High-resolution portable near-infrared brain function imaging system

Abstract

The invention discloses a high-resolution portable near-infrared brain function imaging system. The system comprises a wearable near-infrared device and an upper computer, a controller module, a light source module and a detector module are arranged on the wearable near-infrared device; after a subject wears the wearable near-infrared device, the controller module drives the light source module, LED light sources in the light source module are sequentially lightened, a first light signal is transmitted to a brain region of the subject, a second light signal generated after the first light signal is subjected to diffuse reflection by the brain region is received by the detector module and converted into a current signal, and the current signal is transmitted to the light source module. After the signals are processed, the signals are converted into multichannel original light intensity data, the multichannel original light intensity data are sent to the upper computer through the controller module, the upper computer quantitatively converts the multichannel original light intensity data into blood oxygen concentration, a real-time high-resolution brain blood oxygen change model is constructed, and brain function imaging is achieved. A complete brain function signal processing flow is established, the stability is high, and the method is suitable for complex task scenes.

Description

Technical Field

[0001] The invention relates to the field of functional near-infrared spectral imaging, and in particular to a high-resolution portable near-infrared brain functional imaging system. Background Art

[0002] As an emerging imaging technology, functional near infrared spectroscopy (fNIRS) has the following advantages: the equipment is light and flexible, and can be used in places such as schools or hospitals; the signal is less affected by head movement, and the subjects can even complete the experiment under conditions such as running; it can simultaneously provide information on oxygenated hemoglobin, deoxygenated hemoglobin, and overall hemoglobin concentrations, reflecting changes in brain function activities from multiple angles; it is a non-invasive technology, so repeated measurements can be performed. In addition, many research results in the past decade have fully demonstrated the broad application prospects of fNIRS in basic brain science research. A high-performance fNIRS system can assist in the diagnosis of patients' mental illnesses, detect patients' brain activities in real time at the bedside, and help researchers explore the mysteries of the human brain, etc.

[0003] Continuous Wave functional Near Infrared Spectroscopy (CW-fNIRS) technology is widely used in the field of cognitive ability monitoring as a non-invasive neuroimaging method. The hardware complexity of the common CW-fNIRS brain blood oxygen detection system increases with the improvement of the temporal resolution of the blood oxygen signal. The process of building a multimodal physiological signal acquisition system is cumbersome and it is difficult to achieve clock source synchronization. Moreover, due to the limitations of bulky equipment, high prices and uncomfortable wearing experience, traditional fNIRS monitoring systems still cannot achieve continuous and long-term monitoring.

[0004] In the prior art, an fNIRS near-infrared brain function imaging device and its adjustment method, and a brain imaging system (CN118000721A) are disclosed, which relate to the field of brain imaging technology. The device described in the patent is worn on the forehead, and relies on a pusher (airbag) to achieve the angle and position movement of the light source transmitter and the light source receiver, and a high degree of fit of the device is achieved through the above modules. The structure is complex, resulting in a bulky device and low wearing comfort, making it difficult to achieve continuous monitoring in natural scenarios. Because the patented device or other devices using optical fibers are too thick (the thickness of the device is too high, so that transcranial magnetic stimulation cannot penetrate into the brain area to be regulated), it is impossible to combine the transcranial magnetic stimulation device to achieve closed-loop control of brain stimulation and brain function detection simultaneously.

[0005] The existing technology is based on the arrangement of the transmitting probe and the receiving probe at an optimal distance of 3 cm. The information detected at this distance includes the change information of deep brain function signals and the noise of shallow global oscillation interference. Therefore, there is a defect that the resolution is low and the brain function signals are buried in the noise and difficult to be accurately extracted.

[0006] Near-infrared brain functional imaging technology is based on the light source module emitting light signals to the brain area, which are then received and collected by the detector module after being scattered and absorbed by the brain area. The light signals are easily interfered by sunlight and light in natural situations. The existing technology uses a probe cap to block and cover the light, or uses a receiving optical fiber for transmission. This method cannot achieve miniaturization of the equipment, and it is difficult for the subjects to carry it around for exercise. Summary of the invention

[0007] The purpose of the present invention is to provide a high-resolution, high-reliability, portable near-infrared brain function imaging system. The invention studies and designs a continuous wave near-infrared brain imaging system that supports up to 32-channel light sources and 32-channel detectors. The near-infrared brain function imaging system consists of a light source module, a detector module, a wireless WIFI module, a controller module and a host computer. The controller is controlled by a platform based on STM32H7 to achieve real-time data acquisition capabilities for 32*32 channels of the brain. While achieving miniaturized and convenient design, it provides higher spatial and temporal resolution. The present invention realizes multi-spacing channel signal acquisition based on multi-level light intensity excitation, realizes brain function signal extraction through spline interpolation and wavelet filtering fusion optimization algorithm, effectively improves signal quality and channel quantity, and realizes continuous monitoring and precise extraction of brain function signals in natural scenarios.

