Patent Application
20130150698
1. Field of the Invention
The present invention generally relates to a bio-signal capturing system, and more particularly to an event-based bio-signal capturing system.
2. Description of Related Art
Current bio-signal applications for human beings are mainly focusing in the following areas: (1) bulky systems for formal medical practices, (2) rudimentary gadgets for fitness monitoring, and (3) bulky brainwave-based gadgets for rudimentary applications of brainwave signal. These devices and applications are based largely on bio-signal only. There is no user-friendly and wearable system or product on the market that correlates the bio-signal with other parameters to obtain accurate identification on stimuli of bio-signal for applications require robust and consistent interpretation of bio-signal.
For the foregoing reasons, a need has arisen to propose a novel bio-logical capturing system, for example, to achieve more compact, wearable and suitable for mobile uses like fitness monitoring, sports monitoring and game playing, and also provides a comfortable setting for stationary uses like sleep monitoring and meditation monitoring.
In view of the foregoing, it is an object of the embodiment of the present invention to provide an event-based bio-signal capturing system that leverages both bio-signal of biological beings and event marker from the user or the environment. The embodiment provides accurate identification on stimuli of bio-signal for applications require robust and consistent interpretation of bio-signal.
According to one embodiment, an event-based bio-signal capturing system includes at least one bio-signal capturing device, at least one event capturing device, and a data recording device. The bio-signal capturing device is configured to capture a bio-signal measured from biological beings. The event capturing device is configured to capture an event and generate a corresponding event marker. The data recording device is configured to acquire the bio-signal and the event marker. The bio-signal and the event marker are acquired with corresponding time reference for subsequent event-based data analysis.
The system 100 may further include a data recording device 13 that is configured to acquire the captured bio-signal and event marker from the bio-signal capturing device 11 and the event capturing device 12, respectively. The acquired bio-signal and the event marker may be further stored, with or without data compression, for example, in a storage area 131 associated with the data recording device 13. Moreover, the bio-signal and the event marker are acquired with corresponding time reference for subsequent event-based data analysis. The time reference may, for example, derived from a master clock 132 associated with the data recording device 13.
The acquired bio-signal and the event marker may be subjected to data analysis in the data recording device 13, such as a portable computing device (e.g., a mobile phone) for mobile applications or a personal computer for stationary applications. Alternatively, the data analysis may be performed by a computer 14 that is communicated with the data recording device 13 via a computer network 15 such as the Internet or an intranet.
The bio-signal capturing device 11 mentioned above may capture electrical-activity bio-signals such as Electroencephalography (EEG) signals, Electrocardiogram (ECG) signals or Electromyography (EMG) signals. Alternatively or combinedly, the bio-signal capturing device 11 may capture physiology bio-signals such as skin electrical potential signals, skin conductance (SC) signals, blood flow signals, oxygen content signals or body temperature signals.
As shown in
As mentioned above, there may have a variety of possible bio-signals captured from the bio-signal capturing device 11. Accordingly, the sensing patch 112 may be designed or manufactured as an EEG patch, an ECG patch with attenuator or reduced gain, an EMG patch with attenuator or reduced gain, a skin-conductance patch with active signal source (clock) or power source, a blood flow patch with light emitting diodes (LED) and transducer, an oxygen content patch with LED and transducer, or a general transducer patch with micro controller unit (MCU) and transducers.
The event capturing device 12 mentioned above may capture one or more events in the following (nonexclusive) list: voice recording, sound recording, still imagery, video recording, body/muscle movement or posture, electromagnetic field (EMF) exposure, geographic location, orientation/gesture, finger tapping, altitude, temperature, humidity, and air pressure. The event capturing device 12 may include a rechargeable battery (not shown), and, generally speaking, may be powered by battery, external power, wireless energy source, or energy harvesting mechanism.
The event capturing device 12 mentioned above may be running continuously, or be activated by one or more ways in the following (nonexclusive) list: user, voice, sound, scene change, body/muscle movement or posture, EMF exposure change, geographic location change, orientation change, altitude change, temperature change, humidity change, air pressure change, computer software, and preset conditions.
With respect to the event-based data analysis mentioned above, in one exemplary embodiment, the data recording device 13 performs a bio-signal/event correlation analysis based on the time reference of the captured bio-signal and event marker. In another exemplary embodiment, the data recording device 13 performs multi-dimensional pattern recognition based on a priori characterization and modeling of the captured bio-signal and event marker. A variety of data analyses may be adapted to the event-based data analysis such as sleep analysis, meditation analysis, mood analysis, stress and relaxation analysis, bio-feedback, fitness analysis, attention analysis or interactive game playing.
