Portable near-infrared brain function imaging system

Abstract

The invention relates to the field of wearable medical electronics and nerve monitoring equipment, in particular to a portable near-infrared brain function imaging system, and aims to solve the problem that the real-time signal-to-noise ratio (SNR) cannot meet reliable brain blood oxygen measurement requirements due to insufficient interference resistance of a long wiring harness analog signal chain of a portable fNIRS in a dynamic or high-noise environment. The system comprises a light source subsystem, a detector, a middle adapter plate, a main control board, a power supply module and a shell, a transimpedance amplifier is directly integrated behind a photodiode on a probe of the detector, an analog signal is amplified at a source and preliminarily filtered, the amplified signal is transmitted to the main control board through a wire harness, the anti-interference capability can be remarkably enhanced, and the detection accuracy is improved. The system signal-to-noise ratio and the measurement stability are improved, and meanwhile the load of a mainboard circuit module is relieved. In addition, the system is low in manufacturing cost.

Description

Technical Field

[0001] The present application relates to the field of wearable medical electronics and neural monitoring equipment, and in particular to a portable near-infrared brain function imaging system. Background Art

[0002] Functional near-infrared spectroscopy (fNIRS) has become one of the most active non-invasive brain imaging methods, alongside EEG and fMRI, due to its low light source power requirements and ability to provide long-term physiological monitoring in a natural posture. With the rapid advancement of flexible wearable materials, miniaturized ADCs, and low-power wireless SoCs, portable fNIRS is expanding from niche applications such as performing arts, sports training, and driver fatigue monitoring to larger-scale applications such as classroom group learning, remote postoperative rehabilitation, and home neurological monitoring. The market has maintained a compound annual growth rate of over 18% for five consecutive years.

[0003] However, the core bottleneck of current wearable fNIRS still lies in the insufficient anti-interference of the analog signal chain: the picoampere-level current output by the detector photodiode needs to be transmitted to the main control PCB along a 30-60cm flexible flat wiring harness, which is often affected by 50 / 60Hz power frequency, electromagnetic radiation and head shaking coupled noise in strong outdoor noise or exercise conditions; public tests show that the dynamic SNR often drops to 20-25dB, while high-quality brain blood oxygen signal acquisition requires at least 35dB, resulting in a large amount of data post-processing or even complete abandonment. Summary of the Invention

[0004] In view of this, the embodiments of the present disclosure provide a portable near-infrared functional brain imaging system to solve the problem that the long wiring harness analog signal chain of portable fNIRS in dynamic or high-noise environments is insufficient in anti-interference, resulting in the real-time signal-to-noise ratio (SNR) being unable to meet the requirements of reliable cerebral blood oxygen measurement.

[0005] An embodiment of the present disclosure provides a portable near-infrared brain functional imaging system, including a light source subsystem, a detector subsystem, an intermediate adapter board, a main control board, a power supply module and a casing. A transimpedance amplifier is directly integrated behind the photodiode on the probe of the detector subsystem to amplify the analog signal at the source and perform preliminary filtering. The amplified signal is then transmitted to the main control board through a wiring harness.

[0006] In some embodiments, the constant current driver chip and the detector subsystem output signal line are integrated on the intermediate adapter board, and the constant current accuracy is not less than ±1%; the input equivalent noise current density of the transimpedance amplifier is less than 50 fA / √Hz@1MHz, and the gain-bandwidth product is not less than 250MHz.

[0007] In some embodiments, the thickness of the intermediate adapter plate is no more than 0.5 mm, and the constant current driver chip of the multiple LEDs is embedded therein. 2 C bus feedback to the MCU.

[0008] In some embodiments, the MCU runs a real-time operating system for scheduling LED dimming, data acquisition, wireless transmission and power monitoring tasks, and performs AES 128 encryption on data frames; the constant current driver chip is used for switching control of multiple light sources and stepless control of their brightness.

[0009] In some embodiments, the driving circuit is moved forward to the intermediate adapter board close to the light source probe, and the IO requirements and the number of peripheral discrete components on the main control board are greatly reduced.

[0010] In some embodiments, the portable near-infrared brain functional imaging system also includes a magnetic connection structure, which separates the main control board from the battery control board. When the main battery is exhausted, the old battery is quickly unplugged and the backup battery is inserted, and the power supply of the main control board can continue.