[0008] The purpose of the present invention is achieved by at least one of the following technical solutions.

[0009] A high-resolution portable near-infrared brain function imaging system comprises a wearable near-infrared device and a host computer; the wearable near-infrared device is provided with a controller module, a wireless WIFI module, a light source module and a detector module; After the subject wears the wearable near-infrared device, the controller module drives the light source module, and the LED light sources in the light source module are lit in turn to emit a first light signal to the subject's brain area. The second light signal generated by the diffuse reflection of the first light signal in the brain area is received by the photodiode in the detector module and converted into a current signal. After signal processing, the analog-to-digital converter module (ADC) converts the signal acquisition into multi-channel raw light intensity data and transmits it to the controller module. The controller module sends the multi-channel raw light intensity data to the host computer through the wireless WIFI module. The host computer quantitatively converts the multi-channel raw light intensity data into blood oxygen concentration through the modified Lambert-Beer law, constructs a real-time high-resolution brain blood oxygen change model, and realizes brain function imaging.

[0010] Furthermore, the light source module includes a multi-channel digital-to-analog converter (DAC), a plurality of light source driving circuits and a plurality of LED light sources; the number of the light source driving circuits and the LED light sources are equal and correspond one to one; each channel in the multi-channel DAC controls a light source driving circuit to drive the corresponding LED light source respectively; The controller module sends the digital signal to the multi-channel digital-to-analog converter to convert it into an analog voltage signal, and then the light source driving circuit converts the analog voltage signal into a current signal, thereby driving the LED light source to work under a constant current condition. The multi-channel digital-to-analog converter is used to realize the independent adjustment of the light intensity of the light source, and realizes adaptive LED light source adjustment for personalized human body. In the light source driving circuit, the collector of the transistor is connected to the cathode of the LED light source, the anode of the LED light source is connected to the driving voltage, the analog voltage signal output by the multi-channel digital-to-analog converter is input to the non-inverting input terminal of the operational amplifier, and the operational amplifier is used to realize the role of a follower, relying on the reverse change of the base current of the transistor to offset the change of the collector current, thereby driving the LED light source to emit a first light signal in the form of a constant current.

[0011] Furthermore, the detector module includes a plurality of photodiodes, a plurality of pre-transimpedance amplifiers, a plurality of post-stage programmable gain amplifiers, a plurality of filter processors and an analog-to-digital converter (ADC) module; the number of photodiodes, pre-transimpedance amplifiers, post-stage programmable gain amplifiers and filter processors is equal and one-to-one corresponding; the ADC is multi-channel, each channel receives an analog signal output by a filter processor and converts it into a digital signal; A photodiode and a pre-stage cross-group amplifier form a detector; The detector module uses a photodiode to detect the near-infrared light diffusely reflected by the human body, which is the second light signal. The photodiode converts the second light signal into a current signal. Since the light signal is weak and difficult to detect accurately over long distances, the pre-transimpedance amplifier converts the current signal into an analog voltage signal and performs front-end amplification to improve the signal-to-noise ratio of the weak analog voltage signal and transmit it to the post-stage programmable gain amplifier through a shielded line. The post-stage programmable gain amplifier is programmed to adjust different gains so that the analog voltage signal gain is within the range that the analog-to-digital converter can read. The filter processor removes the high-frequency instrument noise in the signal and inputs it into the analog-to-digital converter module. The analog-to-digital converter module converts the analog voltage signal after noise removal into multi-channel original light intensity data, and finally transmits the multi-channel original light intensity data to the controller module.

[0012] Furthermore, in the wearable near-infrared device, the second light signal generated by the first light signal emitted by each LED light source is received by two detectors, each detector including a photodiode and a pre-transimpedance amplifier; an LED light source and a detector constitute a channel; Among them, the distance between the LED light source and the corresponding first detector is 30mm, forming a 30mm long channel. A second detector is arranged between the LED light source and the corresponding first detector. The second detector and the LED light source form a 15mm short channel. The short-distance channel contains information of shallow physiological interference noise, and the long-distance channel contains shallow interference noise and deep hemodynamic changes, and changes in hemoglobin concentration caused by deep brain signal changes. When collecting data, the noise interference in the long-channel data collected by the first detector is estimated by the short-channel data collected by the second detector, and the noise is removed by the host computer.

[0013] Furthermore, the controller module has SPI communication, and the multi-channel digital-to-analog converter is controlled by the controller module with SPI communication to realize the correct sequence of lighting up the LED light sources in sequence; The controller module controls the analog voltage of each channel to achieve separate control of the optical power of each LED light source; different optical powers are required for light source-detector channels worn at different parts of the body and at different distances. The controller module determines whether the original light intensity signal received by the detector module is oversaturated or too weak, and achieves adaptive optical power regulation by adjusting the analog voltage of the multi-channel analog-to-digital converter.

