Ultrasound · Handheld whole-body point-of-care ultrasound with AI measurement tools
Butterfly iQ+ with AI
Butterfly Network
The device that made ultrasound something a clinician carries rather than books time on — one silicon probe, thirteen cleared applications, roughly a hundred countries — and the catalogue's clearest cautionary tale on AI governance: in 2020 its two flagship AI tools were recalled for having been shipped without any 510(k) clearance at all, then brought back through the front door.
Performance Metrics
Clinical Evidence
The evidence base is broad the way the device is broad — and correspondingly thin at any single point. The strongest independent clinical evaluation is a University of Iowa emergency-department study (Diagnostics, December 2023): across 381 handheld point-of-care exams on the iQ+, clinicians identified 86.5% of prespecified emergent diagnoses (32 of 37, including acute cholecystitis, severely reduced ejection fraction and pericardial effusion) and 84.3% of less-emergent diagnoses against radiology-performed imaging as reference. The manufacturer-authored PNAS platform paper (2021) documents the ultrasound-on-chip engineering, the augmented-reality acquisition guidance and automated ejection-fraction estimation, and the 13-indication clearance breadth. Feature-level validation is more mixed, and honestly so. A prospective emergency-department comparison of bladder-volume measurement (Cureus, 2024; 21 patients in acute urinary retention, catheterisation as gold standard) found the cart-based ultrasound agreed with catheterised volume better than either the Butterfly auto-bladder tool or the nursing bladder scanner — convenience favoured the handhelds, accuracy the cart. The HOUSE-HF feasibility study (ESC Heart Failure, October 2025, University of Pennsylvania) showed patients recently hospitalised for decompensated heart failure could be taught a six-zone lung protocol on the iQ+ in about twenty minutes and sustain self-scanning at home with physician cloud review — a care-model result rather than an accuracy result. No peer-reviewed accuracy study of the 2026 Gestational Age tool was located; its equivalence claim to sonographer-performed biometry (training set reported at over 21 million images) is vendor-reported, resting on the 510(k) clearance rather than published literature.
| Study | Design | n | Sensitivity | Specificity | AUC | Published |
|---|---|---|---|---|---|---|
| Wubben BM, et al. (University of Iowa — independent; single-centre ED cohort) | RetrospectiveRetrospective | 381 | 86.5% emergent-diagnosis identification (32/37) | — | — | Diagnostics (Basel), Dec 2023; 381 handheld POCUS exams on iQ+ in one emergency department, Nov 2021–Nov 2022; 84.3% (43/51) of less-emergent diagnoses also identified vs radiology reference |
| Ho-Gotshall S, Wilson C, et al. (Grand Strand Medical Center — independent; prospective three-way comparison) | ProspectiveProspective | 21 | — | — | — | Cureus, 2024; bladder volume by Butterfly auto-bladder tool vs nursing bladder scanner vs cart-based ultrasound against catheterisation; cart-based agreed best with catheterised volume, handhelds rated more convenient |
Clinical Pulse
Verified clinician sentiment
Verified clinician reviews launching soon · Apply to contribute →
Reviews are signed by clinicians verified against their regulator's public register.
Inside the Auris+ Listing
Five more sections complete this device’s Auris+ Listing.
Decision Ledger
ProPro unlocks a private, cross-vendor log of every case you read on this device — what the AI called, what you concluded, and a one-line reason if you overrode it. Ready for the EU AI Act's deployer logging obligations when they land in 2028.
Clinical Evidence Deep Dive
ProPro unlocks the structured clinical-evidence summary — study count, target patient population, and a tabular accuracy-metrics view drawn from peer-reviewed sources.
Peer-Reviewed Publications
ProPro unlocks the curated peer-reviewed publication list with PubMed cross-links — the citation backbone of every editorial verdict.
Post-Market & Regulatory Conditions
ProPro unlocks the post-market surveillance summary, recall record, and the conditions of approval that bound real-world use.
AI Algorithm Version History
ProPro unlocks the chronological record of algorithm version changes — what changed when, drawn from manufacturer changelogs and regulatory filings.
Regulatory Approvals
ARTG 318962
K252148
Class II
Safety Record
FDA recall Z-1601-2020 (initiated 2020-02-24): Auto Ejection Fraction Tool on Butterfly iQ software versions 1.10.0–1.15.0, distributed without an approved 510(k); tool disabled via correction notice. Terminated.
FDA recall Z-1600-2020: Auto 3D Bladder Volume Tool on Butterfly iQ, distributed without an approved 510(k); tool disabled, then re-cleared as K200980 in June 2020. Terminated.
RA-75979: Butterfly iQ+ units shipped to Canadian customers by clerical error before the iQ+ held a Canadian medical device licence (only the original iQ was licensed at the time).
RC-2025-RN-00012-1: obstetric-preset app defect on iOS, Butterfly app versions 2.35.0–2.38.0, corrected in 2.38.1; no adverse health consequences reported. Carried from the mined research digest; the DRAC record was not independently re-read this run.
Butterfly's safety record is the catalogue's standing exhibit on what happens when AI features outrun their regulatory clearance. In February 2020 the company disabled two AI tools — Auto Ejection Fraction (software versions 1.10.0 through 1.15.0) and Auto 3D Bladder Volume — after the FDA recorded both as Class II recalls (Z-1601-2020 and Z-1600-2020) with the stated reason "distributed without an approved 510(k)". The bladder tool returned through a proper clearance four months later (K200980, June 2020); a discrete re-clearance for Auto EF could not be isolated in the register. Both recalls are terminated. The pattern recurred administratively in Canada, where iQ+ units were shipped to Canadian customers before the device held a licence (Health Canada recall RA-75979, Type II, posted November 2020, attributed to a clerical error; the iQ+ licence followed in June 2021). A 2025 TGA product correction (RC-2025-RN-00012-1, per the mined research digest — not independently re-read this run) covered an obstetric-preset app bug on iOS in versions 2.35.0–2.38.0, fixed in 2.38.1, with no adverse health consequences reported. The mined MAUDE review found four device reports — 2020 charger/overheating malfunctions and two 2026 probe display failures — none with a patient-harm signal.
Intended Use & Indications
Butterfly iQ+ is a battery-powered, phone- or tablet-connected ultrasound system in which a single silicon transducer emulates the linear, curved and phased arrays that conventionally require three separate probes. Per the FDA 510(k) record for the platform (K202406), it is indicated for use by trained healthcare professionals for diagnostic ultrasound imaging and measurement of anatomical structures and fluids in adult and pediatric patients, spanning peripheral vessel, procedural guidance, small organs, cardiac, abdominal, urology, fetal/obstetric, gynaecological, musculoskeletal and ophthalmic applications. The AI tools sit on top of that platform indication as measurement aids: the clinician acquires the images and retains interpretation; the algorithms segment, count or estimate. The exception in operator model is the 2026 Gestational Age tool, whose blind-sweep protocol is expressly designed for users who are not trained sonographers — fundal height is entered, guided sweeps are performed, and the estimate is produced automatically for patients between 16 and 37 weeks.