[0011] In some embodiments, the contact resistance of the magnetic interface before and after disconnection and re-contact does not change by more than 50 mΩ, and the replacement operation time does not exceed 10 seconds.

[0012] In some embodiments, the portable near-infrared brain functional imaging system further includes a wireless communication module, which supports a 40 MHz channel bandwidth and is configured to operate in the 2.4 GHz and / or 5 GHz frequency bands.

[0013] In some embodiments, the wireless communication module uploads encrypted data to a cloud server to achieve multi-node synchronous brain function monitoring.

[0014] In some embodiments, the housing can be combined with a photoelectrode cap and worn behind the head to keep the optical probe stably fitted to the scalp in mobile scenarios, and the system mass is less than 120g.

[0015] In some embodiments, the filtering is based on a combination of Kalman filtering and 0.01 Hz high-pass filtering to remove drift and motion artifacts. The data frame uploaded to the cloud server includes not only the filtered detector signal data but also the LED current, battery power and signal strength information of the wireless communication module for real-time quality control and remote parameter adaptation.

[0016] Compared with the prior art, the embodiments of the present disclosure have at least the following beneficial effects: A portable near-infrared brain imaging system, provided by specific embodiments of the present disclosure, features a transimpedance amplifier integrated directly after the photodiode on the detector's probe. This amplifier amplifies and initially filters the analog signal at the source. The amplified signal is then transmitted to the main control board via a wiring harness. This significantly enhances anti-interference capabilities, improves the system's signal-to-noise ratio and measurement stability, and reduces the load on the mainboard circuit modules. Furthermore, the system offers low manufacturing costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The above and other purposes, features, and advantages of the present disclosure will become more apparent through a more detailed description of the embodiments of the present disclosure in conjunction with the accompanying drawings. The accompanying drawings are intended to provide a further understanding of the embodiments of the present disclosure and constitute a part of the specification. Together with the embodiments of the present disclosure, they are used to explain the present disclosure and are not intended to limit the present disclosure. In the drawings, the same reference numerals generally represent the same components or steps.

[0018] Figure 1 Shown is a schematic diagram of the system function flow provided by an embodiment of the present disclosure.

[0019] Figure 2 Shown is a structural schematic diagram of a portable near-infrared brain functional imaging system provided by an embodiment of the present disclosure.

[0020] Figure 3 Shown is a schematic structural diagram of a probe provided in one embodiment of the present disclosure.

[0021] Figure 4 Shown is a schematic diagram of an RTOS task timing diagram provided by an embodiment of the present disclosure.

[0022] Figure 5 Shown is a schematic diagram of a host magnetic attraction structure provided by an embodiment of the present disclosure.

[0023] Figure 6 Shown is a structural diagram of a host control panel provided by an embodiment of the present disclosure.

[0024] Figure 7 FIG2 is a schematic diagram showing a comparison of bandwidth histograms of BLE and low-power wireless communication modules provided in an embodiment of the present disclosure.

[0025] Figure 8 Shown is a line graph comparing the SNR of a traditional long-wire harness architecture provided by an embodiment of the present disclosure and the present disclosure.

[0026] Reference numerals: 10. A portable near-infrared brain functional imaging system; 100, probe; 110, probe head; 120, probe tail; 130, LED circuit board; 140, filter; 200, host; 210, battery; 220, host control panel; 221, rectangular block; 300, switching circuit components; 400, wiring harness. DETAILED DESCRIPTION

[0027] The following will clearly and completely describe the technical solutions in the embodiments of the present disclosure in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present disclosure, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present disclosure without making any creative efforts shall fall within the scope of protection of the present disclosure.

[0028] Prior art portable near-infrared brain imaging systems have the following technical shortcomings, such as weak anti-interference capabilities, large system size, limited uninterrupted battery life, and poor multi-device synchronous acquisition capabilities. The present invention provides a portable near-infrared brain imaging system that not only effectively addresses these core technical issues but also improves upon these shortcomings. The specific details are as follows: An embodiment of the present disclosure provides a portable near-infrared brain function imaging system 10, which includes a light source subsystem, a detector subsystem, an intermediate adapter board, a main control board, a power supply module, and a housing.