[0014] Furthermore, the controller module transmits the data of each channel in the collected multi-channel raw light intensity data to the host computer through the wireless WIFI module for processing; when the subject moves, the relative movement of the head and the probe will generate motion artifacts and baseline drift errors. The host computer performs real-time artifact removal and noise removal on the input raw light intensity data based on spline interpolation and wavelet function fusion algorithm, and then converts the raw light intensity data after real-time artifact removal and noise removal into blood oxygen concentration data.

[0015] Furthermore, in the host computer, artifact detection evaluation is performed based on the kurtosis of the continuous wavelet transform coefficient distribution, as follows: According to the data of each channel in the input multi-channel raw light intensity data, the kurtosis threshold is estimated and set, and the part of the raw light intensity data exceeding the kurtosis threshold is considered as an artifact; the estimation formula of the kurtosis threshold h is as follows:

[0016] in, is the first i samples, is the sample mean, and n is the total number of samples.

[0017] Furthermore, in the host computer, the short channel data collected by the second detector is used to further filter out the global oscillation interference of the superficial skin, and construct an adaptive filtering error signal. This process is equivalent to using the outer layer hemodynamic changes to estimate the physiological interference related to the hemodynamic change signal induced by it. Therefore, the long channel data collected by the first detector minus the short channel data collected by the second detector is the hemodynamic response of the brain activity process to be sought, that is, the deep brain signal change, as follows: Deep brain signal changes It is expressed as:

[0018] in, Indicates the change in oxygenated hemoglobin concentration calculated from the long channel data collected by the first detector or changes in deoxyhemoglobin concentration , including physiological disturbances and induced hemodynamic changes, is the sampling time; is a column vector, represented by is the adaptive filter order, which defines the number of coefficients contained in the coefficient vector of the adaptive filter, that is, the complexity of the adaptive filter; For signal of The adaptive filter will take into account the signal values ​​at the current moment and the previous N moments to calculate the filter output at the current moment, which can capture the dynamic changes of the signal and improve the filtering performance; represents the change in oxygenated hemoglobin concentration obtained from the short channel data collected by the second detector or changes in deoxyhemoglobin concentration , which is mainly composed of physiological interference and serves as a reference signal for adaptive filtering; is the coefficient vector of the adaptive filter; represents the first coefficient of the adaptive filter coefficient vector w(k) at sampling time k. elements; It changes as the adaptive filter is iteratively updated to minimize the error between the adaptive filter output and the desired response.

[0019] Furthermore, in the host computer, based on the modified Beer-Lambert law, the input is the original light intensity data after real-time artifact removal and noise removal, and the output is the blood oxygen concentration; in a uniform non-scattering matrix, the intensity relationship between the incident light and the outgoing light can be described by the modified Beer-Lambert law, and the expression is as follows:

[0020] In the above formula, Represents the outgoing light intensity, i.e. the original light intensity data, represents the incident light intensity, represents the molecular extinction coefficient, represents the concentration of the transmission medium, It represents the straight-line distance between the detector and the light source; the molecular extinction coefficient and the concentration of the transport medium They are all constants and can be obtained by looking up the relevant table; The optical density attenuation is defined as:

[0021] Introducing a scattering-related light intensity loss parameter To correct the Beer-Lambert law, in addition, DFP (Differential path length factor) is called differential path length factor; is the optical wavelength parameter; is the absorption coefficient, ; is the differential path length factor for a specific wavelength, which can be found from the relevant table; according to the optical density attenuation The change in optical density attenuation ΔOD is calculated; The changes in the concentrations of oxygenated hemoglobin and deoxygenated hemoglobin are:

[0022]

[0023] in, is the change in oxygenated hemoglobin, is the change in deoxyhemoglobin, The first light wavelength of the first light signal emitted by the LED light source Oxygenated hemoglobin The extinction coefficient, The second light wavelength of the first light signal emitted by the LED light source The change in optical density, Differential path length factor, It represents the straight-line distance between the detector and the light source.

[0024] An LED light source will have light signals with two wavelengths. Light is essentially an electromagnetic wave with two characteristics: wavelength and intensity. The wavelengths of the LED light source of the present invention are 730nm and 850nm, and different wavelengths correspond to different extinction coefficients.

[0025] Furthermore, in the host computer, a light source-detector topology map including the distribution of light sources and detectors is obtained, the light source-detector topology map is aligned with the standard template of the cerebral scalp, the spatial distribution map of brain functional activities is drawn in real time, and a real-time high-resolution cerebral blood oxygen change model is constructed to display the activities of different brain regions.

