PulsePad — iPad anesthesia cockpit that OCRs any vet monitor and ends paper charts

1. One-liner

iPad cockpit that reads any anesthesia monitor with the camera and ends handwritten charts for independent US vet clinics.

2. Trend signal — why now?

Three things broke loose in the last 12 months and they all point at the same gap:

• ACVAA published new monitoring guidelines in July 2025 — the first major revision since 2009. The update expands what must be documented every surgery: hemodynamic, respiratory, thermoregulatory, neuromuscular blockade, anesthetic depth, plus a new pre-anesthesia equipment checklist. More boxes to fill on the same paper sheet a vet tech is filling between checking pulses on a sleeping Labrador.
• State boards are citing the combo, not just one or the other. The California Veterinary Medical Board filed accusations against multiple licensees in late 2025 and early 2026 — Naiditch (Dec 2025), Byerly (Dec 2025), Van Volkenburgh (Feb 2026), Ross — each pairing “Anesthesia Violation” with “Recordkeeping Violation” in the same complaint. The legal exposure is real and getting named.
• The PIMS giants moved, but they bundled. Instinct rolled out “Ax Mode” inside Instinct EMR. ezyVet has Vet Radar. Midmark sells an Anesthetic Record Interface tied to its own monitors. Digitail raised a $23M Series B in 2025. Vetcove raised $30M. Money is flowing — but every solution requires either swapping out the PIMS or buying a specific monitor brand. The 50%+ of US small-animal clinics still on AVImark/Cornerstone (Covetrus + IDEXX legacy server-based) have no clean path.

Layer in a vet tech market that’s 15% short on credentialed staff (NAVTA), with 14,300 openings and 7,500 grads sitting the licensing exam each year. Nobody has spare hands to scribble vitals every 5 minutes.

Provenance:

• Signal 1 (Demand/Regulatory): ACVAA Small Animal Anesthesia and Sedation Monitoring Guidelines 2025 — published July 2025, first revision since 2009 — https://acvaa.org/updated-acvaa-small-animal-anesthesia-monitoring-guidelines-are-available/ — observed 2026-05-09
• Signal 2 (Demand/Enforcement): California VMB 2025–2026 disciplinary actions pairing “Anesthesia Violation” with “Recordkeeping Violation” (Naiditch, Byerly, Van Volkenburgh, Ross) — https://www.vmb.ca.gov/consumers/2026_disciplinary.shtml — observed 2026-05-09
• Signal 3 (Economic/Feasibility): Digitail closes $23M Series B led by Five Elms Capital; Vetcove closes $30M; Instinct ships “Ax Mode” inside Instinct EMR — https://digitail.com/blog/digitail-raises-23m-series-b-led-by-five-elms-capital/ + https://instinct.vet/blog/introducing-instincts-digital-anesthesia-mode-for-veterinary-hospitals/ — observed 2026-05-09 Category: Workflow automation

3. The opportunity

Every existing digital anesthesia record is bundled into something the customer doesn’t want to swap. Instinct Ax Mode = upgrade your PIMS. Vet Radar = move to ezyVet. Midmark Advisor = buy Midmark’s $4–8K monitor. The independent two-vet practice on AVImark since 2008, running a SurgiVet or Cardell monitor, gets nothing useful — they keep using the paper sheet and roll the dice on the next AAHA audit or board complaint.

PulsePad attacks the gap directly: a standalone iPad app that reads any anesthesia monitor by pointing the iPad’s rear camera at the screen, transcribes voice notes from the vet tech, generates an ACVAA-formatted PDF chart at the end of the case, and emails it / drops it into the existing PIMS as an attachment. No PIMS migration. No new monitor. No 30-page implementation. $149/mo per clinic, swipe a card, used the same afternoon.

The 10× claim isn’t “AI-powered” — it’s “you don’t have to rip and replace anything.” The clinic stays on AVImark. The monitor stays on the cart. The paper sheet goes in the bin.

4. Target market

• Primary customer: Independent US small-animal veterinary clinics, 1–4 DVMs, AAHA-accredited or pursuing accreditation, running a legacy PIMS (AVImark, Cornerstone, IntraVet, ImproMed) and an off-the-shelf anesthesia monitor (SurgiVet Advisor, Cardell, Mindray, BioLight, Midmark, Edan). Owner-operator economics, $1.5M–$6M annual revenue.
• Why they buy: AAHA surprise audits. State board complaints citing “anesthesia + recordkeeping.” A vet tech who quit because she’s tired of writing in 5-min increments while watching a capnography trace. ACVAA 2025 added more boxes to fill. The malpractice insurer asking, in a renewal questionnaire, whether anesthesia records are typed.
• Rough TAM reasoning: ~32,000 small-animal clinics in the US (AVMA). ~3,000 AAHA-accredited; another ~4,500 affiliated. Independent (non-corporate-rolled-up) is roughly 60% — call it 18,000–20,000 clinics. Realistic serviceable target in years 1–3: the 5,000–7,000 AAHA-touching independents who already self-identify as records-conscious.
• Why now for them: The ACVAA guideline update in July 2025 forced anesthesia CE refreshers across the industry. Tech vacancy is at 15%. The cost of replacing a vet tech who walked is $20K. A $149/mo tool that gives a tech back 30–45 minutes per surgery day pays for itself in the first week.

