# US 2026/0224168 A1 · Denoising biological signals

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2026/0224168 A1
- Document ID: US20260224168A1
- Status: Published application
- Kind: A1
- Application number: 19/454,807
- Filed: 2026-01-21
- Published: 2026-08-06
- Earliest claimed: 2025-02-04
- Assignees: Pacesetter, Inc.
- Applicants: Pacesetter, Inc.
- Inventors: Ruoli Jiang; Shiloh Sison
- Web page: https://patentgrep.com/patent/US20260224168A1

## CPC

- A61B 5/7203: {for noise prevention, reduction or removal}
- A61N 1/3718: {Monitoring of or protection against external electromagnetic fields or currents}
- A61B 5/686: {Permanently implanted devices, e.g. pacemakers, other stimulators, biochips (A61B5/6861 takes precedence)}
- A61B 5/7217: {of noise originating from a therapeutic or surgical apparatus, e.g. from a pacemaker}
- A61B 5/7267: {involving training the classification device}
- A61B 5/7271: {Specific aspects of physiological measurement analysis (specific diagnostics methods using bioelectric or biomagnetic signals A61B5/316)}
- A61B 2562/0223: Magnetic field sensors
- A61N 1/362: Heart stimulators (heart defibrillators A61N1/39)
- A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D)
- A61N: ELECTROTHERAPY; MAGNETOTHERAPY; RADIATION THERAPY; ULTRASOUND THERAPY (measurement of bioelectric currents A61B; surgical instruments, devices or methods for transferring non-mechanical forms of energy to or from the body A61B18/00; anaesthetic apparatus in general A61M; incandescent lamps H01K; infrared radiators for heating H05B)

## Abstract

Disclosed herein is an implantable medical device (IMD), system and method for denoising biological signals. The IMD generates a biological signal representative of a physiologic parameter of a patient and normally analyzes the biological signal to detect a characteristic of interest (COI) indicative of a pathophysiologic condition of the patient. The present technology denoises, when the IMD is exposed to a noise causing condition, noisy biological signal segments of the biological signal generated while the IMD is exposed to the noise causing condition using a denoising prediction model to obtain denoised biological signal segments. The denoising prediction model is trained to predict a denoised biological signal segment from a noisy biological signal segment. The denoised biological signal segments can then be analyzed to detect the COI indicative of the pathophysiologic condition of the patient.

## Claims

### Claim 1

An implantable medical device (IMD) comprising:
a sensing circuitry configured to generate a biological signal representative of a physiologic parameter of a patient;
a memory storing specific executable instructions and a denoising prediction model trained to predict a denoised biological signal segment from a noisy biological signal segment; and
a processor connected to the sensing circuitry and the memory and configured to execute the specific executable instructions to:
normally analyze the biological signal generated by the sensing circuitry to detect a characteristic of interest (COI) indicative of a pathophysiologic condition of the patient;
denoise, when the IMD is exposed to a noise causing condition, noisy biological signal segments of the biological signal generated by the sensing circuitry while the IMD is exposed to the noise causing condition using the denoising prediction model to obtain denoised biological signal segments; and
analyze the denoised biological signal segments to detect the COI indicative of the pathophysiologic condition of the patient.

### Claim 2 (depends on 1)

The IMD of claim 1, further comprising a communication circuitry configured to receive a noise mode programming command from an external device, wherein the processor is connected to the communication circuitry and is configured to execute the specific executable instructions to denoise, in response to reception of the noise mode programming command, the noisy biological signal segments using the denoising prediction model to obtain the denoised biological signal segments.

### Claim 3 (depends on 1)

The IMD of claim 1, wherein the processor is configured to execute the specific executable instructions to:
analyze the biological signal generated by the sensing circuitry during a noise-sampling period to detect noise or interference in the biological signal; and
denoise, in response to detection of the noise or interference in the biological signal, the noisy biological signal segments using the denoising prediction model to obtain the denoised biological signal segments.

### Claim 4 (depends on 1)

The IMD of claim 1, further comprising a communication circuitry connected to a communication coil, wherein the processor is connected to the communication circuitry and is configured to execute the specific executable instructions to:
monitor for noise on the communication coil to thereby detect when the IMD is exposed to the noise causing condition; and
denoise, when noise is detected on the communication coil, the noisy biological signal segments using the denoising prediction model to obtain the denoised biological signal segments.