[0029] Specifically, if Figure 2 As shown in FIG, a portable near-infrared brain function imaging system 10 includes a probe 100, a host 200, a switching circuit component 300 and a wiring harness 400. Figure 1As shown, the LEDs in the light source subsystem emit near-infrared light through a constant-current driver module, which then illuminates the cerebral cortex. After scattering through the skull and brain tissue, the light signal is received by the photodiodes in the detector subsystem. The received light signal is converted to a picoampere current, then amplified and initially filtered by a transimpedance amplifier (TIA) before being converted to a volt-level voltage. The amplified analog signal is then transmitted via a cable to an intermediate adapter board. This adapter board not only performs signal transfer but also integrates a constant-current driver module for the light source, providing stable drive current for each LED channel. The adapter board is connected to the main control board, which integrates an MCU and is responsible for LED dimming control, signal acquisition, data processing, wireless transmission, and power monitoring. The MCU regulates the output current of each light source channel via a control bus and digitizes the photoelectric signal using an ADC module to quantify light intensity. The collected data is uploaded to a cloud server via a wireless communication module. The power monitoring module monitors the system power status in real time. When the battery level drops below 10%, an audible and visual reminder alerts the user to replace the battery. The system's built-in supercapacitor energy storage module provides short-term battery life during battery replacement, ensuring the replacement process is completed within 10 seconds without disrupting system operation. The battery board includes a lithium battery, a battery charge and discharge management module, and a host status display module. The charge and discharge management module monitors battery charging, discharging, and temperature to ensure safe operation. The host status display module indicates the current system status, such as "Acquiring," "Stopped," and "Connecting."

[0030] Furthermore, the light source subsystem is provided with at least two near-infrared light-emitting diodes (LEDs) with different central wavelengths. Each LED is linearly and steplessly dimmed in the range of 0–40 mA via a constant current driver chip. The detector subsystem integrates a transimpedance amplifier after each near-infrared photodiode, which is used to amplify the weak photocurrent by ≥60 dB at the probe end and perform primary filtering. The intermediate adapter board is a multi-layer circuit board, which is used to converge the electrical and signal lines output by the light source subsystem and the detector subsystem, and provide the downstream main control board with light detection electrical signals and I 2 C bus cascade; the main control board is no larger than 45mm × 30mm and is equipped with a 24-bit ΔΣ analog-to-digital converter, a microcontroller unit (MCU), and a low-power wireless communication module. The MCU achieves ±2µs clock synchronization with an external router through the wireless communication module based on the 802.11 timestamp synchronization (TSF) mechanism; the power supply module consists of a lithium battery, microcontroller circuit, and a magnetic interface. At least a 330mF supercapacitor (less than 7mm in diameter and less than 2mm in thickness) is connected in parallel on the main control board to maintain uninterrupted system operation during battery replacement.

[0031] Furthermore, current portable fNIRS devices often use photoelectric probes to transmit analog signals via long-distance wiring harnesses to a main control unit for transimpedance amplification and A / D conversion. For example, many studies have adopted this structure to emphasize the device's portability. However, due to the long transmission lines, the system is highly sensitive to power frequency interference, electromagnetic noise, and motion artifacts, significantly reducing the signal-to-noise ratio (SNR). The detector subsystem disclosed herein integrates a transimpedance amplifier directly after the photodiode on the probe, amplifying the analog signal at the source and performing preliminary filtering. The amplified signal is then transmitted to the main control board via a wiring harness. This significantly enhances anti-interference capabilities, improves the system's signal-to-noise ratio and measurement stability, and reduces the load on the mainboard circuit modules. Furthermore, the disclosed portable near-infrared functional brain imaging system has low manufacturing costs and can be widely adopted.

[0032] In some embodiments, the constant current driver chip and the detector subsystem output signal line are integrated on an intermediate adapter board, and the constant current accuracy is not less than ±1%; the input equivalent noise current density of the transimpedance amplifier is less than 50 fA / √Hz@1MHz, and the gain-bandwidth product is not less than 250MHz.

[0033] Furthermore, a constant-current driver chip is placed directly in close proximity to the LED array, providing a linearly adjustable current of 0–40 mA for the 730 nm and 850 nm dual-wavelength light sources. This avoids the dimming ripple and spectral drift associated with traditional PWM methods, maintaining consistent brightness while reducing system noise and temperature rise. After the light beam penetrates the scalp, InGaAs photodiodes located 30 mm apart convert the reflected light into a microampere current. To suppress noise introduced by long wiring harnesses, a transimpedance amplifier (TIA) is integrated within the detector head. Ultra-low-noise devices such as the OPA1S2384 offer a bandwidth exceeding 250 MHz and an input noise of 50 fA / √Hz at 1 MHz at a gain of 1 MΩ, enabling over 40 dB of amplification and preliminary filtering of the hemodynamic signal at the source.