5. Product sketch (MVP)

• Camera-OCR vital capture. Prop the iPad up next to the anesthesia monitor. Camera reads HR, RR, SpO2, EtCO2, NIBP, MAP, Temp, agent % off the screen. Logs every 30–60 seconds.
• Voice-first event log. Tech taps once and speaks: “0.3 mL ketamine IV, 14:32.” Whisper transcribes, app puts it in the drug column with timestamp. No typing while gloved.
• ACVAA 2025 checklist mode. Pre-anesthesia equipment checklist enforced before “start case.” Personnel name + role required.
• Auto-generated PDF chart. Case ends → PDF with timeline graph, drug log, monitoring narrative, and the AAHA / ACVAA-required fields, signed by the assigned tech and DVM.
• PIMS drop. Email the PDF, push to a watched folder, or drag-drop attach to the AVImark/Cornerstone patient record. No API integration project.
• Multi-OR. One iPad per induction table. Cloud roster of cases for the practice manager.
• Audit binder. One click to pull the last 90 / 180 / 365 days of charts as a single PDF for AAHA audits.
• Fallback typing. Tech can correct OCR by tapping the value. Never blocked by an unreadable screen.

6. AI angle — what’s load-bearing

Two AI capabilities are doing the actual work, and both became reliably cheap in the last 12 months:

1. Multimodal vision reading the monitor screen. A GPT-4o-class model reads HR/RR/SpO2/EtCO2/NIBP/agent % off arbitrary monitor brands and screen layouts. Pre-2024 this was unreliable and per-call expensive. In 2025–2026 it’s pennies per case and accurate enough that human spot-check beats handwriting.
2. Whisper voice transcription tuned for vet drug names. Tech says “1.2 of dexmedetomidine IM” while gloved; transcript classifies dose, route, drug, time. This is the part that kills the paper sheet — gloves + scribbling is the moment the chart turns into chicken scratch.

Strip the AI out and you have… a digital paper form. Nobody wants that. The AI is the reason the workflow collapses from 5-minute manual interrupts to 2-second voice or zero touch.

7. Localization angle (if any)

N/A — this is a US-first play. The US has the unique combination of (a) AAHA accreditation as a paid voluntary standard, (b) state veterinary medical boards with active enforcement, (c) AVImark + Cornerstone legacy entrenchment, and (d) a vet tech credentialing model that creates the documentation burden. UK and Canada are obvious year-2 expansions; AVA / RCVS guidelines parallel ACVAA. Skip them until the US wedge is proven.

8. Business model — path to $1M–$5M ARR

• Pricing: $149/mo per clinic, single OR. $99/mo per additional OR/induction table. Annual prepay 2 months free.
• ACV: Realistic blended $1,800–$2,400/yr per clinic.
• Rough math to $1M ARR: 500 clinics × $1,800 = $900K. 500 clinics × $2,000 ACV = $1M. Achievable in 18 months with one founder doing AAHA-list outreach and one part-time SDR.
• Rough math to $5M ARR: ~2,000 clinics at $2,500 ACV. Requires expansion: voice-AI surgical safety checklist, post-op recovery chart, pain-score logging, anesthesia drug ordering integration with Vetcove / Patterson. Each adds $30–60/mo per clinic.
• Expansion path: Multi-OR upsells, multi-location practices (corporate rollups), then add-on modules (recovery monitoring, controlled-drug log integration, AAHA audit-binder export-as-a-service).

