Koninklijke Philips had 134 patents in artificial intelligence during Q3 2023. Koninklijke Philips NV has filed patents related to various medical imaging technologies. One patent describes a device for localizing stenoses in angiograms by analyzing image data from both angiographic and treatment X-ray images. Another patent focuses on supporting MR imaging by using RF sensors to determine the presence and location of surface RF coils. Additionally, there are patents for a radiology report analysis method that assigns uncertainty scores to radiology findings and a patient appointment notification method that predicts the likelihood of a patient missing an appointment and adjusts notification strategies accordingly. Lastly, there is a patent for an image processing system that utilizes machine learning to generate dark-field and/or phase contrast projection images for tomographic X-ray imaging. GlobalData’s report on Koninklijke Philips gives a 360-degreee view of the company including its patenting strategy. Buy the report here.
Koninklijke Philips grant share with artificial intelligence as a theme is 30% in Q3 2023. Grant share is based on the ratio of number of grants to total number of patents.
Application: Stenosis localization (Patent ID: US20230274437A1)
The patent filed by Koninklijke Philips NV relates to a device for localizing stenoses in angiograms. The device includes an image supply, a data processor, and an output. The image supply provides a first angiographic image and a second treatment X-ray image. The data processor identifies and delineates the stenosis in the first image based on the first image and device-related content present in the second image. It also detects the interventional device in the second image and provides a direct identification of structures in the first image that are most similar to the device as detected in the second image. The output provides an indication of the stenosis.
The device is configured to detect and identify the interventional device in the second image by determining geometric parameters such as shape, size, orientation, bending radius, diameter, and direction. It analyzes the first image and assesses whether any of these geometric parameters can be found in the first image to identify and delineate the stenosis. The device also provides a segmentation of the stenosis in the first image based on the device-related content present in the second image.
The data processor uses a non-registered image-based identification procedure to provide the identification of the stenosis in the first image. It assesses the first image in view of structural parameters relating to the interventional device without geometrical transferring or registering procedure. The device can also use a self-learning algorithm or a convolutional network configuration to learn the relationship between the first and second images.
The patent also describes a medical system that includes the device for localizing stenosis in angiograms and an image acquisition device with an X-ray source and detector. The system is used for annotating medical images of stenosis treatment.
In summary, the patent describes a device and method for localizing stenoses in angiograms. The device uses a combination of angiographic and treatment X-ray images to identify and delineate the stenosis. It detects and identifies the interventional device in the treatment X-ray image and provides a direct identification of structures in the angiographic image that are most similar to the device. The device can also provide a segmentation of the stenosis and use self-learning algorithms or convolutional network configurations for improved identification. The patent also describes a medical system that includes the device and an image acquisition device for annotating medical images of stenosis treatment.