阅读说明：本技术 信号数据提取方法及设备 (Signal data extraction method and device ) 是由 劳里·A·科克斯 于 2013-07-22 设计创作，主要内容包括：一种用于提取重要信号信息的方法和系统被公开。该方法检查一组从患者神经系统测试获得的信号,并且找出关注的信号区域。一旦找出所有具有关注区域的信号组集,系统得出关注区域被定位且有效的结论。拒绝不在组集内的信号,并且在组集内的信号被信号平均以生成信号平均波形。信号平均波形代表测试的结果。(A method and system for extracting important signal information is disclosed. The method examines a set of signals obtained from a patient's nervous system test and finds a signal region of interest. Once all the sets of signal groups with regions of interest are found, the system concludes that the regions of interest are located and valid. Signals that are not within the group set are rejected and the signals within the group set are averaged by the signals to generate a signal average waveform. The signal average waveform represents the results of the test.)

1. A method of constructing a waveform, the waveform representing a test result, the method comprising the steps of:

acquiring a plurality of signals;

storing the plurality of signals;

analyzing the plurality of signals to identify regions of interest of the signals;

generating a histogram corresponding to the signal having the region of interest and selecting a set of signal groups having the region of interest from consecutive bits in the histogram;

calculating a percentage value of the set of signal groups, the percentage value corresponding to a percentage of signals having the region of interest to the total of the plurality of signals;

determining whether the percentage value is greater than a predetermined percentage value;

concluding that the region of interest is valid if the percentage value is greater than a predetermined percentage value; and is

Generating a signal average waveform from the signals in the set of signals.

2. The method of claim 1, wherein the plurality of signals are obtained by one of a VEP test or a PERG test.

3. The method of claim 1, wherein the region of interest is a peak in the signal.

4. The method of claim 1, wherein the region of interest is a portion of a waveform found in a signal within a predetermined time frame.

5. The method of claim 2, wherein the region of interest is a P100 latency.

6. The method of claim 2, wherein the region of interest is the P50 peak.

7. The method of claim 1, wherein the step of analyzing the plurality of signals to identify regions of interest of the signals further comprises the step of signal averaging at least two signals to form a signal averaged waveform and detecting regions of interest of the signals on the signal averaged waveform.

8. The method of claim 1, wherein the predetermined percentage value is in the range of 30-50%.

9. A system for constructing a waveform representative of a test result, comprising:

a stimulus display for displaying a stimulus to an eye of a patient;

a sensor for detecting a signal generated in response to the stimulus;

a computer having a data acquisition system for acquiring the signal;

a digital memory for storing the signal;

the computer further has:

a system for analyzing a plurality of stored signals for detecting a region of interest on each of the stored signals;

a system for determining that a set of individual signals having the region of interest has been collected;

a system for calculating a percentage value of the set, the percentage value corresponding to a percentage of the signals having the region of interest to the total of the plurality of signals;

a system for judging whether the percentage value is larger than a preset percentage value, if the percentage value is larger than the preset percentage value, the concerned area is valid; and

a system for generating a signal-averaged waveform from the signals in the set.

10. The system of claim 9, wherein the sensor is an electrode attached to the scalp of the patient.

11. The system of claim 9, wherein the sensor is an electrode attached to a patient's sub-ocular region.

12. The system of claim 9, wherein the computer is configured to ignore signals not in the set.

13. The system of claim 9, wherein the region of interest is a signal portion of interest located within a predetermined time frame.

Technical Field

The present invention relates to the field of signal data extraction, and more particularly to an improved apparatus and method for extracting signal information from a plurality of signals obtained by visual path testing.

Background

Early disease monitoring is of great importance in improving the success of treatment. However, once a patient develops symptoms of a disease, typically, the disease has worsened or developed to some extent. In order to actually find potential problems at an earlier stage, asymptomatic patients should be examined for early signs of disease.

Glaucoma and other eye and nerve related diseases are all diseases that may be detected at an earlier stage. In the art, there are a variety of tests that are capable of detecting the early onset of glaucoma and other neurological diseases.

Graphic VEPs are well-known diagnostic aids in the detection of glaucoma and other neurologically related diseases. In this test, the patient observed a reversed pattern displayed on the screen and in-field sensory stimulation caused by Visual Evoked Potentials (VEPs). Electrodes were placed on the scalp of the subjects to detect VEPs.

In the diagnosis and/or examination of patients using the VEP technique, the clinician views the N75P100N135 complex set in a composite waveform map. The N75P100N135 complex conveys important information about the state of the nerve cells conducting the nerve impulses.