[0026] Compared with the prior art, the advantages of the present invention are: The present invention is a high-resolution portable near-infrared brain function imaging system. The present invention adopts a dual-wavelength acquisition method and completes device selection. It selects a high-precision DAC and a light source driving circuit to drive the light source, realizes real-time adjustable light source luminous power, and flexibly controls the timing and intensity of the light source to meet the measurement requirements of different distances; the present invention can drive 32 light sources to work at the same time. The detector module adopts a PD photodiode, which reduces the power consumption of the system while amplifying it through a front transimpedance amplifier and a back-end programmable gain amplifier to improve the signal-to-noise ratio, and then eliminates global noise through a filtering processor, thereby increasing the effective information in the multi-spacing channel combination.

[0027] The artifact detection kbWD algorithm used in the present invention uses only a single parameter (the kurtosis of the WC distribution) to determine artifact-related outliers and obtain their moments. After obtaining artifact detection, the good positioning brought by the wavelet method can eliminate spikes and low-amplitude slow drifts. Spline interpolation can remove baseline drifts. Spline interpolation and wavelet filtering are combined to perform high-precision artifact correction, achieve accurate extraction of brain function signals, and establish a complete brain function signal processing flow. The present invention has developed a set of continuous wave near-infrared brain function imaging systems with complete functions, strong stability, and suitable for complex task scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 A light source-detector topology diagram including the distribution of light sources and detectors in an embodiment of the present invention; Figure 2 It is a detection principle diagram of the short pain channel and the long channel formed between the LED light source and the detector in an embodiment of the present invention; Figure 3 This is a schematic diagram of the structure of a wearable near-infrared device in an embodiment of the present invention; Figure 4 Schematic diagram of the structure of a high-resolution portable near-infrared brain function imaging system in an embodiment of the present invention; Figure 5 Schematic diagram of the structure of the light source driving circuit in an embodiment of the present invention. DETAILED DESCRIPTION

[0029] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the specific implementation of the present invention is described in detail below with reference to the accompanying drawings and embodiments.

[0030] Example: A high-resolution portable near-infrared brain functional imaging system, such as Figure 4 As shown, it includes a wearable near-infrared device and a host computer; the wearable near-infrared device is provided with a controller module, a wireless WIFI module, a light source module and a detector module; After the subject wears the wearable near-infrared device, the controller module drives the light source module, and the LED light sources in the light source module are lit in turn to emit a first light signal to the subject's brain area. The second light signal generated by the diffuse reflection of the first light signal in the brain area is received by the photodiode in the detector module and converted into a current signal. After signal processing, the analog-to-digital converter module (ADC) converts the signal acquisition into multi-channel raw light intensity data and transmits it to the controller module. The controller module sends the multi-channel raw light intensity data to the host computer through the wireless WIFI module. The host computer quantitatively converts the multi-channel raw light intensity data into blood oxygen concentration through the modified Lambert-Beer law, constructs a real-time high-resolution brain blood oxygen change model, and realizes brain function imaging.

[0031] Furthermore, the light source module includes a multi-channel digital-to-analog converter (DAC), a plurality of light source driving circuits and a plurality of LED light sources; the number of the light source driving circuits and the LED light sources are equal and correspond one to one; each channel in the multi-channel DAC controls a light source driving circuit to drive the corresponding LED light source respectively; The controller module sends the digital signal to the multi-channel digital-to-analog converter to convert it into an analog voltage signal, and then the light source driving circuit converts the analog voltage signal into a current signal, thereby driving the LED light source to work under a constant current condition. The multi-channel digital-to-analog converter is used to realize the independent adjustment of the light intensity of the light source, and realizes adaptive LED light source adjustment for personalized human body. like Figure 5 As shown, in the light source driving circuit, the collector of the transistor is connected to the cathode of the LED light source, the anode of the LED light source is connected to the driving voltage VCC of 5V, the analog voltage signal output by the multi-channel digital-to-analog converter is input to the in-phase input terminal of the operational amplifier, and the operational amplifier is used to realize the role of a follower, relying on the reverse change of the base current of the transistor to offset the change of the collector current, thereby driving the LED light source to emit a first light signal in the form of a constant current.

