none (and arguably worsens the problem)	Mobile previews are explicitly “for informational purposes only and not intended for diagnostic use beyond notification” — never shown to patients. Imaging outputs (LVO flag, ICH volume) are clinician-to-clinician handoffs and do not get translated to patient-friendly language.	Viz.ai never translates findings to “what this means for you.” Pre-prints compressed images that only clinicians should see.
5	Patients punished for self-advocacy	0 — none	The patient is not in the loop. Viz.ai’s notification flow is strictly clinician → clinician. Patients (or their caregivers) cannot view the alert or send their own DICOM.	No patient portal; no patient-controlled data; no patient-facing second-opinion capability; no caregiver view.
6	”Between-visit” blind spot (wearables, home monitoring, patient-reported events)	0 — none	All inputs come from hospital imaging devices and EHR. No wearables, no smartphone data, no patient-reported outcomes, no remote monitoring.	No data ingestion from Apple Watch, continuous glucose monitors, etc. No patient-reported outcome capture. Viz’s data layer is hospital-bound.
7	Caregivers are the real information source (Alzheimer’s, pediatric, elderly)	0 — none	Caregivers cannot authenticate as a user on Viz.ai’s clinician-facing mobile app. No caregiver role, no family-portal view, no “send to my daughter’s phone.”	No caregiver workflow. The communication graph is fully professional.
8	Medical-education gap / “should I be referred to a subspecialist?“	1 — incidental	Viz does flag incidental findings (AAA → vascular surgery; ICH → neurosurgery; HCM on ECG → cardiology; subarachnoid hemorrhage risk from aneurysm → neuro IR; 71% of lung nodules missed → Viz PE/Pulmonary Suite). So it triggers downstream subspecialty consults. But this trigger is downstream of a positive imaging finding in a hospitalized patient — not pre-referral decision support in an outpatient.	No patient-facing symptom checker; no “you might benefit from seeing X specialist” tool; no clinical-guideline surfacing for the patient; Viz Oncology + NCCN guidelines is the closest, but still clinician-facing and inpatient.

Summary score: Total across 8 pain points = 1 / 24 (≈4%). Viz.ai has effectively zero coverage of the patient-experience pain points that drive our capstone thesis.

4.1 Coverage heatmap

Strong vertical (where Viz.ai dominates): acute-care workflow — faster stroke triage, hub-and-spoke transfer, CMS reimbursement pathway, FDA clearances. This is the gold-standard acute-AI playbook.
Empty quadrant (where the wedge lives): the outpatient, longitudinal, between-visit, patient-facing, caregiver-included, “what is this ambiguous finding” space. None of the 8 pain points are touched.

4.2 The opportunity — articulated crisply

Pain point 2 (“13-year diagnostic journeys”) — impossible to address inside Viz.ai because Viz sees only imaging, not longitudinal outpatient records. A tool that aggregates + summarizes 5-10 years of fragmented outpatient records and proposes “have you considered X specialty?” has zero overlap with Viz and zero Viz moat to compete against.
Pain point 7 (“caregivers are the real information source”) — Viz.ai’s clinician-only architecture actively excludes caregivers; building a caregiver-as-user model would need a fundamentally different product (and FDA pathway). Viz can’t pivot here without cannibalizing its clinician channel.
Pain point 4 (“dismissive comments on ambiguous test results”) — Viz.ai’s mobile previews are explicitly forbidden from patient use (FDA labeling). A patient-facing “what does this scan mean in plain English?” tool reads off Viz-produced structured outputs but is a different product (and a different regulatory posture — patient decision support vs. clinician triage).
Pain point 1 (“doctors attribute to anxiety/psych”) — Viz.ai fires on imaging; by the time Viz runs, the physician has already committed to organic workup. Changing the attribution moment requires a tool that lives at the intake (before imaging), not downstream.

