Ultrasonic diagnostic apparatus with machine learning

ChangeBridge: Patent Grants - Diagnosis & Surgery (A61B)

Published March 31st, 2026

Detected March 31st, 2026

Summary

USPTO granted Patent US12588899B2 to Canon Kabushiki Kaisha for an ultrasonic diagnostic apparatus that uses machine learning models to generate estimated images from reduced transmission/reception ultrasonic scan data. The patent covers technology that trains models on data from varying numbers of ultrasonic transmissions to produce high-quality images from fewer scans.

View original document View source feed page

What changed

USPTO granted Patent US12588899B2 to Canon Kabushiki Kaisha on March 31, 2026. The patent covers an ultrasonic diagnostic apparatus using machine learning where a model is trained on learning data including first data from a first transmission/reception and second data from a second transmission/reception with a larger number of transmissions. The trained model generates estimated images equivalent to data from the higher-number-of-transmissions scan using data from a reduced transmission/reception equivalent to the first transmission/reception. Inventors include Shoya Sasaki, Naoya Iizuka, and Kenichi Nagae.

This is a patent grant notice, not a regulatory rulemaking. It does not create compliance obligations for any party. Medical device manufacturers and healthcare providers utilizing ultrasonic diagnostic equipment may wish to review this patent when developing new products or acquiring imaging technology to understand existing intellectual property rights in this technical space.

Source document (simplified)

← USPTO Patent Grants

Ultrasonic diagnostic apparatus, learning apparatus, and image processing method

Grant US12588899B2 Kind: B2 Mar 31, 2026

Assignee

Canon Kabushiki Kaisha

Inventors

Shoya Sasaki, Naoya Iizuka, Kenichi Nagae

Abstract

An ultrasonic diagnostic apparatus, comprising: an ultrasonic probe which scans an observation region in an object with an ultrasonic wave; and an estimated image generating unit which, by using a model having been machine-learned using learning data including first data based on a first received signal that is obtained by first transmission/reception of an ultrasonic wave and second data based on a second received signal that is obtained by second transmission/reception that represents a larger number of transmissions/receptions than the first transmission/reception of the ultrasonic wave, generates an estimated image equivalent to image data obtained by the second transmission/reception from third data based on a third received signal that is obtained by transmission/reception equivalent to the first transmission/reception of the ultrasonic wave by the ultrasonic probe.