Abstract: A method of extracting a representative waveform of a bio-signal includes receiving an input of the bio-signal; dividing the bio-signal into a plurality of sub-signals; selecting at least one sub-signal for extracting a representative waveform from among the divided sub-signals; extracting a representative waveform by using the at least one selected sub-signal; evaluating a quality of the extracted representative waveform; and based on the representative waveform satisfying a predetermined quality criterion corresponding to the evaluation, determining that the representative waveform is a final representative waveform.

Abstract: A method and device with image feature generation are provided. The method includes extracting a reference feature of a reference frame image, among a plurality of frame images, using a feature extraction model, generating partial compression information corresponding to a portion of a target frame image, where the portion of the target frame has a determined difference with a corresponding portion of the reference frame image that is greater than or equal to a predetermined threshold, generating a compression feature from the generated partial compression information using a feature derivation model, and generating, based on the reference feature and the compression feature, a target feature of the target frame image.

Abstract: A method of determining a final path of a moving object includes acquiring pieces of sensor data, determining a first path of the moving object based on the pieces of sensor data, inputting at least one piece of sensor data among the pieces of sensor data into an encoder that encodes the at least one piece of sensor data, inputting the encoded at least one piece of sensor data into a generative neural network model that generates guide information on a path of the moving object, and determining the final path of the moving object, based on the first path and the guide information.

Abstract: An apparatus for estimating blood pressure may include a sensor configured to measure a bio-signal from an object; and a processor configured to: obtain a cardiac output (CO) feature based on the bio-signal; obtain a total peripheral resistance (TPR) feature by combining an amplitude of a propagation wave component and amplitudes of at least two reflection wave components of the bio-signal, in response to a variation of the CO feature at a blood pressure measurement time relative to a calibration time, being within a predetermined range; obtain the TPR feature by combining the amplitude of the propagation wave component and an amplitude of one reflection wave component of the at least two reflection wave components, in response to the variation of the CO feature not being within the predetermined range; and estimate blood pressure based on the CO feature and the TPR feature.

Abstract: A method for predicting failure of bio-information estimation is provided. The method for predicting failure of bio-information estimation includes: receiving a bio-signal; obtaining failure prediction indicators from the bio-signal that is received until a current time, the failure prediction indicators comprising at least one of a current number of pulses of the bio-signal until the current time, a current signal quality of the bio-signal until the current time, a maximum number of pulses of the bio-signal that are predicted to be measured until a time limit, and a maximum signal quality of the bio-signal that is predicted to be measured until the time limit; and predicting whether bio-information estimation will fail based on the failure prediction indicators before the time limit is elapsed.

Abstract: An apparatus for non-invasively estimating bio-information is provided. According to one exemplary embodiment, the apparatus may include a bio-signal acquirer configured to acquire a bio-signal; and a processor configured to extract a plurality of characteristic points from the bio-signal, determine internally dividing points of the plurality of characteristic points, and extract feature values from the bio-signal based on the internally dividing points to perform bio-information estimation.

Abstract: A blood pressure monitor may include a photoplethysmogram (PPG) sensor to obtain a PPG sensor and at least one processor configured to extract characteristic points from the PPG signal, determine internally dividing points of the characteristic points based on time values of the characteristic points, extract feature values from the PPG signal based on the internally dividing points, and determine a blood pressure based on the extracted feature values of the PPG signal. The blood pressure monitor may further include a display configured to provide a graphical user interface that displays a vertical line that is movable in a timeline over a graph that shows a change in the blood pressure over time, a blood pressure value corresponding to a position of the vertical line, and component pulses constituting the PPG signal, together with the graph and the vertical line.

Abstract: An apparatus for estimating blood pressure may include: a photoplethysmogram (PPG) sensor configured to measure a PPG signal from a user; and a processor configured to: extract a cardiovascular feature from the PPG signal; calculate a first variation in the extracted cardiovascular feature compared to a reference cardiovascular feature measured at a reference time; determine a measurement posture of the PPG signal based on a time interval between two waveform components of the PPG signal; in response to the measurement posture corresponding to the reference posture, estimate the blood pressure based on the first variation; and in response to the measurement posture not corresponding to the reference posture, obtain a second variation by correcting the first variation; and estimate blood pressure based on the second variation.

Abstract: An apparatus for non-invasively estimating blood pressure is provided. The apparatus for estimating blood pressure may include a bio-signal measurer configured to measure a bio-signal from a user and a processor configured to estimate blood pressure using the measured bio-signal. The processor may extract a first feature and a second feature from the bio-signal at an extraction time, estimate changes in the first feature and the second feature which have occurred during a time period from a calibration time at which the first feature and the second feature are calibrated to the extraction time at which the first feature and the second feature are extracted, and estimate a blood pressure based on the changes in the first feature and the second feature.

Abstract: An apparatus for estimating blood pressure may include: a memory storing one or more instructions; and a processor configured to execute the one or more instructions to: extract a cardiac output (CO) feature, a first candidate total peripheral resistance (TPR) feature, and a second candidate TPR feature from a bio-signal; determine one of the first candidate TPR feature and the second candidate TPR feature as a TPR feature based on a direction of change in the CO feature and a direction of change in the first candidate TPR feature between a blood pressure measurement time and a calibration time; and estimate the blood pressure based on the TPR feature and the CO feature.