[0034] Furthermore, if Figure 3 As shown, the probe 100 includes a probe head 100, a probe tail 120, an LED circuit board 130, and a filter 140. The 730nm and 850nm dual-wavelength LEDs are linearly driven by a constant current chip. The 0–40mA stepless current output avoids PWM ripple while maintaining spectral purity. A transimpedance amplifier is mounted next to the detector, amplifying the picoampere or microampere photocurrent by 60dB and performing primary filtering. The amplified millivolt voltage is sent to the main control board via a multi-layer adapter board only 0.5mm thick. This adapter board connects the LED power supply, TIA output, and I 2The C monitoring lines are isolated in layers, shielding against crosstalk and preserving interfaces for subsequent channel expansion. The main control board is compacted to 45mm × 45mm, retaining only the SoC containing a 24-bit ΔΣADC, MCU, supercapacitor, and low-power wireless communication module. The node clock maintains ±2µs accuracy with the router via the 802.11TSF mechanism. Up to 256 devices can operate in parallel on the same base station. With each device sampling at 500Hz and 16 channels, the overall bandwidth is still less than 30Mbps.

[0035] In some embodiments, the thickness of the intermediate adapter plate is no more than 0.5 mm, and the constant current driver chip of the multiple LEDs is embedded therein. 2 C bus feedback to the MCU.

[0036] Furthermore, if existing multi-channel devices can be expanded to 32-channel or even hundreds-channel systems, the main control MCU can be used to output PWM waves to drive MOS tubes to control LED light intensity. This results in dense motherboard components and complex wiring. Although multi-board stacking solutions or modular designs attempt to reduce the pressure on the motherboard, it is still difficult to achieve true miniaturization. The constant current driver chip disclosed in this paper replaces the traditional MCU PWM drive method to achieve smooth and stepless regulation of LED brightness from 0-100%. At the same time, the drive circuit is moved forward to an intermediate adapter board close to the light source probe, which greatly reduces the IO requirements and the number of peripheral discrete components on the main control board, greatly improving the overall volume and lightness.

[0037] In some embodiments, the MCU runs a real-time operating system for scheduling LED dimming, data acquisition, wireless transmission and power monitoring tasks, and performs AES 128 encryption on data frames; the constant current driver chip is used for switching control of multiple light sources and stepless control of their brightness.

[0038] Furthermore, if Figure 4 As shown in the figure, the RTOS concurrently executes tasks such as LED dimming, data acquisition, wireless transmission, power monitoring, and data encryption on the microcontroller unit (MCU) through time slice rotation and priority scheduling.

[0039] Furthermore, the microcontroller unit (MCU) is equipped with a real-time operating system (RTOS), such as FreeRTOS or uC / OS. This system leverages task prioritization and a time-slice rotation method to sequentially schedule multiple tasks, including LED dimming, data acquisition, wireless transmission, and power monitoring. This system utilizes a scheduling strategy that combines time-triggered and event-triggered operations, reducing dimming response times to less than 10ms and ensuring data acquisition cycles accurate to ±1ms. Furthermore, data frames on the SPI or I2C bus are encrypted using the AES-128 encryption algorithm, achieving an encrypted throughput of up to 280kbps, ensuring data transmission security. The constant-current driver chip utilizes PWM dimming technology with a frequency range of 200Hz to 20kHz, enabling stepless brightness adjustment from 0.1% to 100%. Furthermore, its multi-channel independent control architecture supports mixed light operation with different light sources.

[0040] In some embodiments, the driving circuit is moved forward to the intermediate adapter board close to the light source probe, and the IO requirements on the main control board and the number of peripheral discrete components are greatly reduced.

[0041] Furthermore, the driver circuit is integrated into an intermediate adapter board, located only 5-10 cm from the light source probe. This design effectively shortens the signal transmission path and reduces EMI interference by 40%. By communicating with the main control board via the SPI / I2C bus, the drive requirements that originally required 32 GPIOs can now be met with just four communication lines, reducing the number of peripheral components on the main control board by 65% ​​and the PCB area by 30%.