The graphic electroretinogram (PERG) is a diagnostic tool for ocular diseases similar to VEP. However, unlike the use of VEP, which derives its signal from the visual cortex from behind the skull, with PERG, the signal is collected directly from the retina by placing a transducer under the patient's eye. Using the PERG test, diagnostically significant waveform profiles were shown with the N35-P50-N95 complex.

Typically, a signal is generated whenever the pattern displayed on the screen is flipped. After a certain number of signals have been acquired, the software program signal averages a number of acquired signals, as the case may be, and generates a waveform representation of the N75P100N135 complex or the N35-P50-N95 complex.

One problem with this method of generating a signal-averaged signal is that signals that are not associated with the N75P100N135 complex or the N35-P50-N95 complex may be included in the signal-averaged waveform map. For example, signal oscillations caused by patient blinks or movement may be erroneously included in the signal-averaged waveform map. As a result, the sensitivity and specificity of the VEP or PERG test may be reduced or impaired.

Disclosure of Invention

The present invention aims to provide a new method and apparatus for improving the accuracy of a signal-averaged waveform map derived from a plurality of signals. The system identifies a time frame within which a region of interest is expected to appear in the signal (a "region of interest" being a peak or trough). The system then examines the plurality of signals to determine whether the current region of interest is in each of the plurality of signals. A region of interest is valid once it is detected in each of the plurality of signals within the desired time frame. For example, if the region of interest is a peak and the desired time frame is between 90 and 120 milliseconds, the system may determine that the region of interest is located and valid when a predetermined number of peaks (e.g., 5 peaks) or a predetermined percentage of peaks are detected within the 90 to 120 millisecond time frame. Once this determination is made, peaks that occur outside of a given timeframe (i.e., earlier than 90 milliseconds or later than 120 milliseconds) will be ignored.

The system then identifies signals that are supported by the waveform of interest and signals that are not supported by the waveform of interest. Signals that are not supported by the waveform of interest are not included in the averaged signal test results.

Drawings

FIG. 1 is a schematic diagram of a system architecture according to an embodiment of the present invention.

Fig. 2 shows a low contrast checkerboard pattern according to an embodiment of the present invention.

Fig. 3 shows a high contrast checkerboard pattern according to an embodiment of the present invention.

Fig. 4 shows a readout of two signal average waveforms obtained from the left and right eyes of a patient, respectively, according to an embodiment of the present invention.

Figure 5 illustrates a readout of the signal average waveform obtained from a patient during a PERG test according to an embodiment of the present invention.

Detailed Description

Embodiments of the present invention will be described below with reference to the above-described drawings. However, the drawings and their description of the present invention are not intended to limit the scope of the present invention. It will be appreciated that various modifications can be made without departing from the spirit of the invention. Moreover, features described herein may be omitted, additional features may be included, and/or features described herein may be combined in ways other than the specific combinations set forth herein, without departing from the spirit of the invention.

Referring to FIG. 1, a system of hardware components is shown, according to an embodiment of the present invention. As shown, a display device 102, or stimulus display, is provided to display a pattern to a viewer 104. Sensors 10, 12 and 14, such as disposable electrodes, non-invasively attached to the patient's scalp or under the patient's eye detect responses to the stimuli produced by the optic nerve or retina, respectively. The response is amplified, digitized, recorded and analyzed by the data acquisition component 107. Conductors 116, 118 and 120 collect information from sensors 10, 14 and 12, respectively, which sensors 10, 14 and 12 are located on the scalp around the visual cortex, frontal cortex and parietal cortex of patient 104, respectively. Conductors 116, 118 and 120 are connected to a VEP recording measurement device. Conductor 121 is positioned under the patient's eye and connected to the PERG measurement device.

A VEP or PERG recording and measuring device (or data acquisition component) 107 acquires the brain or retina response to the stimulus. The data acquisition component includes an amplifier 106 that receives signals from sensors 116, 118, 120, and/or 121 and amplifies the VEP/PERG signals. The amplified signal is then provided to an analog-to-digital converter 108 for converting the analog VEP/PERG signal to digital form. The data acquisition component is connected to the central processor 110 of the computer for controlling the operation and function of the VEP/PERG recording and measuring device. The computer has and/or is coupled to a digital storage medium for storing signal data and other desired information. The CPU is connected to a Graphical User Interface (GUI) or display 112 that displays the data acquired by the VEP/PERG recording and measuring device and displays information about the operation of the test performed by the operator. A keyboard 114 connected to CPU110 allows an operator to input information to the computer relating to the item to be tested. While a printer 124 connected to the CPU allows the test results to be printed out. A visual stimulus generating device 122, such as a graphics card, is also connected to and controlled by the CPU. The stimulus generating device 122 generates a pattern on the display 102 that is perceived by the patient. Stimulus generating device 102 connects to data acquisition component 107 via a synccard (Sync card) 126. The synchronization card 126 synchronizes the periodic visual stimulus to the sample rate at which the VEP/PERG signal response is recorded. Software running on the computer is programmed to perform the process steps described herein.