[0032] In one embodiment, in order to accurately control the current to achieve adaptive optical power regulation, the multi-channel digital-to-analog converter uses TI's 8-channel 16-bit digital-to-analog converter DAC8568; Furthermore, the detector module includes a plurality of photodiodes, a plurality of pre-transimpedance amplifiers, a plurality of post-stage programmable gain amplifiers, a plurality of filter processors and an analog-to-digital converter (ADC) module; the number of photodiodes, pre-transimpedance amplifiers, post-stage programmable gain amplifiers and filter processors is equal and one-to-one corresponding; the ADC is multi-channel, each channel receives an analog signal output by a filter processor and converts it into a digital signal; A photodiode and a pre-stage cross-group amplifier form a detector; In one embodiment, the photodiode is a silicon photodiode; The detector module uses a photodiode to detect the near-infrared light diffusely reflected by the human body, which is the second light signal. The photodiode converts the second light signal into a current signal. Since the light signal is weak and difficult to detect accurately over long distances, the pre-transimpedance amplifier converts the current signal into an analog voltage signal and performs front-end amplification to improve the signal-to-noise ratio of the weak analog voltage signal and transmit it to the post-stage programmable gain amplifier through a shielded line. The post-stage programmable gain amplifier is programmed to adjust different gains so that the analog voltage signal gain is within the range that the analog-to-digital converter can read. The filter processor removes the high-frequency instrument noise in the signal and inputs it into the analog-to-digital converter module. The analog-to-digital converter module converts the analog voltage signal after noise removal into multi-channel original light intensity data, and finally transmits the multi-channel original light intensity data to the controller module.

[0033] In one embodiment, the front-end transimpedance amplifier module uses a TLC2201 module, and the post-stage programmable gain amplifier uses an AD7606C-18 module; In one embodiment, the signal received by the photodiode is 7-9 times less than the input light intensity and contains noise. A low-pass filter is selected in the filter processor for filtering, eliminating various noise interference signals generated in the circuit. Through the analysis of the types and performance of low-pass filters, the Butterworth low-pass filter has the characteristics of smooth amplitude characteristics and no fluctuations in the passband, and the present invention does not require strict requirements on the linear phase of the low-pass filter, which meets the requirements of the system for the design of the filter circuit, and finally the Butterworth low-pass filter TLC14 module is selected.

[0034] Furthermore, in wearable near-infrared devices, such as Figure 1 and Figure 2 As shown, the second light signal generated by the first light signal emitted by each LED light source 2 is received by two detectors 1, and each detector 1 includes a photodiode and a pre-transimpedance amplifier; an LED light source 2 and a detector 1 constitute a channel; Among them, the distance between the LED light source 2 and the corresponding first detector 101 is 30 mm, forming a 30 mm long channel. A second detector 102 is arranged between the LED light source 2 and the corresponding first detector 101. The second detector 102 and the LED light source 2 form a 15 mm short channel. The short-distance channel contains information on shallow physiological interference noise, and the long-distance channel contains shallow interference noise and deep hemodynamic changes, and changes in hemoglobin concentration caused by deep brain signal changes. When collecting data, the noise interference in the long channel data collected by the first detector 101 is estimated by the short-channel data collected by the second detector 102, and the noise is removed by the host computer.

[0035] In one embodiment, 24 light sources and 32 detectors are provided to form 24*32 channels. Different post-stage programmable gain adjustments are required for channels at different distances. A post-stage programmable gain amplifier is used to achieve 1-3 times voltage amplification, and adaptive adjustments are performed on short-distance channels and long-distance channels to meet the same signal-to-noise ratio.

[0036] In one embodiment, Figure 3 As shown, a flexible wearable device is arranged in the wearable near-infrared device to ensure good contact with biological tissues through the ductility of the device itself under the condition of electrical stability. The present invention adopts an FPC flexible circuit board to ensure the stability of the device while enhancing good coupling with biological tissues, ensuring that the internal optical elements are closely attached to the scalp to avoid the generation of relative displacement between the scalp and the probe itself. Miniaturization and lightweight are ensured to reduce the burden on the subject's head. The probe structure consists of three parts: a tubular shell, a gland, and an internal circular PCB, wherein the tubular shell and the gland are manufactured by 3D printing, and the material is lightweight polyester fiber. The outer shell has a diameter of no more than 1cm and a height of about 5mm. An LED light source or detector is welded on the circular PCB inside. The packaged probe is fixed on a silicone fixing belt according to a certain layout. The softness and elasticity of the silicone ensure the stability of the fixing module, ensuring that the probe fits perfectly with the scalp during the test. A narrow-band filter with a wavelength of 700nm-900nm is placed between the detector and the biological tissue to filter out ambient light and reduce the impact of human bioelectricity on near-infrared signals, thereby increasing the possibility of continuous and long-term brain detection under daily conditions, and realizing a flexible, wearable and portable near-infrared brain function imaging system.

[0037] Aiming at the detection needs of the frontal brain area, a lightweight probe fixing cover was designed through 3D printing; The source-detection topology based on flexible circuit boards and narrow-band filters ensures close fit with the human body and reduces the impact of ambient light, increasing the number of channels and signal quality, and extracting reliable brain function signals.