[0042] In some embodiments, the portable near-infrared brain functional imaging system also includes a magnetic connection structure, which separates the main control board from the battery control board. When the main battery is exhausted, the old battery is quickly unplugged and the backup battery is inserted, and the power supply of the main control board can continue.

[0043] Furthermore, if Figure 5 and Figure 6As shown, the host 200 includes a battery 210 and a host control panel 220. A magnetic rectangular block 221 is located at the top of the host control panel 220. The magnetic properties of the rectangular block 221 are opposite to those of the rectangular block at the bottom of the battery 210, and the shapes of the two rectangular blocks match each other. Under the influence of magnetic force, the host control panel 220 and the battery 210 are magnetically attached to each other. Most portable fNIRS devices use an integrated internal battery design, which only lasts for a few hours and requires downtime for charging. This is extremely inconvenient for research and clinical applications requiring multi-hour or long-term brain function monitoring. The power supply module of the present disclosure adopts a magnetically separated design. A 330mF tantalum capacitor array is welded to the main control side to provide ripple filtering. When the battery power drops to a threshold, the old battery can be replaced with a new one within 10 seconds. The capacitors maintain power to the entire device during this period. In addition, the motherboard includes a reserved 120 mF ultra-low ESR tantalum capacitor array, which prompts a replacement request when the main battery charge drops below 20%. Users can remove and insert a backup battery within 1 second without interrupting power, theoretically achieving unlimited battery life. Similar hot-swappable designs have proven viable in medical mobile workstation power supplies, ensuring continuous operation in hospitals and research settings.

[0044] In some embodiments, the contact resistance of the magnetic interface before and after disconnection and re-contact does not change by more than 50 mΩ, and the replacement operation time does not exceed 10 seconds.

[0045] Furthermore, the magnetic terminal maintains a contact resistance fluctuation of less than 50mΩ throughout the entire process, ensuring safe battery replacement and preventing loosening during exercise. The housing is 3D-printed from PA12 nylon powder, with internal positioning pins and an adjustable headband for dual fixation, ensuring the probe's forward and backward movement within 1mm during running or cycling. Slots in the housing allow for air circulation, further reducing heat buildup during extended wear.

[0046] In some embodiments, the portable near-infrared brain functional imaging system further includes a wireless communication module, which supports a 40 MHz channel bandwidth and is configured to operate in the 2.4 GHz and / or 5 GHz frequency bands.

[0047] Furthermore, to save power, mainstream lightweight systems choose BLE connections, but their bandwidth, transmission speed, and number of concurrent connections are limited. In scenarios involving multi-device synchronous data collection, real-time feedback, or group research, BLE becomes a bottleneck. For example, studies have shown that BLE can cause problems such as unstable connections and data freezing in the context of synchronous data collection during music performances. The low-power wireless communication module disclosed in this disclosure can be used to replace Bluetooth transmission to achieve higher data bandwidth and greater concurrent access capabilities. The system can upload collected data to the network in real time and supports simultaneous data collection by hundreds of devices, meeting the needs of large-scale brain function monitoring and multi-device joint use scenarios.

[0048] In some embodiments, the wireless communication module uploads encrypted data to a cloud server to achieve multi-node synchronous brain function monitoring.

[0049] Furthermore, the wireless communication module uses time-slot-based uplink scheduling and data compression transmission strategies to control the system's total uplink data throughput to 30 Mbps or below when up to 64 acquisition nodes upload measurement data simultaneously, thereby reducing power consumption and network delay jitter, thereby improving multi-node signal synchronization accuracy.

[0050] In some embodiments, the housing can be combined with a photoelectrode cap and worn behind the head to keep the optical probe stably fitted to the scalp in mobile scenarios, and the system mass is less than 120g.

[0051] Furthermore, the shell is made of 3D-printed PA12 and supplemented by an adjustable elastic headband, with a total mass of approximately 120g, which can stably fit the scalp in mobile environments such as walking, exercising and in the classroom; the porous heat dissipation and positioning pin structure take into account both comfort and repeatable positioning accuracy.

[0052] In some embodiments, the filtering is based on a combination of Kalman filtering and 0.01 Hz high-pass filtering to remove drift and motion artifacts. In addition to the filtered detector data, the data frame uploaded to the cloud server also contains LED current, battery power, and signal strength information of the wireless communication module for real-time quality control and remote parameter adaptation.