### Claim 5 (depends on 1)

The IMD of claim 1, further comprising a magnetic field sensor configured to detect a magnetic field, wherein the processor is connected to the magnetic field sensor and is configured to execute the specific executable instructions to:
denoise, in response to the magnetic field sensor detecting the magnetic field, the noisy biological signal segments using the denoising prediction model to obtain the denoised biological signal segments.

### Claim 6 (depends on 1)

The IMD of claim 1, further comprising a communication circuitry connected to a communication coil, wherein the processor is connected to the communication circuitry and is configured to execute the specific executable instructions to:
determine, when the IMD is exposed to electromagnetic interference (EMI), an EMI-induced voltage on the communication coil; and
denoise, when the IMD is exposed to EMI, the noisy biological signal segments to obtain the denoised biological signal segments by inputting the noisy biological signal segments and the determined EMI-induced voltage on the communication coil into the denoising prediction model.

### Claim 7 (depends on 1)

The IMD of claim 1, wherein:
the memory stores a segment classification model trained to classify biological signal segments as noisy or non-noisy biological signal segments; and
the processor is configured to execute the specific executable instructions to:
segment, when the IMD is exposed to the noise causing condition, the biological signal generated by the sensing circuitry while the IMD is exposed to the noise causing condition into a plurality of biological signal segments;
classify the plurality of biological signal segments using the segment classification model; and
denoise the biological signal segments classified as noisy by the segment classification model using the denoising prediction model to obtain denoised biological signal segments.

### Claim 8 (depends on 1)

The IMD of claim 1, further comprising a communication circuitry configured to:
transmit, to an external device comprising a segment classification model trained to classify biological signal segments as noisy or non-noisy biological signal segments, the biological signal generated by the sensing circuitry while the IMD is exposed to the noise causing condition or a plurality of biological signal segments thereof; and
receive, from the external device, a respective classification of the plurality of biological signal segments as noisy or non-noisy, wherein the processor is connected to the communication circuitry and is configured to execute the specific executable instructions to denoise biological signal segments classified as noisy by the segment classification model using the denoising prediction model to obtain denoised biological signal segments.

### Claim 9

A system for denoising biological signals comprising:
an implantable medical device (IMD) comprising:
a sensing circuitry configured to generate a biological signal representative of a physiologic parameter of a patient;
a first communication circuitry configured to perform wireless communication with an external device;
a first memory storing first specific executable instructions; and
a first processor connected to the sensing circuitry, the first communication circuitry and the first memory and configured to execute the first specific executable instructions to:
normally analyze the biological signal sensed by the sensing circuitry to detect a characteristic of interest (COI) indicative of a pathophysiologic condition of the patient;
control, when the IMD is exposed to a noise causing condition, the first communication circuitry to transmit, to the external device, a biological signal generated by the sensing circuitry while the IMD is exposed to the noise causing condition, or biological signal segments thereof; and
the external device comprising:
a second communication circuitry configured to perform wireless communication with the IMD;
a second memory storing second specific executable instructions and a denoising prediction model trained to predict a denoised biological signal segment from a noisy biological signal segment; and
a second processor connected to the second communication circuitry and the second memory and configured to execute the second specific executable instructions to:
denoise, using the denoising prediction model, noisy biological signal segments of the biological signal, or of the biological signal segments thereof, received by the second communication circuitry from the IMD to obtain denoised biological signal segments; and
control the second communication circuitry to transmit, to the IMD, the denoised biological signal segments or information of the COI indicative of the pathophysiologic condition of the patient detected by analysis of the denoised biological signal segments.

### Claim 10 (depends on 9)

The system of claim 9, wherein the second processor is configured to execute the second specific executable instructions to:
analyze the denoised biological signal segments to detect the COI indicative of the pathophysiologic condition of the patient; and
control the second communication circuitry to transmit the information of the COI indicative of the pathophysiologic condition of the patient to the IMD.

