GE HealthCare Technologies has been granted a patent for an x-ray image laterality detection system. The system uses a neural network model to analyze x-ray images and determine their laterality. If the assigned laterality class is not the target laterality, the system adjusts the image to derive a corrected x-ray image. The system can alert the user whether the unclassified medical image has the target laterality. GlobalData’s report on GE HealthCare Technologies gives a 360-degree view of the company including its patenting strategy. Buy the report here.
According to GlobalData’s company profile on GE HealthCare Technologies, Nucleoside chemical synthesis was a key innovation area identified from patents. GE HealthCare Technologies's grant share as of September 2023 was 44%. Grant share is based on the ratio of number of grants to total number of patents.
Laterality detection in medical images using a neural network
A recently granted patent (Publication Number: US11776150B2) describes a method for detecting the laterality of a medical image using a neural network model. The method involves executing the neural network model to analyze an unclassified medical image and determine its laterality. If the detected laterality is not the target laterality, the method includes adjusting the image to derive a corrected medical image with the desired laterality. The user is then alerted about the laterality of the image based on the analysis.
The method also includes the use of a user interface manager to receive user input and generate a user policy. The laterality of the unclassified medical image can be adjusted based on this user policy. The neural network model used in the method is trained with training medical images and observed laterality classes associated with those images. This allows the model to assign a laterality class to the unclassified medical image and determine if it matches the target laterality. If the assigned laterality class is not the target laterality, the image is adjusted to derive a corrected medical image with the desired laterality. If the assigned laterality class is the target laterality, the unclassified medical image is outputted as is.
The method also includes the use of a metadata editor to update metadata associated with the unclassified medical image based on the detected laterality class. Additionally, a digital marker indicating the laterality of the output medical image can be generated and overlaid on the image.
Overall, this patented method provides a way to automatically detect and adjust the laterality of medical images using a neural network model. It offers potential benefits in improving the accuracy and efficiency of laterality detection in medical imaging, particularly in cases where lead markers are not present. The method also allows for user input and customization through the user interface manager, providing flexibility in adjusting the laterality based on specific requirements or preferences.