[0053] Furthermore, if Figure 7 As shown, the bandwidth capabilities of Bluetooth Low Energy (BLE) and the low-power wireless communication module of the present invention are compared. The transmission rate of BLE is usually lower than 1 Mbps, while the low-power wireless communication module of the present invention can reach 28 Mbps in actual tests, with a theoretical upper limit of 150 Mbps (based on the 802.11n protocol).

[0054] Furthermore, a high-performance MCU runs a Kalman filter and a 0.01 Hz high-pass drift reduction algorithm, enabling 500Hz multi-channel real-time sampling and noise suppression. A low-power Wi-Fi SoC is also deployed on the main control board, leveraging the 802.11 Transient Switching Frequency (TSF) mechanism to lock node clocks to the master router within ±2 µs, enabling parallel "superscanning" brain network experiments with ≥256 devices. In actual testing, 64 nodes operating at 16 channels at 500 Hz utilized only approximately 28 Mbps of bandwidth, far below the 150 Mbps theoretical limit of 802.11n. In contrast, traditional wearable fNIRS systems using Bluetooth Low Energy (BLE) are significantly limited in terms of connection count and throughput. Both literature and engineering reports indicate that reliable connections to a single device typically exceed ten, and transmission rates fall below 1 Mbps, making them difficult to achieve for high-sampling synchronous acquisition.

[0055] Furthermore, in the signal flow, a constant-current LED emits a pulsed beam, which is scattered by the scalp and brain before returning to the detector. The analog voltage amplified at the source undergoes 24-bit sampling on the main control board. The MCU then runs a Kalman filter and a 0.01Hz high-pass algorithm to remove drift and artifacts, calculating optimized detector data for each channel in real time. The encrypted data packets are uploaded to the cloud via a wireless communication module. Researchers can view brain blood oxygen waveforms in computer software or download the raw frames for offline network analysis. The cloud transmits control commands for the LEDs in each channel, and the MCU automatically adjusts gain and transmit power based on this feedback, creating a closed-loop quality control system.

[0056] Furthermore, compared to the approximately 21dB dynamic SNR achieved by traditional long-wire solutions in running switching scenarios, this solution achieves an average of 37dB, a more than fourfold improvement. The number of components and PCB area on the main control board are reduced by 35% and 28%, respectively, and the overall weight is reduced from 155g to 118g. This higher signal-to-noise ratio and unlimited battery life lay the hardware foundation for large-scale, multi-time applications such as synchronized classroom learning, sports fatigue monitoring, and remote post-operative brain function assessment.

[0057] Furthermore, if Figure 8 As shown in the figure, the dynamic signal-to-noise ratio (SNR) performance of a traditional long-wire architecture is compared with that of the disclosed technical solution in different test scenarios (stationary, walking, running, and high noise). The disclosed technical solution demonstrates significantly higher SNR in all scenarios, especially in dynamic and high-noise environments, far exceeding the 35 dB required for high-quality cerebral oximetry. Notably, except in the extremely high-noise scenario, where it slightly fell below 35 dB, it still far outperforms the traditional solution.

[0058] Furthermore, by moving the transimpedance amplifier (TIA) and constant-current LED driver to the probe end, the weak photocurrent undergoes ≥40dB gain amplification and primary filtering before leaving the probe. In high-artifact scenarios such as walking and running, we conducted comparative tests on 10 subjects: the dynamic SNR of the traditional long-wiring structure was approximately 21dB, while the solution of the present invention achieved an average of 37dB, an improvement of approximately 4.2 times. After the amplification and dimming circuits were "relocated", the main control board only retained the ADC, MCU, Wi-Fi SoC, and buffer capacitors, reducing the component count by approximately 35% and the PCB area by approximately 28%. The weight of the entire device without the shell was reduced from 155g to approximately 120g, significantly reducing the pressure on the head when worn for long periods of time.

[0059] Furthermore, the portable near-infrared brain function imaging system disclosed in the present invention has the characteristics of high anti-interference, high precision, hot-swappable power supply and low-latency wireless synchronization. It is directly applicable to application scenarios such as cognitive monitoring and educational science, clinical evaluation and brain function monitoring, support for neurology, psychiatry, intensive care, intraoperative monitoring, etc., brain blood oxygen tracking, brain-computer interface (BCI) applications, wearable and mobile environment research, etc. At the same time, it also has certain application space in the fields of large-sample group brain science research, remote health and psychological intervention, wearable metabolic and physiological joint monitoring, long-term brain dynamic tracking in family and educational scenarios, etc.

