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.
1.1Posture sensing by room camera
What ifWhat if a wall-mounted depth camera identifies the first posture while the ECG monitor records?
Claim language“an on-body posture component”
The camera estimates the trunk angle from the three-dimensional positions of the shoulders and hips. A processor uses timestamps to label the corresponding ECG interval when the trunk angle stays within the range assigned to the first posture.
1.2Posture entry at bedside
What ifWhat if an operator records the second posture through a bedside touchscreen while ECG acquisition continues?
Claim language“from the on-body posture component”
The operator selects the second posture from a menu and marks the start and end of the recording interval. The acquisition software attaches the selected posture to ECG samples within that interval using the monitor clock.
1.3One representative beat
What ifWhat if the processor extracts one representative beat from the first ECG recording?
Claim language“a plurality of first ECG waveforms”
A detector identifies R peaks and scores their surrounding samples for clipping, baseline drift and high-frequency noise. The processor copies a 600-millisecond window centered on the highest-scoring peak and stores that single beat as the first posture reference.
1.4Beat timing without waveform extraction
What ifWhat if the processor extracts R-peak times from the second ECG recording and builds a heartbeat-interval reference?
Claim language“a plurality of second ECG waveforms”
A detector emits a timestamp for each accepted R peak in the continuous ECG stream. The processor computes the median interval between successive peaks and stores it as the second posture reference for beat timing.
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
Posture calibration for equine ECG monitoring
A girth-mounted inertial sensor records trunk orientation while skin electrodes and a portable ECG recorder acquire a single ECG lead from a horse. The recorder timestamps both signals on a shared clock during level standing and standing on a sloped examination platform. Software extracts beats around their R peaks and computes a pointwise median waveform for each posture. A veterinary review screen displays both baselines with their measured trunk angles.
Component substitution
Joint angle sensors for posture measurement
Wearable electrogoniometers replace an inertial posture sensor and span the hip and knee on one side of the body. A synchronized logger samples joint angles and ECG while the individual holds an extended-knee standing position and a supported partial squat. Software groups the ECG beats by the recorded hip and knee angle ranges and aligns the beats at their R peaks. The system reports a median ECG waveform and the measured angle ranges for each position.
Parameter or scale variation
Four angle calibration with fixed windows
A chest-mounted accelerometer and ECG monitor acquire paired data at supported trunk inclinations of 15, 30, 45 and 60 degrees above horizontal. At each setting, the monitor records ECG at 500 samples per second for 90 seconds and uses the final 60 seconds for beat extraction. For each R peak with a complete window, the processor extracts 200 milliseconds before the peak and 400 milliseconds after it. The processor calculates a pointwise median across the extracted beats at each angle and reports all four waveforms.
Combination with existing systems
Calibration integrated with ambulatory telemetry
A wearable ECG transmitter and an on-body motion sensor send timestamped samples to a bedside telemetry gateway. The gateway groups calibration segments by posture, extracts R-peak-centered beats and stores their median waveforms under the active monitoring session. During subsequent monitoring, the gateway associates each stable-posture ECG segment with its stored baseline. The central telemetry viewer displays the current ECG and the associated baseline on the same voltage and time scales.
Automated or data-driven variant
Beat clustering for posture baseline generation
A processor extracts R-peak-centered beats from synchronized ECG and on-body posture recordings and groups them by recorded posture. Within each posture group, it subtracts each beat's median pre-QRS voltage and computes pairwise distances from the root-mean-square voltage difference between aligned beats. It applies agglomerative clustering with a distance cutoff estimated from repeated low-motion calibration recordings. The processor reports the pointwise median waveform of the largest cluster and the number of beats in that cluster for each posture.
Signal quality weighted waveform baselines
An ECG front end records a lead-off indicator alongside the ECG, and a chest-mounted accelerometer records movement during two calibration postures. Software aligns extracted beats at their R peaks and assigns zero weight to any beat window containing a lead-off indication. For the remaining beats, it assigns a weight inversely proportional to one plus the window's root-mean-square dynamic acceleration divided by a reference acceleration measured during quiet calibration. The system reports the pointwise weighted mean waveform and total beat weight for each posture.
Specific version of a part
Reference values for ECG measurements
The processor measures QRS duration and ST amplitude for each accepted beat in the first posture. It stores the 10th and 90th percentiles for each measurement and compares later beats in that posture against those ranges.
Report posture matched differences
The processor selects the reference waveform using the current posture label and calculates the difference in QRS duration and ST amplitude between the current beat and the reference. The display shows the posture label and signed differences in milliseconds and millivolts.
