Pulsenmore Ltd. (NASDAQ: PLSM) said on August 17, 2026 that the Israel Innovation Authority had approved a grant of $1.27 million for its SmartScan AI research and development programme. It is the second Israel Innovation Authority approval the company has disclosed in a month. The company said the two approvals together come to approximately $1.6 million of grant funding against programmes with a combined scope of approximately $3.12 million, which implies that roughly half the cost sits with Pulsenmore.

The SmartScan AI programme runs for 12 months. Its stated purpose is to combine artificial intelligence with real-time feedback so that a pregnant woman can guide herself through an ultrasound scan at home. The company said the planned system will combine position data from a new transducer and guide the user until a suitable image is obtained, and that the objective is to reduce operator dependency so clinical teams can support a greater volume of remote examinations. Development is expected to draw on data from more than 250,000 clinical scans accumulated on the platform, and to include optimisation of models for on-device processing.

Founder and Chief Executive Elazar Sonnenschein said the second approval “represents another vote of confidence in our technology and the Company’s strategic direction” and that the company is building an AI layer running from image acquisition through the clinical workflow. Pulsenmore named three intended applications: fetal well-being and growth monitoring, visual feedback on fetal head position in advanced pregnancy, and monitoring of follicles and endometrial thickness during fertility treatment.

The first grant, and what it is for

The earlier approval came on July 13, 2026, when Pulsenmore said it had been selected for Israel’s Healthcare AI Regulatory Sandbox Program, established by the Israel Innovation Authority and the Ministry of Health, to lead a $1 million AI development project. That programme addresses a different point in the chain. Today every examination performed through the platform is reviewed in its entirety by a physician before results go back to the patient, and the sandbox project is aimed at software that automatically identifies key ultrasound parameters and presents findings to the physician, shortening review time. Pulsenmore said it will work with Beilinson Hospital, part of the Clalit Health Services network, on development and clinical validation.

Taken together, one grant funds acquisition and the other funds interpretation. The company said its technology is deployed by Clalit Health Services, Israel’s largest health maintenance organisation, and by other clinical partners, and that more than 250,000 home ultrasound scans have been performed on the platform.

What the regulatory record establishes

The device the software would sit on has a defined United States regulatory identity, and it is a recent one. The Food and Drug Administration granted a De Novo classification request for the Pulsenmore ES, submission DEN240074, with a decision on October 31, 2025 on a request received on December 11, 2024. The order created a new device type, an ultrasound imaging system for acquiring images at home by lay users, under 21 CFR 892.1590, Class II, product code SGJ.

The indications are tightly drawn. The Pulsenmore ES is intended to enable acquisition of ultrasound images that allow interpreting healthcare providers to determine fetal heart rate, and for limited diagnostic imaging in B-Mode and M-Mode in fetal and obstetric applications where traditional scanning at a clinic is impractical. It is intended for pregnant women with a singleton pregnancy at a gestational age of 14-38 weeks, on the order of a physician, and is designed for patients with a body mass index under 40. A physician interprets the images in a remote access setup, and access to the device must be granted by healthcare professionals. The FDA order states that the device is not intended to detect defects, replace a routine scan or pregnancy test, or replace a medical consultation or a visit to the emergency room or clinic.

Two of the special controls FDA attached to the new device type map directly onto what SmartScan AI proposes to do. One requires that non-clinical performance testing validate device system guidance functionality in a simulated use environment. Another requires that technological characteristics incorporate design features limiting the number and duration of scans by the lay user, to mitigate unnecessary ultrasound exposure. A guidance algorithm that keeps prompting a user until a suitable image is obtained interacts with both.

What the grants cost

Israel Innovation Authority money is not free capital, and Pulsenmore’s annual report on Form 20-F sets out the terms. As of December 31, 2025 the company had received royalty-bearing grants totalling NIS 15.6 million, approximately $4.9 million. It is committed to pay royalties at a rate of 3% on sales proceeds from products developed under Israel Innovation Authority programmes, up to the total amount of grants received, with interest. For grants approved on or after January 1, 2024, the annual interest is the higher of the 12 month Secured Overnight Financing Rate plus 1%, or a fixed annual rate of 4%.

The restrictions matter more than the royalty. Under the Research Law, transferring know-how or manufacturing rights outside Israel requires approval from an Israel Innovation Authority research committee. If manufacturing rights move abroad with approval, royalties can rise to a total equal to 300% of the grant amount plus interest. If know-how moves abroad, the ceiling is up to six times the total grants received in respect of that know-how, plus interest. The company said it does not anticipate paying at the increased rate.

Analysis: non-dilutive funding with a jurisdictional lock

The grants are small next to the company’s own spending, and the disclosure invites a comparison the release does not make. Research and development expenses were NIS 17.4 million, approximately $5.4 million, in 2025, against NIS 20.1 million in 2024 and NIS 31.4 million in 2023. The combined $1.6 million of new grant money is therefore meaningful relative to a single year of research spending that has been falling, not marginal, but it also funds roughly half of two programmes whose remaining cost the company carries itself.

Revenue is the harder number to read. Sales were NIS 40 million, approximately $12.5 million, in 2025, against NIS 9.7 million in 2024. The 20-F attributes NIS 30.5 million of the 2025 figure to a settlement agreement with GEHC, including amounts recognised on the cancellation of orders for 15,000 units and on the cancellation of remaining distribution rights. Stripping that out leaves an underlying product revenue base close to the prior year. Against that base, a 3% royalty on products developed under grant programmes is a modest claim, but the aggregate NIS 15.6 million of grants already taken is a real accrued obligation.

The binding constraint is jurisdictional rather than financial. Each grant deepens the Research Law attachment on the resulting know-how, and the SmartScan AI work is by description core intellectual property: guidance algorithms, a new transducer’s position data and models trained on the company’s own scan archive. A company whose device is regulated in the United States under a brand new FDA classification, and whose largest disclosed deployment is with an Israeli health maintenance organisation, is developing that software under a programme whose terms place any future transfer of the resulting know-how abroad behind an Israel Innovation Authority approval requirement.

What a careful reader would look for next is narrow. Neither announcement gives a milestone schedule, a disbursement profile or a clinical validation endpoint, and neither grant agreement has been filed. The De Novo order’s special controls set out what clinical and non-clinical testing an acquisition-guidance feature would have to satisfy, so the first checkable signal is whether Pulsenmore files a new submission covering guidance software rather than treating it as a change to the cleared system.