[0060] The core concept of this disclosure is to complete the light source driving and signal amplification within the probe itself, pre-packing the most noise-sensitive and space-consuming circuits into the flexible probe. This creates a high-gain shield for the weak photocurrent right at the source, while also significantly reducing the size of the main control board. This involves moving the noise-sensitive amplification process to the front end of the probe, allowing the signal to be "armored" before routing. This also reduces the size of the main control board, resulting in lightweight wear, stable operation, and clean data. Current portable fNIRS devices transmit the weak photoelectric signal, which is only at the picoampere level, from the probe via a 30-40 cm flexible cable to the main control board before amplification and dimming are performed. This "naked" long-distance analog transmission acts like a portable antenna, absorbing all noise generated by 50 / 60 Hz power frequencies, electromagnetic radiation, and head movement. This results in a dynamic signal-to-noise ratio (SNR) often dropping to around 20dB, making it difficult to obtain usable cerebral blood oxygen waveforms when walking or running outdoors. To improve these fragile signals, traditional solutions rely on stacking bulky amplifiers, MOS tubes, and PWM dimming circuits on the main control board. This significantly increases the PCB area, component count, and heat dissipation requirements. The weight of the entire device often exceeds 150g, making it tiring and prone to heat when worn for long periods of time. This series of issues, including high noise, heavy equipment, and dirty signals, ultimately forced researchers to perform extensive offline data cleaning and even abandon the use of fNIRS in real-world sports scenarios. The core technical challenge lies in achieving high-gain, low-noise amplification at the signal source, "armoring" the signal before routing, while also reducing the size of the main control board, fundamentally improving anti-interference capabilities and achieving a lighter, more reliable wearable experience.

[0061] Furthermore, for portable fNIRS to truly move beyond the laboratory and into dynamic scenarios such as sports training, classroom teaching, and remote rehabilitation, it must first address the paradox of weak signal transmission over long cables combined with the burden of head-mounted wear. The entire device consists of a light source subsystem, a detector subsystem, an intermediate adapter board, a main control board, a hot-swappable power supply module, and a 3D-printed housing. The mass-produced version weighs less than 120g, ensuring users experience no noticeable pressure or heat even when worn for extended periods. Furthermore, the embedded firmware uses an RTOS to manage multiple tasks, including light source dimming, signal acquisition, wireless communication module transmission, and power monitoring. Data frames (including filtered detector channel data, LED current, and RSSI) are uploaded to a cloud database after AES-128 encryption. Researchers can view cerebral blood oxygenation waveforms in real time in a browser or import them into MATLAB / Python toolboxes for brain network, synchronization, and fatigue analysis. This overall solution comprehensively overcomes existing technical bottlenecks in portable fNIRS in terms of interference rejection, miniaturization, battery life, and multi-device synchronization, providing a reliable hardware foundation for long-term, high-density, and large-scale group brain function research.

[0062] Furthermore, the portable near-infrared brain functional imaging system disclosed in the present invention adopts a modular hardware architecture and is composed of seven major components: a light source subsystem, a detector subsystem, an intermediate adapter board, a main control board, a power supply module, a wireless communication module, and a shell and a wearing mechanism. The light source subsystem has a built-in 730 nm and 850 nm dual-wavelength LED array, and the constant current driver chip is directly mounted on the flexible light board, which can achieve linear stepless dimming in the range of 0-40 mA. The detector subsystem integrates a transimpedance amplifier (gain 1 MΩ, input noise <50 fA / √Hz) behind each InGaAs photodiode, so that the weak photocurrent is amplified and preliminarily filtered inside the probe. The two subsystems are connected by a six-layer flexible adapter board with a thickness of only 0.5 mm. In addition to distributing the LED current and detector signal, the board also integrates LED drive, analog signal transfer interface and I 2 The C-bus cascade interface maximizes space on the main control board. The main control board, measuring 45 mm × 30 mm, utilizes a high-performance MCU architecture to simultaneously control a 24-bit delta-sigma ADC, LED current closed-loop control, and data cache. It also utilizes an on-chip low-power Wi-Fi SoC for millisecond-level clock synchronization and encrypted transmission. The power supply module consists of a 3.7 V / 1500 mAh lithium battery and a magnetic hot-swappable connector. A 330 mF ultra-compact supercapacitor is located on the mainboard, enabling rapid, uninterrupted battery replacement. The wireless communication module utilizes the 802.11n protocol and can simultaneously manage over 256 nodes at 40 MHz bandwidth. The entire device is housed in a 3D-printed PA12 housing with an adjustable elastic headband and positioning pins. The unit weighs approximately 120 g, ensuring a secure fit and excellent heat dissipation for everyday use or exercise. Each component is plug-and-play through blind-plug FPC connectors and magnetic interfaces, which not only facilitates maintenance and upgrades, but also reserves sufficient interfaces for subsequent channel expansion and multimodal fusion (such as EEG and ECG).