Fig. 2 and 3 show two examples of patterns displayed to a patient for stimulating the retina and the visual pathway. Fig. 2 shows a low contrast checkerboard pattern, while fig. 3 shows a high contrast pattern. In either case, "pattern flipping" is performed by reversing the light and dark blocks on the checkerboard. It will be appreciated by those skilled in the art that other patterns are possible and within the scope of the invention.

In the diagnosis and or examination of patients using the VEP technique, the N75-P100-N135 complex group delivers important information about the status of the ganglion cells of the retina and the optic nerve that conducts nerve impulses. The N75-P100-N135 complex, which is well known in the VEP art, roughly corresponds to the depolarization and repolarization phases of the action potential. The latency or reduced amplitude of the N75-P100-N135 response may indicate, for example, a failure caused by hypomyelination, such as a nerve damage that loses conductance. Thus, the N75-P100-N135 complex was the focus of the VEP test. The signal portions not associated with the N75-P100-N135 complex were mostly not of diagnostic interest.

As noted, to include only signals related to the N75-P100-N135 complex, the software is designed to ignore effects and signals that are not related to the N75-P100-N135 complex. In one embodiment, the software is designed to analyze two or more time frames simultaneously and the signal averages them. The automatic cursor position program locates the P100 peak for each signal averaged paired time frame. P100 was evaluated using a histogram. A set of groups found within the range of the desired P100 is identified. If the group set contains a sufficient percentage of frames from the entire test, P100 is identified. Frames outside the set are not included in the signal average test results. Since signals unrelated to P100 are not included in the final signal average waveform map, the VEP response is rendered clearer and more accurate.

For example, given a pattern of a particular contrast, the P100 peak is expected to occur in approximately 93 to 118 milliseconds. In this test environment, the software is designed to generate a histogram or similar plot to identify the set of peaks that occur within 93 to 118 milliseconds. If a predetermined number of peaks are identified at the expected time, the software concludes that the P100 peak is located and valid. Any peaks outside the expected time frame are rejected and are not included in the signal-averaged waveform map. In one embodiment of the invention, the expected time frame for the P100 peak is in the range of 80 to 140 milliseconds. It will be appreciated by those skilled in the art that any of a variety of different scopes are possible in different embodiments of the invention, all of which are within the teachings of the invention.

In one embodiment of the present invention, a set of signal groups located in two or three (or more) consecutive bits is used to identify the N75-P100 complex group. Any peaks that are located before or after a particular three (or more) bits are identified as outliers and are therefore not included in the signal average profile. It is understood that the number of sequential bits can vary, and can be more or less than three in different embodiments of the invention.

If an insufficient number of time frames are found in the set, P100 is not identified. This is important because neural noise alone can tend to create peaks that can be mistaken for a P100 response. The result is a more stable P100 response, since repeated testing is not affected by random frames intended to move the actual P100 more or less.

It is to be understood that in various embodiments of the present invention, the latency of the response may be obtained in any of a variety of ways. For example, rather than examining a single frame to detect P100 latency, the system instead signal averages two or more response frames (results of different stimuli each), and it includes the latency of the signal-averaged waveform as one frame in bits with a corresponding range of latencies. More specifically, assuming that a signal is generated each time the pattern flips, the system does not analyze each generated signal to detect the P100 peak, instead, the system generates a composite signal average waveform and it analyzes the signal average waveform to determine whether the P100 peak is present. When a set of signal-averaged P100 peaks is identified, the system concludes that P100 has been located. It will be appreciated by those skilled in the art that for the P100 peak, a single frame may be analyzed, and it is not necessary to analyze a combination of two or more frames. It is further understood that different embodiments of the present invention may analyze combinations of more than two signals.

