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.
1.1Electrical sensing of cardiac timing
What ifWhat if skin electrodes supplied the cardiac timing signal?
Claim language“a PPG sensor”
Two skin electrodes feed a differential amplifier that measures cardiac electrical activity. A band-pass filter and a QRS detector locate ventricular depolarizations, and the intervals between detections yield heart rate.
1.2Analog perfusion based pulse gating
What ifWhat if analog circuitry performed perfusion based pulse gating?
Claim language“a computer system”
A low-pass circuit tracks the optical baseline, and a peak-to-trough detector measures the pulsatile amplitude. An analog divider forms the amplitude-to-baseline ratio, and a comparator controls an analog switch that passes accepted pulses from a delayed signal path.
1.2Motion reference drives artifact subtraction
What ifWhat if the filter estimated the optical artifact from an accelerometer reference?
Claim language“determines a perfusion index”
A three-axis accelerometer and the optical front end acquire synchronized samples. An adaptive least-mean-squares filter combines delayed acceleration samples to estimate the motion component and subtracts that estimate from the optical waveform.
1.2Pulse weighting from perfusion rank
What ifWhat if the filter continuously weighted pulses by their perfusion ranks within each window?
Claim language“a predetermined filter threshold”
The filter ranks the pulse perfusion indices within each eight-second window and assigns smaller positive weights to higher ranks. It blends each measured pulse with a time-aligned median pulse template using its assigned weight, then joins the resulting segments into a waveform.
1.2Filter outputs accepted beat times
What ifWhat if the filter emitted accepted beat timestamps as its only output?
Claim language“generates a filtered PPG signal”
A peak detector timestamps pulse candidates, and the processor calculates each candidate's pulsatile amplitude divided by its local baseline. The processor emits timestamps for candidates that pass the perfusion threshold, and a rate estimator uses successive accepted timestamps to calculate beats per minute during runs of consecutive accepted beats.
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
Pulse filtering on a training manikin
A reflectance photoplethysmography sensor clips over a translucent tube beneath the surface of a training manikin. A pump drives periodic flow of an optically absorbing fluid through the tube while trainees move or reposition the manikin. A controller divides the root mean square amplitude of the pulsatile signal by the mean optical level over each two-second window and marks windows above a stored perfusion-index threshold as missing in the recorded waveform.
Component substitution
Camera replaces the contact optical probe
A monochrome CMOS camera and a continuous green light source replace the contact optical probe. The camera captures a fixed patch of forehead skin at 60 frames per second with exposure and gain held constant, and a processor averages pixel intensities within the patch for each frame. The processor divides the root mean square amplitude of the band-pass filtered intensity trace by the mean intensity in each two-second window and masks waveform segments whose ratio exceeds a stored threshold.
Parameter or scale variation
Fixed windows with quarter-second updates
A reflectance photoplethysmography sensor samples at 128 Hz, and the processor evaluates a 256-sample window every 32 samples. The processor divides the root mean square amplitude of the 0.5–5 Hz signal in each window by the mean raw signal level from the same window. At each update, it releases the central 32 raw samples if the ratio is at or below the stored threshold and emits 32 missing-sample flags otherwise.
Three optical channels with sample selection
A wrist probe places three photodiodes around one green emitter, at emitter-to-detector separations of 3, 5, and 7 mm, and samples each detector at 100 Hz. Every 500 milliseconds, a controller computes the ratio of pulsatile root mean square amplitude to mean optical level over the preceding two seconds for each channel and compares the ratios with stored channel-specific thresholds. For the next 500 milliseconds, it outputs the baseline-normalized waveform from the passing channel with the largest ratio and marks the output as missing when all three channels exceed their thresholds.
Combination with existing systems
Optical waveform filtering in sleep recorders
A finger photoplethysmography probe connects to a sleep recorder that also acquires respiratory belt and airflow signals on a shared clock. An optical processing module calculates the ratio of pulsatile root mean square amplitude to mean optical level for each one-second interval and compares it with a stored perfusion-index threshold. The recorder stores the accepted optical samples as a waveform track and marks rejected intervals as gaps aligned with the respiratory tracks.
