Back to results
US 12,608,976 B2Granted
Method and system for assisting in determining identity of scanned person in CT scanning, and CT scanner
AssigneeSiemens Shanghai Medical Equipment Ltd.
InventorsZe Ben Li, Li Na Li, Jia Feng, Jie Qing Liu
Life of this patent
Estimating the term…
Abstract
Techniques are provided for determining an identity of a scanned person in computed tomography (CT) scanning is provided. The techniques include the use of a system for assisting in determining an identity of a scanned person in CT scanning and a CT scanner.
Claims13
A method for determining an identity of a person in computed tomography (CT) scanning, comprising:
- obtaining identity information of a person;
- collecting facial image information of the person after the person is positioned;
- determining, using a facial recognition algorithm, whether the person is a match based on the identity information of the person;
- performing, using natural language processing (NLP), a voice dialog with the person based on the identity information of the person after the person is positioned;
- determining whether the person is a match based on the voice dialog;
- in response to (i) the person being determined to not be a match based on the facial recognition algorithm, or (ii) the person being determined to not be a match based on the voice dialog, transmitting a reminder to an operator to determine an identity of the person;
- and
- in response to (i) the person being determined to be a match based on the facial recognition algorithm, and (ii) the person being determined to be a match based on the voice dialog both, initiating a CT scan.
5 more elements
A system for determining an identity of a person in computed tomography (CT) scanning, comprising:
- a camera configured to collect facial image information of a person after the person is positioned;
- a vision hardware platform implementing a convolutional neural network (CNN) model,
- wherein the vision hardware platform is configured to:
- receive identity information of the person and the facial image information;
- extract facial feature information corresponding to the identity information of the person in the facial image information via the CNN model;
- compare the facial feature information with the identity information of the person;
- and
- generate an image mismatch signal when a mismatch between the facial feature information with the identity information of the person is determined;
- a speaker configured to receive a sound signal and to play a voice;
- a microphone configured to receive a voice and to convert the voice into a sound signal;
- a voice hardware platform implementing a voice dialog processor, wherein the voice dialog processor is configured to:
- receive the identity information of the person;
- convert question information corresponding to the identity information of the person into a sound signal;
- and
- transmit the sound signal to the speaker;
- receive, via the microphone, an answer sound signal answered by the person;
- extract, from the answer sound signal, voice feature information corresponding to the identity information of the person;
- compare the voice feature information with the identity information of the person;
- and
- generate a voice mismatch signal when a mismatch between the voice feature information and the identity information of the person is determined;
- and
- reminder circuitry configured to transmit a reminder to determine an identity of the person to an operator in response to receiving the image mismatch signal or the voice mismatch signal.
19 more elements
Dependencies are inferred from the claim text.