As mentioned, the system identifies a set of P100 peaks to conclude that P100 is located and valid. The set of groups may be established in any of a variety of ways. In one embodiment of the invention, the system determines the ratio of valid to invalid signals and uses the ratio to determine whether a sufficient number of valid signals are obtained. If a sufficiently large percentage of the response signals are valid, a group set is established. For example, if the waveform of interest contains a P100 peak between 80 and 140 milliseconds of a time frame, the signal having that peak at a given time frame is a "valid signal". Peaks that fall outside the 80 to 140 millisecond timeframe are referred to as "nulls". After all the signals from the entire test are acquired and stored, the system determines what the percentage of valid signals is. If the percentage is above a predetermined percentage value, a set of P100 peaks is established and P100 is located and valid. If the percentage of valid signals is below a predetermined percentage, no group set is established and P100 is invalid. In one embodiment of the invention the predetermined percentage value ranges from 30-50%, but different ranges and different percentage values may be used in different embodiments of the invention.

The software of the present invention is programmed to run a number of primary functions for each test. All signals originating from the neural stimulation are stored on a digital storage medium. The software then analyzes the stored signal data to determine whether the resulting set of signals is associated with a particular waveform (e.g., P100 latency). The software then takes two independent signals and averages their signals into a signal average waveform. The signal averaged waveform is then analyzed to detect the P100 peak within a given time frame. The software continues the procedure until all or a predetermined number of the acquired signals are analyzed to identify P100 (or other areas of interest). The software then calculates the number of signals of P100 identified from the total number of signals obtained from the given test. The software then determines whether the percentage of the signal identifying P100 is above a predetermined percentage value. If the number is above a predetermined percentage value, the software concludes that the signal of P100 or other field of interest is localized and valid. The software then signal averages all the signals that P100 was detected within a given time frame to generate a final signal average waveform representing the test results. Any outlier signals are ignored and are not incorporated into the final signal average waveform.

As a result, the system of the present invention identifies and removes frames that are not associated with the VEP N75P100 complex group from the signal averaging results. The response results have several benefits: good responses, whether with the desired P100 latency or with delayed P100 latency, tend to be low noise (making P100 more recognizable), and P100 latency tends to be more stable in repeated tests. Poor responses (whether caused by pathology or patient inattention or poor patient connection) can be identified as such so as to not calculate a wrong P100 latency.

Fig. 4 shows two signal average waveforms obtained from the right and left eyes of a patient, respectively. Showing that the P100 peak occurs approximately between 85 and 120 milliseconds (time frame B). In this example, the set of peaks that appear in time frame B is identified and included into the signal-averaged waveform. Once a critical number of signals are detected within time frame B, P100 is established. Any peaks that appear outside of time frame B (e.g., time frames a and C) are ignored and not included in the final resulting signal average waveform.

Graphic electroretinograms (PERG) are another diagnostic modality for the detection of glaucoma. PERG is similar to VEP in that the signal is generated by a stimulus caused by the patient observing a flipping pattern on the screen. However, this signal is acquired by a specific sensor located under the patient's eye. In the PERG test, the important waveforms for diagnosis have the N35-P50-N95 complex. The N35-P50-N95 complex is similar to the expected waveform of VEP, except that it is generally smaller in amplitude and shifted to the left in time. Because the PERG sensor receives stimulation directly from the retina (VEP acquires signals behind the skull as opposed to VEP), the signals are acquired earlier in time than the VEP, and the result is that the N35-P50-N95 complex is shifted left. Thus, when the peak of the VEP signal is at or about 100 milliseconds, the peak of the PERG signal is about 50 milliseconds.

In an embodiment of the present invention, the system is programmed to identify the P50 peak in the PERG signal. In a similar approach to the VEP approach described, time frames are identified during which P50 is expected to occur. Once a predetermined number of peaks are detected in the desired time frame, the software concludes that P50 is located and valid. For example, in one embodiment of the invention, the software is set up to identify the peak that comes between 40 and 60 milliseconds as the P50 peak. Once the set of peaks is detected at that time frame, the P50 peak is located. Any peaks outside of the time frame (e.g., earlier than 40 or later than 60 milliseconds) are ignored and are not included in the final signal average waveform.

Fig. 5 shows an average waveform of signals obtained from a patient during a PERG test. The P50 peak is shown to occur approximately between 48 and 52 milliseconds (time frame E). In this example, the set of groups of peaks that appear in time frame E are identified and included into the signal-averaged waveform. Once a critical number of signals are detected within time frame E, the P50 peak is established and valid. Any peaks that appear outside of time frame B (e.g., time frames D and F) are ignored and are not included in the signal average waveform representing the test results.

It will be appreciated that the methods and systems described herein may be used with multiple signals obtained in any of a variety of different test environments, and VEP and PERG are merely examples.

While the invention has been described in terms of specific embodiments, it is to be understood that the description is not intended to be limiting, since further changes and modifications may be apparent to, or may be construed by, persons skilled in the art. It is intended that the present application cover all such modifications and variations.