[0063] The operating process of the portable near-infrared brain function imaging system disclosed herein is as follows: when in use, first wear the system on the head, and connect the main control board and the battery board through the magnetic interface to ensure stable power supply. After startup, the MCU controls the light source to emit near-infrared light, and the detector receives the transmitted signal and completes signal amplification and acquisition. The collected data is uploaded to the cloud server via the wireless module to realize real-time brain function monitoring. When the battery is low, the system will emit sound and light prompts, and the user can quickly replace the battery without interrupting the system.

[0064] Finally, it should be noted that: in the various embodiments of the present application, if the form of connection is not clearly defined, the form of connection may be a detachable connection form such as bolts and nuts, screws, snaps, magnets, etc. If there is no special requirement for the form of non-detachable matching in some connections, non-detachable connections may be made by welding, bonding, gluing, etc. In addition, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them; although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein; and these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present application.

Claims

1. A portable near-infrared brain function imaging system, comprising a light source subsystem, a detector subsystem, an intermediate adapter board, a main control board, a power supply module and a housing, characterized in that , a transimpedance amplifier is directly integrated after the photodiode on the probe of the detector subsystem, which amplifies the analog signal at the source and performs preliminary filtering. The amplified signal is then transmitted to the main control board through a wiring harness.

2. A portable near-infrared brain function imaging system according to claim 1, characterized in that: The constant current driver chip and the detector subsystem output signal line are integrated on the intermediate adapter board, and the constant current accuracy is not less than ±1%; The input equivalent noise current density of the transimpedance amplifier is less than 50 fA / √Hz@1MHz, and the gain-bandwidth product is not less than 250MHz.

3. The portable near-infrared brain function imaging system according to claim 1, characterized in that: The thickness of the intermediate adapter plate is no more than 0.5 mm, and the constant current driver chip of the multi-channel LED is embedded, and the constant current driver chip is connected to the intermediate adapter plate through I 2 C bus feedback to the MCU.

4. A portable near-infrared brain function imaging system according to claim 3, characterized in that: The MCU runs a real-time operating system for scheduling LED dimming, data acquisition, wireless transmission, and power monitoring tasks, and performs AES 128 encryption on data frames. The constant-current driver chip is used for switching and brightness control of multiple light sources. The drive circuit is moved forward to the intermediate adapter board, which is close to the light source probe. The IO requirements and the number of peripheral discrete components on the main control board are greatly reduced.

5. The portable near-infrared brain function imaging system according to claim 1, characterized in that: It also includes a magnetic connection structure, which separates the main control board from the battery control board. When the main battery is exhausted, the old battery can be quickly unplugged and a backup battery can be inserted to continuously supply power to the main control board. The contact resistance of the magnetic interface before and after disconnection and re-contact does not change by more than 50m ohms, and the replacement operation time does not exceed 10 seconds.

6. The portable near-infrared brain function imaging system according to claim 1, characterized in that: It also includes a wireless communication module, which supports 40 MHz channel bandwidth and is configured to operate in the 2.4 GHz and / or 5 GHz frequency bands; the wireless communication module uploads encrypted data to a cloud server to achieve multi-node synchronous brain function monitoring.

7. The portable near-infrared brain function imaging system according to claim 1, characterized in that: The shell can be combined with a photoelectrode cap and worn behind the head to keep the optical probe stably fitted to the scalp in mobile scenarios. The system mass is less than 120g.

8. A portable near-infrared brain function imaging system according to claims 1 to 7, characterized in that: The filtering is based on a combination of Kalman filtering and 0.01Hz high-pass filtering to remove drift and motion artifacts. The data frame uploaded to the cloud server not only contains the filtered detector signal data, but also includes LED current, battery power and signal strength information of the wireless communication module for real-time quality control and remote parameter adaptation.

Images