Chest mounted acceleration sensor
An adhesive chest patch holds a three-axis accelerometer in a fixed orientation relative to the torso. The monitor samples acceleration at 25 Hz and ECG at 500 Hz on a common clock, then estimates trunk tilt from the low-pass filtered acceleration during stationary intervals.
Beat windows centered on R peaks
A QRS detector locates R peaks in the ECG recorded during the second posture. For each accepted peak, the processor extracts a segment spanning 200 milliseconds before the peak and 400 milliseconds after it, omitting windows with clipped samples.
Median waveform for second posture
The processor aligns the accepted second-posture beat segments at their R peaks and subtracts each segment's pre-QRS voltage offset. It computes the median voltage across the segments at each aligned sample position to form the second posture baseline.
Shared axes for posture baselines
The display plots both baseline waveforms on shared voltage and time axes. Separate line patterns and posture labels identify the traces, and zero milliseconds marks the R-peak alignment point.
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.
1.1obtaining, while a body of an individual is positioned in a first posture, both: (i) first posture data from an on-body posture component positioned on the body of the individual; and (ii) first ECG data from an ECG monitor configured to obtain ECG data from the individual;
For example, when applied to a body of patient 103, the body sensor 141 captures biometric data (e.g., heart rate, ECG data, etc.) in real-time.
B · US 2020/0289063 A1 · ¶ 32
1.2obtaining, while the body of the individual is positioned in a second posture different from the first posture, both: (i) second posture data from the on-body posture component; and (ii) second ECG data from the ECG monitor;
The sensor 604 can be attached to a user 612 via electrodes that are attached to the user's skin. Thus, the wearable device 602 can, via the sensor 604, measure electrical activity of the user's heart over a period of time.
A · US 2016/0089047 A1 · ¶ 50
For example, when applied to a body of patient 103, the body sensor 141 captures biometric data (e.g., heart rate, ECG data, etc.) in real-time.
B · US 2020/0289063 A1 · ¶ 32
1.3extracting, from the first ECG data, a plurality of first ECG waveforms;
Signal processing can be performed on the ECG signal affected by baseline wander noise in order to determine a start time and an end time for individual waveforms in the ECG signal affected by baseline wander noise.
A · US 2016/0089047 A1 · abstract
For example, R-R intervals in the ECG data can be identified to detect heartbeats, and the ECG data relating to the identified heartbeats can be analyzed to classify the heartbeats and identify potentially irregular heartbeats and cardiac issues.
B · US 2020/0289063 A1 · ¶ 18
1.4extracting, from the second ECG data, a plurality of second ECG waveforms;
Signal processing can be performed on the ECG signal affected by baseline wander noise in order to determine a start time and an end time for individual waveforms in the ECG signal affected by baseline wander noise.
A · US 2016/0089047 A1 · abstract
For example, R-R intervals in the ECG data can be identified to detect heartbeats, and the ECG data relating to the identified heartbeats can be analyzed to classify the heartbeats and identify potentially irregular heartbeats and cardiac issues.
B · US 2020/0289063 A1 · ¶ 18
1.5generating, from the extracted plurality of first ECG waveforms, a first ECG waveform baseline;
Rather than using a unique reference for all the waveforms in the noisy ECG signal, the local reference can be set per each individual waveform. In one example, the local reference for each ECG waveform can be identified using a slope technique (i.e., the onset value for setting the local reference can be determined based on slope changes in the individual waveform).
A · US 2016/0089047 A1 · ¶ 44
1.6generating, from the extracted plurality of second ECG waveforms, a second ECG waveform baseline; and
Rather than using a unique reference for all the waveforms in the noisy ECG signal, the local reference can be set per each individual waveform. In one example, the local reference for each ECG waveform can be identified using a slope technique (i.e., the onset value for setting the local reference can be determined based on slope changes in the individual waveform).
A · US 2016/0089047 A1 · ¶ 44
1.7reporting one or both of the first ECG waveform baseline and the second ECG waveform baseline.
An empty cell means only that no passage was reproduced there.
Statements in these documents
For example, wearable technology can provide health monitoring and health metrics, music listening, global positioning system (GPS) capabilities, activity tracking, telephony services, internet browsing, etc. for the user that is wearing the wearable technology.
A · US 2016/0089047 A1 · ¶ 2
MCT empowers physicians with valuable information regarding the occurrence and regularity of a variety of heart conditions and irregularities.
B · US 2020/0289063 A1 · ¶ 2
For example, ECG data may be inaccurate due to problems in collecting the data.
B · US 2020/0289063 A1 · ¶ 3
Cite aspatent.fyi, US 2026/0354732 A1, claim 1, first published 2026-11-19T02:47:19-05:00, https://patent.fyi/US20260354732/, SHA-256 176e8a6bbc461521fe95ef70de9306eb7b76bb6dd386667685872980c0b98fae