[0038] Furthermore, the controller module has SPI communication, and the multi-channel digital-to-analog converter is controlled by the controller module with SPI communication to realize the correct sequence of lighting up the LED light sources in sequence; The controller module controls the analog voltage of each channel to achieve separate control of the optical power of each LED light source; different optical powers are required for light source-detector channels worn at different parts of the body and at different distances. The controller module determines whether the original light intensity signal received by the detector module is oversaturated or too weak, and achieves adaptive optical power regulation by adjusting the analog voltage of the multi-channel analog-to-digital converter.

[0039] Furthermore, the controller module transmits the data of each channel in the collected multi-channel raw light intensity data to the host computer through the wireless WIFI module for processing; when the subject moves, the relative movement of the head and the probe will generate motion artifacts and baseline drift errors. The host computer performs real-time artifact removal and noise removal on the input raw light intensity data based on spline interpolation and wavelet function fusion algorithm, and then converts the raw light intensity data after real-time artifact removal and noise removal into blood oxygen concentration data.

[0040] Furthermore, in the host computer, artifact detection evaluation is performed based on the kurtosis of the continuous wavelet transform coefficient distribution, as follows: According to the data of each channel in the input multi-channel raw light intensity data, the kurtosis threshold is estimated and set, and the part of the raw light intensity data exceeding the kurtosis threshold is considered as an artifact; the estimation formula of the kurtosis threshold h is as follows:

[0041] in, is the first i samples, is the sample mean, and n is the total number of samples.

[0042] Furthermore, in the host computer, the short channel data collected by the second detector is used to further filter out the global oscillation interference of the superficial skin, and construct an adaptive filtering error signal. This process is equivalent to using the outer layer hemodynamic changes to estimate the physiological interference related to the hemodynamic change signal induced by it. Therefore, the long channel data collected by the first detector minus the short channel data collected by the second detector is the hemodynamic response of the brain activity process to be sought, that is, the deep brain signal change, as follows: Deep brain signal changes It is expressed as:

[0043] in, Indicates the change in oxygenated hemoglobin concentration calculated from the long channel data collected by the first detector or changes in deoxyhemoglobin concentration , including physiological disturbances and induced hemodynamic changes, is the sampling time; is a column vector, represented by is the adaptive filter order, which defines the number of coefficients contained in the coefficient vector of the adaptive filter, that is, the complexity of the adaptive filter; For signal of The adaptive filter will take into account the signal values ​​at the current moment and the previous N moments to calculate the filter output at the current moment, which can capture the dynamic changes of the signal and improve the filtering performance; represents the change in oxygenated hemoglobin concentration obtained from the short channel data collected by the second detector or changes in deoxyhemoglobin concentration , which is mainly composed of physiological interference and serves as a reference signal for adaptive filtering; is the coefficient vector of the adaptive filter; represents the first coefficient of the adaptive filter coefficient vector w(k) at sampling time k. elements; It changes as the adaptive filter is iteratively updated to minimize the error between the adaptive filter output and the desired response.

[0044] Furthermore, in the host computer, based on the modified Beer-Lambert law, the input is the original light intensity data after real-time artifact removal and noise removal, and the output is the blood oxygen concentration; in a uniform non-scattering matrix, the intensity relationship between the incident light and the outgoing light can be described by the modified Beer-Lambert law, and the expression is as follows:

[0045] In the above formula, Represents the outgoing light intensity, i.e. the original light intensity data, represents the incident light intensity, represents the molecular extinction coefficient, represents the concentration of the transmission medium, It represents the straight-line distance between the detector and the light source; the molecular extinction coefficient and the concentration of the transport medium They are all constants and can be obtained by looking up the relevant table; The optical density attenuation is defined as:

[0046] Introducing a scattering-related light intensity loss parameter To correct the Beer-Lambert law, in addition, DFP (Differential path length factor) is called differential path length factor; is the optical wavelength parameter; is the absorption coefficient, ; is the differential path length factor for a specific wavelength, which can be found from the relevant table; according to the optical density attenuation The change in optical density attenuation ΔOD is calculated; The changes in the concentrations of oxygenated hemoglobin and deoxygenated hemoglobin are:

[0047]

[0048] in, is the change in oxygenated hemoglobin, is the change in deoxyhemoglobin, The first light wavelength of the first light signal emitted by the LED light source Oxygenated hemoglobin The extinction coefficient, The second light wavelength of the first light signal emitted by the LED light source The change in optical density, Differential path length factor, It represents the straight-line distance between the detector and the light source.

[0049] An LED light source will have two wavelengths of light signals. Light is essentially an electromagnetic wave with two characteristics: wavelength and intensity. In one embodiment, the two wavelengths of the LED light source are 730nm and 850nm, respectively, and different wavelengths correspond to different extinction coefficients.

