patent.fyi
Applications  /  US 2026/0354731
United States Patent Application Publication

US 2026/0354731 A1

Published
Nov 12, 2026
Filed
May 7, 2026
Earliest date on the face
May 9, 2025
Classification
A61B 5/346
First published here
Nov 19, 2026

Generated automatically. Not reviewed by a person. Not legal advice or an opinion.

Section 1 · machine-generated

What if?

Questions about how individual parts of claim 1 might be changed. A question may describe something different from what the claim language recites, or a version of it; no view is offered on which.

1.1Training from unannotated recordings

What ifWhat if training used unannotated recordings and rhythm decisions came from timing rules?
Claim language“the third data labeling at least one medical or technical state”

A reconstruction network learns to recover masked waveform intervals from surrounding samples. A peak detector operates on the reconstructed waveform, and timing rules flag runs of irregular beat intervals.

1.3Analysis with locked parameters

What ifWhat if the second step applied fixed model parameters to classify each incoming segment?
Claim language“a learning and analyzing step”

Training produces a waveform classifier and fixed normalization constants. During analysis, the processor applies those constants to each segment and runs a forward pass with the stored classifier weights.

1.4Direct beat feature classification

What ifWhat if state predictions came directly from beat timing and shape measurements without a waveform prediction?
Claim language“a deviation between the first data of the first biological signals”

A detector identifies ventricular peaks and measures successive beat intervals, ventricular complex widths, and waveform amplitudes. A decision tree trained on these measurements assigns a rhythm label to each segment.

1.5Independent fitting from calibration beats

What ifWhat if a separate process fitted the individualized model from labeled calibration beats using newly initialized parameters?
Claim language“to transform the generalized model used in the first step”

A peak detector extracts beat intervals and ventricular complex widths from a labeled calibration recording. A logistic classifier fitted to those measurements predicts beat labels in subsequent segments from the same subject.

Section 2 · machine-generated

Variations

Further arrangements of the same subject matter, described here as of the publication date at the foot of this page.

Other field of use

Vibration monitoring for industrial pumps

Housing accelerometers record pump vibration, and each training record includes its sensor mounting position and a maintenance inspection label such as normal operation or shaft misalignment. A waveform predictor learns features through an adversarial classifier that predicts mounting position, and a fault model learns to classify inspection states from residual spectra. For a newly instrumented pump, unlabeled vibration windows update the predictor, and per-band residual scaling from those windows adapts the fault model to that pump.

Component substitution

Boosted trees for residual classification

A gradient-boosted decision tree ensemble replaces the generalized state predictor and receives residual features from a waveform predictor trained with an electrode-configuration classifier. For each electrocardiogram window, the feature extractor computes the median absolute residual, its 95th percentile, and residual power in fixed frequency bands. At enrollment, unlabeled windows update the waveform predictor and yield per-feature percentile mappings from the enrollment distribution to the training distribution, which the individualized model applies before evaluating the trees.

State space prediction of cardiac waveforms

A linear dynamical state space model with a Kalman filter serves as the waveform abstraction model, and each electrode configuration has its own observation matrix. Training fits the latent dynamics and observation matrices to predict the next sample while an adversarial classifier penalizes electrode-configuration information in the latent state. A state classifier learns from labeled windows summarized by the magnitude and autocorrelation of the one-step prediction errors. On a target recording, unlabeled windows fit the observation matrix and noise covariance for the recorded configuration, and the fitted covariance normalizes prediction errors before the state classifier evaluates them.

Parameter or scale variation

Short overlapping windows for rhythm analysis

The processor resamples electrocardiogram waveforms to 250 samples per second and segments them into 8-second windows with a 2-second stride. During training, an adversarial classifier predicts the lead configuration from the waveform features, and a rhythm model maps each window's prediction residual to state probabilities. The first 60 seconds of an unlabeled target recording update the waveform predictor and set per-band residual scales for the individualized rhythm model. The model evaluates each subsequent window and averages the latest five probability vectors before assigning a rhythm label.

Combination with existing systems

Integration with bedside telemetry alarms

A bedside telemetry receiver supplies timestamped electrocardiogram samples, lead configuration, and electrode contact status to a waveform analysis service. The service starts with a waveform predictor trained against a lead-configuration classifier and a residual-based state model trained on annotated telemetry records. The service uses unlabeled windows with stable electrode contact to update the waveform predictor and calibrate the state model's residual inputs for the current recording. It sends state probabilities and window timestamps to the telemetry alarm manager, which applies its persistence timer before issuing a notification.

Automated or data-driven variant

Automatic recalibration after electrode changes

An acquisition controller records active channel identifiers and contact impedance, then starts recalibration when the channel set changes or impedance remains outside its configured range. During initial training, an adversarial classifier predicts electrode configuration from the waveform features, and a state model learns from the prediction residuals. During recalibration, software selects unlabeled windows with stable contact and no clipping, then updates the waveform predictor and computes new per-band residual offsets for the individualized state model. The controller activates the updated predictor and offsets together at a window boundary.

Specific version of a part

Beat interval histogram input

An acquisition module detects ventricular peaks and accumulates counts in 20-millisecond beat interval bins over 30 seconds. The analyzer computes the median interval and interquartile range from the bin counts to flag fast or irregular rhythms.

Lead metadata and rhythm labels

A training record stores a 10-second electrocardiogram segment sampled at 250 hertz as signed 16-bit integers. The record also stores the electrode positions, amplifier gain, and an annotated rhythm code for the segment.