Abstract: A method of determining regularity of a bio-signal is provided. The method of determining regularity of a bio-signal according may include acquiring a plurality of pulse waveforms of the bio-signal, acquiring a plurality of slope waveforms corresponding to the plurality of pulse waveforms, binarizing the plurality of slope waveforms, acquiring synchronization information of the plurality of pulse waveforms based on binarizing the plurality of pulse waveforms; acquiring a synchronization rate of a reference interval based on the synchronization information, and determining whether the bio-signal is regular or irregular based on the synchronization rate of the reference interval.

Abstract: An adaptive bio-signal feature combining apparatus includes: a feature extractor configured to extract first feature values and second feature values from a bio-signal of an object; a stable interval determiner configured to determine at least one stable interval in the bio-signal; a statistical variable calculator configured to calculate a statistical variable value of a first feature and a statistical variable value of a second feature for each of the at least one stable interval based on the first and second feature values extracted from the at least one stable interval; and a feature combiner configured to calculate an integrated combining coefficient that is used to combine the first feature and the second feature, based on the statistical variable value of the first feature and the statistical variable value of the second feature.

Abstract: An apparatus for estimating bio-information is provided. The apparatus for estimating bio-information by using an ultrasonic signal according to an embodiment of the present disclosure includes: an ultrasonic sensor configured to acquire an ultrasonic signal from an object; and a processor configured to detect peaks from the acquired ultrasonic signal, and to determine false detection of a peak, among the detected peaks, by using at least one of amplitudes of the detected peaks or a left waveform shape and a right waveform shape of the peak.

Abstract: A method of obtaining a feature for blood pressure estimation is provided. The method includes extracting a reference feature from a bio-signal; obtaining a feature for blood pressure estimation by: obtaining, as the feature for blood pressure estimation, a first candidate feature based on the reference feature being less than a first threshold; obtaining, as the feature for blood pressure estimation, a second candidate feature based on the reference feature being greater than or equal to a second threshold that is greater than the first threshold; or obtaining, as the feature for blood pressure estimation, a combination of the first candidate feature and the second candidate feature based on the reference feature being less than the second threshold; and estimating a blood pressure by using the obtained feature.

Abstract: An apparatus for non-invasively estimating bio-information may include a bio-signal acquirer configured to acquire a bio-signal; and a processor configured to extract a plurality of characteristic points from the bio-signal, determine internally dividing points of the plurality of characteristic points, and extract feature values from the bio-signal based on the internally dividing points to perform bio-information estimation. The plurality of characteristic points may include a first time point and a second time point when the pulse wave signal includes a first component pulse wave that has a maximum amplitude at the first time point, and a second component pulse wave that has a maximum amplitude at the second time point.

Abstract: An apparatus for estimating blood pressure is provided. The apparatus for estimating blood pressure according to an example embodiment includes: a sensor configured to measure a photoplethysmogram (PPG) signal from an object; and a processor configured to detect a position of a dicrotic notch based on a local maximum point and a local minimum point of a second-order derivative signal of the PPG signal, to obtain a first value and a second value from a first section and a second section of the PPG signal divided based on the dicrotic notch of the PPG signal, to obtain a pulse pressure feature value based on the first value and the second value, and to estimate blood pressure based on a variation in the obtained pulse pressure feature value compared to a reference pulse pressure feature value.

Abstract: An apparatus for estimating bio-information of a user may include a first sensor configured to measure a first signal from the user; a second sensor configured to measure a second signal from the user; and a processor configured to obtain a first characteristic point from the first signal; obtain a second characteristic point from the second signal based on the first characteristic point and a pre-defined probability distribution function; and estimate the bio-information of the user based on the second characteristic point.

Abstract: An apparatus for generating a blood pressure estimation model includes: a signal acquirer configured to receive input a first signal and a second signal from a user; and a processor configured to obtain a pulse transit time (PTT) as a first predictor variable value based on the first signal and the second signal, to extract at least one feature from the second signal, to obtain a second predictor variable value based on the extracted at least one feature, and to generate a blood pressure estimation model based on the first predictor variable value and the second predictor variable value.

Abstract: Provided is an apparatus for estimating bio-information. The apparatus for estimating bio-information according to an embodiment includes a processor configured to obtain a pulse wave signal from a user; extract components of a plurality of element waveforms which constitute a waveform of the pulse wave signal; and obtain a cardiovascular feature based on the extracted components of the plurality of element waveforms. The processor is configured to extract a component of at least one element waveform based on a component of an adjacent element waveform.

Abstract: An apparatus for estimating blood pressure is provided. According to one embodiment, the apparatus for estimating blood pressure may include a sensor configured to measure a bio-signal from a subject, and a processor configured to acquire, form the bio-signal, a first feature value associated with cardiac output (CO) and a second feature value associated with total peripheral resistance (TPR), determine whether a current mechanism of blood pressure change of the subject is a post-exercise hypotension mechanism based on the first feature value and the second feature value, and estimate blood pressure according to determination of whether the current mechanism is the post-exercise hypotension mechanism.