[0050] Furthermore, in the host computer, a light source-detector topology map including the distribution of light sources and detectors is obtained, the light source-detector topology map is aligned with the standard template of the cerebral scalp, the spatial distribution map of brain functional activities is drawn in real time, and a real-time high-resolution cerebral blood oxygen change model is constructed to display the activities of different brain regions.

[0051] In one embodiment, the cerebral scalp standard template adopts a 10-20 lead system.

[0052] The preferred embodiments of the present application disclosed above are only used to help understand the present invention and its core ideas. For those skilled in the art, according to the ideas of the present invention, there will be changes in specific application scenarios and implementation operations, and this specification should not be interpreted as limiting the present invention. The present invention is only limited by the claims and their full scope and equivalents.

Claims

1. A high-resolution portable near-infrared brain function imaging system, characterized in that: It includes a wearable near-infrared device and a host computer; the wearable near-infrared device is provided with a controller module, a wireless WIFI module, a light source module and a detector module; After the subject wears the wearable near-infrared device, the controller module drives the light source module, and the LED light sources in the light source module are lit in turn to emit a first light signal to the subject's brain area. The second light signal generated by the diffuse reflection of the first light signal by the brain area is received by the photodiode in the detector module and converted into a current signal. After signal processing, it is converted into multi-channel original light intensity data and transmitted to the controller module. The controller module sends the multi-channel original light intensity data to the host computer through the wireless WIFI module. The host computer quantitatively converts the multi-channel original light intensity data into blood oxygen concentration, constructs a real-time high-resolution brain blood oxygen change model, and realizes brain function imaging.

2. A high-resolution portable near-infrared brain function imaging system according to claim 1, characterized in that: The light source module includes a multi-channel digital-to-analog converter (DAC), a plurality of light source driving circuits and a plurality of LED light sources; the number of the light source driving circuits and the number of the LED light sources are equal and correspond one to one; Each channel of the multi-channel analog-to-digital converter controls a light source driving circuit to drive the corresponding LED light source respectively; The controller module sends the digital signal to the multi-channel digital-to-analog converter to convert it into an analog voltage signal, and then the light source driving circuit converts the analog voltage signal into a current signal, thereby driving the LED light source to work under a constant current condition. The multi-channel digital-to-analog converter is used to realize the independent adjustment of the light intensity of the light source, and realizes adaptive LED light source adjustment for personalized human body. In the light source driving circuit, the collector of the transistor is connected to the cathode of the LED light source, the anode of the LED light source is connected to the driving voltage, the analog voltage signal output by the multi-channel digital-to-analog converter is input to the non-inverting input terminal of the operational amplifier, and the operational amplifier is used to realize the role of a follower, relying on the reverse change of the base current of the transistor to offset the change of the collector current, thereby driving the LED light source to emit a first light signal in the form of a constant current.

3. A high-resolution portable near-infrared brain function imaging system according to claim 1, characterized in that: The detector module includes a plurality of photodiodes, a plurality of pre-transimpedance amplifiers, a plurality of post-stage programmable gain amplifiers, a plurality of filter processors and an analog-to-digital converter (ADC) module; the number of photodiodes, pre-transimpedance amplifiers, post-stage programmable gain amplifiers and filter processors is equal and one-to-one corresponding; the ADC is multi-channel, each channel receives an analog signal output by a filter processor and converts it into a digital signal; A photodiode and a pre-stage cross-group amplifier form a detector; The detector module uses a photodiode to detect the near-infrared light diffusely reflected by the human body, that is, the second light signal. The photodiode converts the second light signal into a current signal. The front transimpedance amplifier converts the current signal into an analog voltage signal and performs front-end amplification to improve the signal-to-noise ratio of the weak analog voltage signal and transmit it to the post-stage programmable gain amplifier through a shielded line. The post-stage programmable gain amplifier is programmed to adjust different gains so that the analog voltage signal gain is within the range that the analog-to-digital converter can read. The filter processor removes the high-frequency instrument noise in the signal and inputs it into the analog-to-digital converter module. The analog-to-digital converter module converts the analog voltage signal after noise removal into multi-channel original light intensity data, and finally transmits the multi-channel original light intensity data to the controller module.

4. A high-resolution portable near-infrared brain function imaging system according to claim 1, characterized in that: In the wearable near-infrared device, the second light signal generated by the first light signal emitted by each LED light source is received by two detectors, each detector includes a photodiode and a pre-transimpedance amplifier; an LED light source and a detector constitute a channel; Among them, the distance between the LED light source and the corresponding first detector is 30mm, forming a 30mm long channel. A second detector is arranged between the LED light source and the corresponding first detector. The second detector and the LED light source form a 15mm short channel. When collecting data, the short channel data collected by the second detector is used to estimate the noise interference in the long channel data collected by the first detector, and the noise is removed by the host computer.