9. Go-to-market wedge — first 100 customers

Concrete. The first 100 are surgically targeted:

1. AAHA-accredited member directory scrape, focused on independents. Public list, 3,000 hospitals. Filter out corporate rollups (Mars, NVA, Thrive, BluePearl, VCA) using the chain-name dictionary. ~1,200–1,500 independents. Personalized cold email + Loom: “Saw your AAHA listing, you’re due for re-accreditation in 14 months — here’s a 2-minute video of how PulsePad makes the anesthesia binder a non-event.” Expect 4–6% reply, 1.5–2% close = 25–30 customers.
2. VTS (Anesthesia/Analgesia) credentialed techs as champions. AVTAA membership is small (~250 active VTS-Anesthesia). DM the public list on LinkedIn + AVTAA member directory. Free Pro for any VTS-Anesthesia who pilots in their clinic and writes a 200-word review. Each champion drags 2–4 clinics behind them. = 20–40 customers.
3. VETgirl + dvm360 sponsored content. Both publish CE on the 2025 ACVAA guidelines. Sponsor a single piece tying “the new ACVAA guidelines” to “your anesthesia chart” with a 30-day pilot CTA. VETgirl’s email list is ~70K. Realistic 0.05% direct conversion = 35 trials, 30% close = 10 customers.
4. VMX 2026 + AVMA Convention 2026 + VHMA 2026 booth crawl. Skip the booth, walk the floor with an iPad demo. Anesthesia track sessions are where the buyers are. 200–300 in-person demos across the three cons = 30–40 trials, 15–20 closes.
5. State board complaint / disciplinary action subscribers. Subscribe to CA, TX, FL, NY VMB enforcement RSS. When a clinic gets cited for “recordkeeping,” send a templated note to the practice owner: “We can’t fix the past, but here’s how the next chart writes itself.” Sympathetic close — small numbers, but high signal. 5–10 customers.

Total path: 90–120 paying clinics in 9 months. No paid ads, no SEO play, no enterprise quota. ~$60–80K spend.

10. Build complexity — justification

Medium. Three off-the-shelf APIs (OpenAI multimodal vision, Whisper, a PDF lib), one iPad app, one thin web admin for practice managers. The non-trivial piece is the monitor library: each major model (SurgiVet Advisor, Cardell 9405, Mindray uMEC, BioLight, Edan iM, Midmark Cardell-9402) has a different screen layout. Build a per-model template / few-shot example library starting with the top 5 covering ~70% of US clinics. 3–4 months for v1 with one full-stack engineer + a vet tech advisor on retainer. v1 ships covering 5 monitor models + AAHA PDF; long tail extended monitor-by-monitor as customers report theirs.

11. Gating checklist

Gate	Pass?	Note
Legal in target market	✅	Records tool, not a medical device. No FDA pathway needed because nothing is alarming, dosing, or controlling the patient. PulsePad observes; it does not treat.
Ethical — no harm / dark patterns	✅	Augments documentation; tech still actively monitors patient. Fallback to manual entry always available.
Market exists (evidence above)	✅	ACVAA guidelines + state board citations + funded competitors all confirm.
1–5 person team can build this	✅	Solo engineer + part-time vet tech advisor for 3–4 months.
Launchable with <$50K / ₹40L	✅	iPad-based, off-the-shelf APIs, no hardware to ship. ~$25–35K to MVP and 5 pilot clinics.

All five pass.

12. Feasibility score

Axis	Weight	Score	Notes
Problem intensity	20	17/20	Felt every surgery day. State boards naming the violation. AAHA audit risk. ACVAA 2025 added documentation burden. Not on-fire daily but acute when it bites.
Demand evidence	15	13/15	Multiple independent signals: ACVAA July 2025 publication; 4+ named CA disciplinary actions 2025–2026; Digitail $23M, Vetcove $30M; Instinct + ezyVet + Midmark + VetMo all already shipping competing-but-bundled products.
Build feasibility	15	11/15	Solid: vision + Whisper + PDF + iPad. Friction: per-monitor screen-template library. 3–4 months v1, but the monitor long-tail will be ongoing.
Distribution clarity	15	12/15	AAHA list + AVTAA list + VMX/VHMA shows + state-board enforcement RSS = a named channel with conversion math. Niche is small enough to walk every door.
Revenue mechanics	15	12/15	$149/mo benchmark sane against $300–1,000/mo PIMS spend. 500 clinics → $1M ARR is grindable, not aspirational. Multi-OR upsell is real.
Time to first revenue	10	8/10	Pilot in 6 weeks, paying in 8–10 weeks. Pre-sale to 2–3 friendly clinics is realistic before code is done.
Defensibility	10	6/10	Soft moat: monitor-OCR template library compounds with every clinic; AAHA-format PDF templates; workflow lock-in once techs trained. PIMS giants could clone but won’t because they prefer to bundle into their own PIMS upgrade. Execution-led 9-month head start is the real moat.
Total	100	79/100

13. Qualitative modifiers

Founder-fit tags

technical-heavy · domain-expertise-required — Need a builder comfortable with multimodal vision prompting + iPad/iOS, plus a vet tech (ideally VTS-Anesthesia) on payroll or co-founder-equity from week 1. Without the domain advisor, the chart format is wrong, the monitor library starts in the wrong order, and the pitch lands flat.