5. Summary

Moat

Regulatory + reimbursement moat, not just technical. Viz.ai is the only AI imaging company with a CMS NTAP (up to $1,040/Medicare use for Viz LVO). This creates an installed-base moat through ICD-10-PCS procedure code 4A03×5D integrated into hospital revenue-cycle workflows, and a competitive moat because subsequent AI stroke products have to either get their own NTAP or compete as inferior non-reimbursed tools.
Scale + workflow dependency: ~2,000 hospitals, 230M lives, 50+ FDA-cleared algorithms, 120+ peer-reviewed publications, 90% clinician click-through rate on alerts, and clinicians have built their on-call workflows around the parallel-alert pattern. This is essentially a two-sided network between radiologists/neurointerventionalists and ED physicians, and it’s a deeply sticky enterprise SaaS.
First-mover in the FDA SaMD “notification-only, not diagnostic” category: every imaging-AI competitor enters through the same FDA pathway Viz.ai established, but Viz has the head-start of 8 years of accumulated clinical-evidence publications and hospital integrations.

Weakness

Accuracy gap in MeVO / distal occlusion vs RapidAI. Per RapidAI’s ESOC 2025 study: RapidAI detected 93% of MeVOs vs 70% by Viz.ai. Per Diagnostic Imaging DUEL study, ISC 2025: Rapid LVO detected 146 LVOs vs 110 by Viz LVO in 1,525 code strokes, with Viz missing 26% of LVOs (23% attributed to “failure of the software to process cases”). This is the most damaging independent peer-reviewed finding against Viz in recent years and is being amplified by competitors.
Workflow tool, not diagnostic tool. Every Viz.ai module’s FDA labeling expressly forbids using it “in-lieu of full patient evaluation or relied upon to make or confirm diagnosis.” Critics argue this makes it worklist prioritization with a marketing wrapper.
Acute-only. Zero outpatient, zero longitudinal, zero patient-facing, zero caregiver. Vulnerable to a parallel product category entirely outside its FDA architecture.
Customer-count growth flattening as it approaches large-account saturation (1,400 in 2022 → 1,500 in Feb 2024 → 1,700 on homepage → 1,800 on /vizai-one → ~2,000 in Jan 2026). Implied YoY growth ~25–33% but slowing as the long tail of community hospitals signs on.
Pricing opaque. Sales-led creates friction in B2B procurement; competitors with cleaner total-cost-of-ownership claims (e.g., RapidAI’s per-scan bundle pricing in some quotes) have a sales-cycle wedge.

Our opportunity (capstone wedge)

The Viz.ai teardown reveals a $0-competition quadrant in the AI healthcare landscape:

Outpatient, longitudinal, patient/caregiver-in-the-loop tools that operate before and after acute imaging — addressing pain points 1, 2, 3, 4, 5, 6, 7, 8 of our capstone rubric.
Patient-facing translation of AI-generated structured outputs (e.g., “your MRI showed a 3mm lesion on the [X]; here are 3 questions to ask your neurologist in plain English”) — directly leveraging Viz.ai-style imaging outputs but presented in patient-voice.
Specialty-guideline-as-patient-coach (“Based on your MRI findings + symptoms, the American Academy of Neurology referral criteria suggest a movement-disorder specialist visit within 6 weeks. Here are 3 in your insurance network. Here’s what to bring.”) — operationalizes pain point 8 in a way no imaging AI vendor does.
Caregiver-as-licensed-user (POA / HIPAA-authorized-rep workflow) — Viz.ai’s clinician-only architecture creates a hole big enough to drive a separate product through.

Bottom line: Viz.ai is the unbeatable acute stroke AI category leader. The capstone wedge is the everything else — the 95% of neurology patients who are NOT having a thrombectomy-eligible LVO, and the 100% of those patients plus their caregivers who need a tool that lives outside the ED.

Sources

Primary (Viz.ai)

Customer / hospital / press

Regulatory

Clinical / academic

Competitive / independent

End of Viz.ai teardown. Total verbatim customer/clinician quotes captured: 11. Total FDA clearances documented: 10+. Total real-world clinical studies cited: 7. Coverage of 8 patient pain points: 1 / 24.

Quartz 5

Explorer

Competitor Teardown - Viz.ai