5. A high-resolution portable near-infrared brain function imaging system according to claim 3, characterized in that: The controller module has SPI communication, and the multi-channel digital-to-analog converter is controlled by the controller module with SPI communication to realize the correct sequence of LED light sources lighting up in sequence; The controller module controls the analog voltage of each channel to achieve separate control of the optical power of each LED light source; different optical powers are required for light source-detector channels worn at different parts of the body and at different distances. The controller module determines whether the original light intensity signal received by the detector module is oversaturated or too weak, and achieves adaptive optical power regulation by adjusting the analog voltage of the multi-channel analog-to-digital converter.

6. A high-resolution portable near-infrared brain function imaging system according to claim 1, characterized in that: The controller module transmits the data of each channel in the collected multi-channel raw light intensity data to the host computer through the wireless WIFI module for processing; when the subject moves, the relative movement of the head and the probe will produce motion artifacts and baseline drift errors. The host computer performs real-time artifact removal and noise removal on the input raw light intensity data based on spline interpolation and wavelet function fusion algorithm, and then converts the raw light intensity data after real-time artifact removal and noise removal into blood oxygen concentration data.

7. A high-resolution portable near-infrared brain function imaging system according to claim 6, characterized in that: In the host computer, artifact detection and evaluation are performed based on the kurtosis of the continuous wavelet transform coefficient distribution, as follows: According to the data of each channel in the input multi-channel raw light intensity data, the kurtosis threshold is estimated and set, and the part of the raw light intensity data exceeding the kurtosis threshold is considered as an artifact; the estimation formula of the kurtosis threshold h is as follows: in, is the first i samples, is the sample mean, and n is the total number of samples.

8. A high-resolution portable near-infrared brain function imaging system according to claim 6, characterized in that: In the host computer, the short channel data collected by the second detector is used to further filter out the global oscillation interference of the superficial skin, and construct an adaptive filtering error signal. This process is equivalent to using the outer layer hemodynamic changes to estimate the physiological interference related to the induced hemodynamic change signal. Therefore, the long channel data collected by the first detector minus the short channel data collected by the second detector is the hemodynamic response of the brain activity process to be sought, that is, the deep brain signal change, as follows: Deep brain signal changes It is expressed as: in, Indicates the change in oxygenated hemoglobin concentration calculated from the long channel data collected by the first detector or changes in deoxyhemoglobin concentration , including physiological disturbances and induced hemodynamic changes, is the sampling time; is a column vector, represented by is the adaptive filter order; For signal of Unit extension; represents the change in oxygenated hemoglobin concentration obtained from the short channel data collected by the second detector or changes in deoxyhemoglobin concentration , which is mainly composed of physiological interference and serves as a reference signal for adaptive filtering; is the coefficient vector of the adaptive filter; represents the first coefficient of the adaptive filter coefficient vector w(k) at sampling time k. elements.

9. A high-resolution portable near-infrared brain function imaging system according to claim 6, characterized in that: In the host computer, based on the modified Beer-Lambert law, the input is the original light intensity data after real-time artifact removal and noise removal, and the output is the blood oxygen concentration; in a uniform non-scattering matrix, the intensity relationship between the incident light and the outgoing light can be described by the modified Beer-Lambert law, and the expression is as follows: In the above formula, Represents the outgoing light intensity, i.e. the original light intensity data, represents the incident light intensity, represents the molecular extinction coefficient, represents the concentration of the transmission medium, It represents the straight-line distance between the detector and the light source; the molecular extinction coefficient and the concentration of the transport medium They are all constants and can be obtained by looking up the relevant table; The optical density attenuation is defined as: Introducing a scattering-related light intensity loss parameter To correct the Beer-Lambert law, in addition, DFP (Differential path length factor) is called differential path length factor; is the optical wavelength parameter; is the absorption coefficient, ; is the differential path length factor for a specific wavelength, which can be found from the relevant table; according to the optical density attenuation The change in optical density attenuation ΔOD is calculated; The changes in the concentrations of oxygenated hemoglobin and deoxygenated hemoglobin are: in, is the change in oxygenated hemoglobin, is the change in deoxyhemoglobin, The first light wavelength of the first light signal emitted by the LED light source Oxygenated hemoglobin The extinction coefficient, The second light wavelength of the first light signal emitted by the LED light source The change in optical density, Differential path length factor, It represents the straight-line distance between the detector and the light source.

10. A high-resolution portable near-infrared brain function imaging system according to claim 1, characterized in that: In the host computer, a light source-detector topology map including the distribution of light sources and detectors is obtained, and the light source-detector topology map is aligned with the standard template of the cerebral scalp. The spatial distribution map of brain functional activities is drawn in real time, and a real-time high-resolution cerebral blood oxygen change model is constructed to display the activities of different brain regions.

Images