Key assumptions to validate (3–5)

1. Assumption: Multimodal vision can read 5 mainstream monitor brands with ≥97% per-field accuracy on every-30s captures under realistic OR lighting. How to test: Borrow or buy used SurgiVet Advisor + Cardell + Mindray uMEC. Run 50 simulated cases each, log accuracy by field. Kill if median accuracy <95% even after few-shot calibration per model.
2. Assumption: Independent AAHA clinics on AVImark/Cornerstone will pay $149/mo without an integration. How to test: 30 cold pricing-included pitches to AAHA-list practice owners. ≥5 verbal commits to a paid 30-day pilot before any code shipped. Kill if <3.
3. Assumption: Vet techs will adopt the camera + voice workflow over paper after one full surgery day, not abandon back to clipboard. How to test: 4 in-clinic shadow sessions with a vet tech using a Wizard-of-Oz prototype (researcher does OCR by hand). Net Promoter ≥8 from the tech, plus willingness to use again unprompted. Kill if techs say “honestly the paper is faster.”
4. Assumption: The ACVAA-format PDF satisfies the AAHA accreditation evaluator. How to test: Pay an ex-AAHA evaluator (consulting market exists) for a 2-hour audit of the generated chart format. Kill if material gaps require rebuild.

Risk flags

1. Platform dependency: OpenAI / Anthropic vision API pricing or reliability shifts. Mitigation: build for multi-provider; benchmark Gemini 2.5 + Claude Opus 4.7 + GPT-4o quarterly.
2. Regulatory drift: If FDA or USDA reclassifies any animal-monitoring software as a “device,” the legal posture changes. Mitigation: stay strictly observational, never alarm or interpret, no “the patient is light” predictions. Keep the boundary visible.
3. PIMS-giant clone: Instinct or Digitail extends their anesthesia mode to support arbitrary monitors. Mitigation: lean into PIMS-agnostic positioning + AAHA audit-binder feature; deepen monitor library.
4. Tech adoption inertia: Vet techs are famously skeptical of “just one more tablet.” Mitigation: make the camera literally the only required interaction the first hour; voice second; no setup wizards.

14. Structured verdict

Score:                  79/100
Verdict:                GO
Confidence:             Medium
Best-fit builder:       Solo technical founder (iOS + multimodal-vision experience) plus a credentialed vet tech (VTS-Anesthesia preferred) as co-founder or weekly advisor
Time to revenue:        8–10 weeks from build start; 4–6 weeks if pre-sold
Capital to launch:      $25–35K to MVP + 5 pilot clinics
Top 3 assumptions to validate first:
  1. Vision-OCR ≥95% per-field accuracy on top-5 monitor brands — bench test against borrowed/used hardware
  2. ≥5 of 30 AAHA-list cold-pitched practice owners commit to a paid 30-day pilot
  3. Vet techs prefer the workflow over paper after one full day — Wizard-of-Oz prototype in 4 clinics
Kill criteria:
  - Abandon if monitor-OCR accuracy <95% after 2 months of prompt + few-shot tuning
  - Abandon if <3 of 30 cold AAHA pitches verbally commit to a paid pilot
  - Abandon if Instinct or Digitail ships PIMS-agnostic standalone anesthesia mode at <$199/mo before MVP launch

15. Next step — 1-week validation sprint

If I had one week to prove this is real before building:

• Day 1: Pull AAHA-accredited member list. Filter to ~600 likely independents. Draft and send 30 highly personalized cold emails to practice owners, each citing their specific clinic and a known anesthesia-related dvm360/AAHA topic. Target ≥3 reply meetings booked by end of day 5.
• Day 2: Buy or borrow access to a SurgiVet Advisor + a Cardell. Set up an iPad with Claude / GPT-4o vision and run 100 captures of varying lighting, distance, and screen content. Score per-field accuracy.
• Day 3: Wizard-of-Oz prototype — Figma + an iPad propped up. Visit 1 friendly clinic (an existing connection) and shadow 2 surgeries doing the OCR by hand. Watch the tech’s reaction.
• Day 4: Schedule and run 4–6 of the booked AAHA-clinic Zoom calls. Pitch with pricing. Ask for verbal pilot commits.
• Day 5: Tally:
  • GO if (a) vision accuracy ≥95% on 2 monitors + (b) ≥3 verbal pilot commits + (c) the in-clinic tech says “I’d use this tomorrow.”
  • NO-GO if any of those three fails. The cluster doesn’t survive without all three.

The validation is falsifiable on three concrete numbers. No vibes.