In order to make the event-based bio-signal capturing system 100 more compact and mobile, some energy saving techniques are deployed in the following exemplary embodiments. The bio-signal capturing device 11 and the event capturing device 12 commonly utilize an analog-to-digital converter (ADC) for converting input continuous physical quantity to a digital number that represents the quantity's amplitude.
In one exemplary embodiment, at least one of the bio-signal capturing device 11 and the event capturing device 12 adopts dynamic gain switching during ADC sampling. Specifically speaking, a high resolution ADC may consume much higher power than a medium resolution ADC. For example, a 16-bit resolution ADC may consume up to 64 times more power than a 10-bit resolution ADC. Since many bio-signals combine occasional large swings (that carry less information content) and mostly smaller swings (that carry more information content), a sampling algorithm using dynamic gain switching (10X for small swing signals and 1X resampling for large swing signals if the 10X gain causes out-of-range condition) using 10-bit resolution ADC can achieve effective 13.5-bit dynamic range that uses only slightly more than the power consumption of 10-bit resolution ADC.
In another exemplary embodiment, at least one of the bio-signal capturing device 11 and the event capturing device 12 adopts non-uniform quantization for ADC sampling. Specifically speaking, by using an algorithm with gain cross-over hysteresis (to reduce the probability of double sampling), the occasional double-sampling power increase is less than 5% of that of a 10-bit ADC, and a power savings of up to 64 times than that when a 16-bit ADC is used.
In a further exemplary embodiment, at least one of the bio-signal capturing device 11 and the event capturing device 12 adopts data scrambling to prevent consecutive sampling data points being closest neighbors during communication; and the data recording device 13 adopts spike removal within a corrupted data packet to reduce noise energy. Specifically speaking, for communications in a noisy environment, there may be frequent interferences from many RF spike/bursting sources (such as WiFi, cell-phone, Blue-Tooth, WiMAX devices) as well as other communication noise sources. Traditional communication techniques rely on Cyclic Redundancy Check (CRC) or other error-detection coding and detection methods to detect whether a data packet is corrupted. The corrupted data packets will either be lost or require energy-consuming communication handshake and re-transmissions. Since bio-signal sampling rate is generally much higher than the frequency contents of bio-signals to minimize sampling noise and Nyquist aliasing, the present exemplary embodiment thus implements algorithm with data scrambling within data packet (that is, the captured bio-signal sampled data points are rearranged within data packet, so that no two consecutive sampled data points are nearest neighbors to each other, and thus resulting in very few consecutive sampled data point corruptions), plus spike removal algorithm to repair corrupted data packet (that is, to detect which sampled data point is corrupted and to repair it with minimum noise energy). This algorithm enables high-quality signal reception in a noisy environment.
In a further exemplary embodiment, at least one of the bio-signal capturing device 11 and the event capturing device 12 adopts collision detection, and adopts skip-forward algorithm upon collision detection. Specifically speaking, in order to conserve energy consumption and to minimize interferences from multiple bio-signal and event capturing devices 11/12, the present exemplary embodiment thus implements collision detection and time hopping communication algorithm. To conserve energy, each capturing device 11/12 will only turn on momentarily (less than 10% for ADCs) to capture the intended signal or event, while staying in low power mode most of the time. Similarly, the communication or storage circuits will only turn on momentarily (less than 1% for communication circuit and less than 5% for storage circuit) to transmit or store the captured signals and events, while staying in low power mode most of the time. Because all capturing devices 11/12 operate with their own fixed routines independently (that is, asynchronously), and with their time bases not perfectly matched (some faster and some slower), all these capturing devices 11/12 will gradually run into each other (systematically) causing collisions in communications and possible data packet corruptions. In the present embodiment, each capturing device 11/12, prior to start of communication transmission phase, will first check if there is any other device transmitting (possible collision) without receiving any data (that is, just to detect carrier signal that may corrupt its transmission). As any device 11/12 detecting such carrier is most likely the faster device (for it's chasing up from behind), it will skip one or more sampling data points (therefore “hasten up” or skip-forward) for its next data packet transmission without causing any actual collision nor losing any data packet. For devices with <1% communication circuit duty cycle, this algorithm allows more than 50 devices per data channel operating simultaneously (and asynchronously) without having collision or data corruptions.
Although specific embodiments have been illustrated and described, it will be appreciated by those skilled in the art that various modifications may be made without departing from the scope of the present invention, which is intended to be limited solely by the appended claims.
| Number | Date | Country | |
|---|---|---|---|
| 61568255 | Dec 2011 | US |