Electrode spacing with waveform samples

An input record contains a 30-second differential electrocardiogram sampled at 500 hertz. The record specifies the center-to-center electrode spacing in millimeters and identifies which electrode connects to the positive amplifier input.

Adversarial encoding with residual classification

A convolutional encoder and decoder reconstruct each waveform while a lead classifier uses a gradient reversal layer to penalize lead information in the encoder features. A rhythm classifier receives the sample-by-sample reconstruction errors and predicts the annotated rhythm code.

Subject calibration with updated normalization

A two-minute recording updates the channel means and variances in the trained abstraction model. The generalized classifier uses the same recording to update its input normalization statistics for the resulting reconstruction errors, forming an individualized classifier that predicts rhythm labels on subsequent segments.

Section 3 · quoted from USPTO records

Other work

Passages reproduced word for word from two earlier-published U.S. patent documents, placed beside the language of claim 1. No characterization is offered. Each passage is marked A or B for the document it comes from; ¶ numbers are the paragraph numbers in the USPTO’s published full text of that document.

Document AUS 2021/0145368 A1
Published
May 20, 2021
Filed
Apr 2, 2019
Class
A61B 5/7278
Document BUS 2022/0222529 A1
Published
Jul 14, 2022
Filed
Feb 18, 2022
Class
G06N 3/08
Claim 1
Document A
Document B
1.1the first biological signals comprise first data, second data and third data, the first data being values measured over a period of time, the second data being values specifying a feature of the manner used to measure the first data, and the third data labeling at least one medical or technical state assigned to the first biological signals; and
A biological-information evaluating device according to an embodiment of the present disclosure includes a signal processing circuit that generates a feature value of an observation target waveform for each of pieces of observation data obtained through observation of a living body for a predetermined period, on a basis of a plurality of pieces of partial observation data contained in each of the pieces of observation data and each having a period shorter than an observation
A · US 2021/0145368 A1 · abstract
1.2the second biological signal comprises first data and second data, the first data being values measured over a period of time and the second data being values specifying a feature of the manner used to measure the first data;
A biological-information evaluating device according to an embodiment of the present disclosure includes a signal processing circuit that generates a feature value of an observation target waveform for each of pieces of observation data obtained through observation of a living body for a predetermined period, on a basis of a plurality of pieces of partial observation data contained in each of the pieces of observation data and each having a period shorter than an observation
A · US 2021/0145368 A1 · abstract
1.3the method comprises a first step and a second step, the first step being a learning step and the second step being a learning and analyzing step, wherein
1.4the first step processes the first data and the second data of the first biological signals by using an abstraction model and a generalized model, wherein the abstraction model is forced to learn features for a prediction of the first data of first biological signals by using a classifier, the features being independent of the second data, and wherein the generalized model uses a deviation between the first data of the first biological signals and their prediction to create a prediction of the third data of the first biological signals; and
The classifier is proposed to predict a label of a direct exchange domain so as to minimize the loss function of the exchange domain; the domain labels in the loss function are interchanged, and the domain information is implicitly encoded in the prediction probability of the adversarial domain classifier, so the optimization of the loss function based on the interchanged domain makes the learned features independent of the domain.
B · US 2022/0222529 A1 · ¶ 33
1.5the second step processes the second biological signal by the abstraction model used in the first step to adapt the abstraction model and to transform the generalized model used in the first step into an individualized model to create a prediction of third data of the second biological signal, wherein the third data labelling at least one medical or technical state connected with the second biological signal.
According to the present application, meta-knowledge based on domain-invariant features is introduced, and common domain features of the same kind of tasks are learnt from different domains corresponding to different data sets to obtain highly transferable domain-invariant knowledge; after meta-knowledge fine-tuning, the model can be fine-tuned for different domains of the same kind of tasks, with a better parameter initialization ability and a higher generalization ability,
B · US 2022/0222529 A1 · ¶ 4

An empty cell means only that no passage was reproduced there.

Statements in these documents
In addition to a human, the target from which the biological information is acquired may include an animal or the like.
A · US 2021/0145368 A1 · ¶ 2
The biological-information evaluating device and the method of evaluating biological information according to the embodiment of the present disclosure make full use of biological information in which periodic variation is able to be seen, which makes it possible to know a difference between states with a small number of samples and a short measuring period.
A · US 2021/0145368 A1 · ¶ 8
In the fine-tuning stage, the existing compression method of the pre-trained language model for a downstream task is to fine-tune the specific data set of the downstream task, and the effect of the compressed model obtained by training is limited by the specific data set of this kind of tasks.
B · US 2022/0222529 A1 · ¶ 3
In view of the shortcomings of the prior art, the purpose of the present application is to provide a method for meta-knowledge fine-tuning and platform based on domain-invariant features.
B · US 2022/0222529 A1 · ¶ 4
Cite aspatent.fyi, US 2026/0354731 A1, claim 1, first published 2026-11-19T02:47:19-05:00, https://patent.fyi/US20260354731/, SHA-256 3db53138e3a8c4ed0ace59715c6f219e77a748b523ec77e882941147019ef8cb
First published
Permanent addresshttps://patent.fyi/US20260354731/
Content fingerprint (SHA-256)3db53138e3a8c4ed0ace59715c6f219e77a748b523ec77e882941147019ef8cb
Methodv2026-11-18.2 · automated, not reviewed by a person
Correctionscorrections@patent.fyi