Automated or data-driven variant
Factory thresholds derived from motion recordings
A calibration workstation collects raw optical recordings from representative units during a scripted sequence of probe movements and labels each interval using a synchronized reference pulse trace and the movement log. Software computes the ratio of pulsatile root mean square amplitude to mean optical level for each interval and selects an upper threshold that maximizes the fraction of labeled motion intervals rejected while retaining at least 95 percent of labeled clean intervals. The selected threshold is stored in device firmware, which applies the same interval calculation and masks waveform segments above the threshold.
Specific version of a part
Optical sensor reports beat times
A photodiode and transimpedance amplifier feed an analog band-pass stage and a peak detector inside the sensor. A timer records each detected peak, and the sensor interface transmits only the timestamps.
Reflective green wrist sensor
Two 525 nm light-emitting diodes illuminate the wrist beside a silicon photodiode, with opaque barriers limiting direct optical coupling. The front end subtracts an LED-off ambient sample from each LED-on sample and outputs the resulting raw PPG samples at 100 Hz.
Transmissive infrared finger sensor
A spring clip holds a 940 nm light-emitting diode opposite a silicon photodiode across the fingertip. A transimpedance amplifier converts the transmitted-light photocurrent to a voltage, and an analog-to-digital converter samples it at 200 Hz.
Peak to trough perfusion measurement
The processor measures pulsatile amplitude from each local maximum and its preceding trough, and divides it by the average optical baseline over that beat. It compares the resulting percentage with a stored threshold and marks samples belonging to rejected beats as missing in the filtered waveform.
Resting calibration sets threshold
During a 60-second resting calibration, the processor measures a perfusion index for each detected beat and stores the median plus four median absolute deviations as the threshold. During monitoring, the filter excludes beat segments above the stored value and preserves accepted samples at their original timestamps.
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.
1.1a PPG sensor for generating a raw PPG signal; and
FIG. 1 illustrates a photoplethysmograph intensity signal 100 output by an oximeter sensor.
A · US 8,983,564 B2 · ¶ 6
Although optical signal 210 is attenuated as it travels through the tissue, a portion of the transmitted optical signal 210 will be reflected from pulsating artery 220 back to PPG sensor module 20 as a reflected, amplitude modulated (AM) optical signal 230, where it is received by a detector module comprising one or more photodetectors 100, analog-to-digital converter(s) (ADCs) 110 and optionally filter(s) 120.
B · US 2017/0281027 A1 · ¶ 36
1.2a computer system comprising a motion artifact filter that receives the raw PPG signal, determines a perfusion index for the raw PPG signal, and generates a filtered PPG signal corresponding to the raw PPG signal by filtering the motion artifacts from the raw PPG signal based on the determined perfusion index and a predetermined filter threshold.
In some embodiments, the reflected blue light may be used as a reference signal for reducing or eliminating motion artifacts or other noise components in the reflected green light signal, which is detected by the photodetectors 100.
B · US 2017/0281027 A1 · ¶ 38
An empty cell means only that no passage was reproduced there.
Statements in these documents
Once calculated, the PI 204 may advantageously be displayed in a wide number of ways, including rising LEDs or other display elements, text, graphics, or other visual elements including color, flashing, and the like, trended data, trace data, or the like.
A · US 8,983,564 B2 · ¶ 9
However, using the foregoing scale, the PI bar 324 can be used as a diagnostic tool during low perfusion for the accurate prediction of illness severity, especially in neonates.
A · US 8,983,564 B2 · ¶ 9
Therefore, a need exists for an improved biometric monitoring device that provides accurate heart rate monitoring in a space-constrained wearable device, while reducing power consumption and extending battery life.
B · US 2017/0281027 A1 · ¶ 7
Although described herein as a wearable biometric monitoring device, and more specifically, a wrist-worn biometric monitoring device, the sensor modules described herein could be used in a variety of wearables, including activity-tracking fitness bands and straps, pedometers and smart watches, as well as fitness equipment, bathroom scales and patient monitoring devices.
B · US 2017/0281027 A1 · ¶ 7
Cite aspatent.fyi, US 2026/0354767 A1, claim 1, first published 2026-11-19T03:31:31-05:00, https://patent.fyi/US20260354767/, SHA-256 ab59fba05247022ed5fd669fb2b92818183d2be3b181725f00ee72be7427a7f8