### Claim 11 (depends on 9)

The system of claim 9, wherein:
the second processor is configured to execute the second specific executable instructions to control the second communication circuitry to transmit the denoised biological signal segments to the IMD; and
the first processor is configured to execute the first specific executable instructions to analyze the denoised biological signal segments to detect the COI indicative of the pathophysiologic condition of the patient.

### Claim 12 (depends on 9)

The system of claim 9, wherein:
the second communication circuitry is configured to transmit a noise mode programming command to the IMD; and
the first processor is configured to execute the first specific executable instructions to control, in response to reception of the noise mode programming command, the first communication circuitry to transmit, to the external device, the biological signal generated by the sensing circuitry while the IMD is exposed to the noise causing condition, or the biological signal segments thereof.

### Claim 13 (depends on 9)

The system of claim 9, wherein the first processor is configured to execute the first specific executable instructions to:
analyze the biological signal generated by the sensing circuitry during a noise-sampling period to detect noise or interference in the biological signal; and
control, in response to detection of noise or interference in the biological signal, the first communication circuitry to transmit, to the external device, the biological signal generated by the sensing circuitry while the IMD is exposed to the noise causing condition, or the biological signal segments thereof.

### Claim 14 (depends on 9)

The system of claim 9, wherein:
the second memory stores a segment classification model trained to classify biological signal segments as noisy or non-noisy biological signal segments; and
the second processor is configured to execute the second specific executable instructions to:
classify the biological signal segments using the segment classification model; and
denoise, using the denoising prediction model, the biological signal segments classified as noisy by the segment classification model to obtain the denoised biological signal segments.

### Claim 15

A method for denoising biological signals comprising:
an implantable medical device (IMD) generating a biological signal representative of a physiologic parameter of a patient;
the IMD normally analyzing the biological signal to detect a characteristic of interest (COI) indicative of a pathophysiologic condition of the patient;
denoising, when the IMD is exposed to a noise causing condition, noisy biological signal segments of the biological signal generated while the IMD is exposed to the noise causing condition using a denoising prediction model to obtain denoised biological signal segments, wherein the denoising prediction model is trained to predict a denoised biological signal segment from a noisy biological signal segment; and
analyzing the denoised biological signal segments to detect the COI indicative of the pathophysiologic condition of the patient.

### Claim 16 (depends on 15)

The method of claim 15, wherein the denoising comprises denoising, in response to reception of a noise mode programming command, the noisy biological signal segments using the denoising prediction model to obtain the denoised biological signal segments.

### Claim 17 (depends on 15)

The method of claim 15, further comprising analyzing the biological signal generated during a noise-sampling period to detect noise or interference in the biological signal, wherein the denoising comprises denoising, in response to detection of noise or interference in the biological signal, the noisy biological signal segments using the denoising prediction model to obtain the denoised biological signal segments.

### Claim 18 (depends on 15)

The method of claim 15, further comprising:
segmenting, when the IMD is exposed to the noise causing condition, the biological signal generated while the IMD is exposed to the noise causing condition into a plurality of biological signal segments; and
classifying the plurality of biological signal segments using a segment classification model trained to classify biological signal segments as noisy or non-noisy biological signal segments, wherein the denoising comprises denoising the biological signal segments classified as noisy by the segment classification model to obtain the denoised biological signal segments.

### Claim 19 (depends on 18)

The method of claim 18, wherein the analyzing the denoised biological signal segments comprises analyzing the denoised biological signal segments and the biological signal segments classified as non-noisy biological signal segments by the segment classification model to detect the COI indicative of the pathophysiologic condition of the patient.

### Claim 20 (depends on 18)

The method of claim 18, further comprising determining, when the IMD is exposed to electromagnetic interference (EMI), an EMI-induced voltage on a communication coil of the IMD, wherein the classifying comprises classifying the plurality of biological signal segments by inputting the plurality of biological signal segments and the determined EMI-induced voltage on the communication coil into the segment classification model.

### Claim 21 (depends on 18)

The method of claim 18, further comprising determining, when the IMD is exposed to electromagnetic interference (EMI), an EMI-induced voltage on a communication coil of the IMD, wherein the denoising comprises denoising, when the IMD is exposed to EMI, the noisy biological signal segments to obtain the denoised biological signal segments by inputting the noisy biological signal segments and the determined EMI-induced voltage into the denoising prediction model.
