# US 11,992,322 B2 · Heart rhythm detection method and system using radar sensor > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 11,992,322 B2 - Document ID: US11992322B2 - Status: Granted - Kind: B2 - Application number: 17/218,142 - Filed: 2021-03-30 - Published: 2022-10-06 - Granted: 2024-05-28 - Earliest claimed: 2021-03-30 - Assignees: Ionetworks Inc. - Applicants: ioNetworks INC. - Inventors: Jing-Ming Guo; Ting Lin; Chia-Fen Chang; Jeffry Susanto; Yi-Hsiang Lin; Po-Cheng Huang; Yu-Wen Wei - Web page: https://patentgrep.com/patent/US11992322B2 ## CPC - A61B 5/346: Analysis of electrocardiograms - A61B 5/0507: using microwaves or terahertz waves - A61B 5/333: Recording apparatus specially adapted therefor - A61B 5/7203: {for noise prevention, reduction or removal} - A61B 5/7225: {Details of analogue processing, e.g. isolation amplifier, gain or sensitivity adjustment, filtering, baseline or drift compensation (input circuits for detecting, measuring, or recording bioelectric or biomagnetic signals A61B5/30; specific diagnostic methods using bioelectric or biomagnetic signals A61B5/316)} - A61B 5/7257: {using Fourier transforms} - A61B 5/7264: {Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems} - A61B 5/7267: {involving training the classification device} - G01S 7/415: {Identification of targets based on measurements of movement associated with the target} - G01S 7/417: {involving the use of neural networks} - G01S 13/34: using transmission of continuous, frequency-modulated waves while heterodyning the received signal, or a signal derived therefrom, with a locally-generated signal related to the contemporaneously transmitted signal - G01S 13/583: {using transmission of continuous unmodulated waves, amplitude-, frequency-, or phase-modulated waves and based upon the Doppler effect resulting from movement of targets} - G01S 13/88: Radar or analogous systems specially adapted for specific applications (electromagnetic prospecting or detecting of objects, e.g. near-field detection, G01V3/00) - G06N 3/0464: Convolutional networks [CNN, ConvNet] - G06N 3/084: Backpropagation, e.g. using gradient descent - A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D) - G01S: RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract A heart rhythm detection method and system by using radar sensor is capable of collecting an original signal using a radar sensor toward at least one subject, and converting the original signal to a two dimensional image information (i.e., spectrogram) using the concept of image vision. Then, the neural network automatically learns which heartbeat frequency should be focused on and which heartbeat frequency should be filtered out in the two dimensional image information through deep learning, so that the heartbeat frequencies can be extracted effectively. ## Claims ### Claim 1 A heart rhythm detection method using radar sensor, comprising the following steps: collecting a one dimensional raw signal by aligning at least one radar sensor toward at least one subject; and converting the one dimensional collected raw signal to two dimensional image information, then learning from the obtained two dimensional image information and automatically filtering out a noise signal by means of a neural network model so as to learn a relationship between a heartbeat frequency and non-heartbeat frequency of the two dimensional image information thereby extracting the heartbeat frequency; wherein the one dimensional raw signal is first filtering processed then further converted to time-frequency dimension by means of a Short-Time Fourier Transform (STFT), thereby transforming the one dimensional raw signal to the two dimensional image information, and the two dimensional image information is a time-frequency view of a two dimensional image; and wherein the neural network model applies AvgPooling for training, and finally adding up a layer of GlobalAvgPooling2D and two layers of Dense for a final characteristic extraction. ### Claim 2 (depends on 1) The heart rhythm detection method using radar sensor according to claim 1, wherein the radar sensor is a one-millimeter wave radar. ### Claim 3 (depends on 1) The heart rhythm detection method using radar sensor according to claim 1, wherein the filtering process is applied to keep the one dimensional raw signal within the frequency domain of heart rhythm by means of the high-pass filtering and/or band-pass filtering process. ### Claim 4 (depends on 1) The heart rhythm detection method using radar sensor according to claim 1, further comprising first detecting and determining whether the subject is existed to maintain the correctness of signal sampling. ### Claim 5 A heart rhythm detection system using radar sensor, comprising: at least one radar sensor, used for collecting a one dimensional raw signal by aligning toward at least one subject; and at least one server equipment, used for connecting to the radar sensor and capable of receiving the one dimensional raw signal thus converting the collected one dimensional raw signal to two dimensional image information, then learning from the obtained two dimensional image information and automatically filtering out a noise signal by means of a neural network model so as to learn a relationship between a heartbeat frequency and non-heartbeat frequency of the two dimensional image information thereby automatically extracting the heartbeat frequency; wherein the one dimensional raw signal is first filtering processed then further converted to time-frequency dimension by means of a Short-Time Fourier Transform (STFT), thereby transforming the one dimensional raw signal to the two dimensional image information, and the two dimensional image information is a time-frequency view of a two dimensional image; and wherein the neural network model applies AvgPooling for training and finally adding up a layer of GlobalAvgPooling2D and two layers of Dense for a final characteristic extraction. ### Claim 6 (depends on 5) The heart rhythm detection system using radar sensor according to claim 5, wherein the server equipment comprises at least one processor and at least one computer readable recording media, and the computer readable recording media stores at least one application program and further stores computer readable instructions such that, upon running the stored computer readable instructions by the processor, the application program can be executed thus converting a raw signal to two dimensional image information, then learning from the obtained two dimensional image information and automatically filtering out the noise signal by means of the neural network model so as to learn the relationship between the heartbeat frequency and non-heartbeat frequency of the two dimensional image information thereby automatically extracting the heartbeat frequency.