Biomagnetic measurement system for sensing biomagnetic signal

Abstract

A biomagnetic sensor includes an integrated active noise cancellation unit that cancels ambient magnetic noise from a detection signal obtained from a tunneling magnetoresistance (TMR) sensor unit using a gradiometer, the detection signal being indicative of a magnetic field proximate biological tissue. Thus, ambient magnetic noise may be cancelled in real time at the sensor front end (e.g., as part of the sensor body itself circuitry). The active noise cancellation techniques presented herein may be sufficiently effective to enable biomagnetic sensors to be used in unshielded environments (e.g., environments affected by the Earth's magnetic field), thereby significantly extending the potential use of the sensors.

Description

Technical Field

[0001] The present invention relates to a system for measuring biomagnetism, and more particularly but not limited to a wearable device for sensing biomagnetic signals. The system may include a magnetic field sensor chip utilizing a tunnel magnetoresistance (TMR) array and active on-chip noise cancellation technology. Background Art

[0002] As is well known, muscle activity can be recorded using electrical sensors by surface (on the skin) or depth (within muscle tissue) electromyography (EMG). A major challenge faced by EMG is the relatively low spatial resolution of the sensors, making it difficult to selectively record individual motor units within the muscle. Intramuscular electromyography (inserting needle electrodes into muscle tissue) can be used to improve spatial resolution, but this causes pain and impairs muscle function. Additionally, during long-term implantations such as in sports rehabilitation, the interface between the metal contacts of the sensor and human tissue changes over time, leading to infections and body rejection.

[0003] With the rapid development of micro-nano technologies, non-invasive biomagnetic assessment has emerged as an alternative to EMG. Biomagnetic signals have the same temporal resolution as the corresponding bioelectrical signals, but much higher spatial resolution. During recording, the induced magnetic field also does not require electrical contact, so the sensor can be fully encapsulated in a biocompatible material, thus minimizing the risk of infection [6] .

[0004] However, there are still technical challenges in providing practical devices capable of detecting biomagnetic signals, mainly because the amplitude of biomagnetic signals is very small. Although high temporal resolution can be achieved, the achievable spatial resolution is limited due to limitations such as the number of sensors, motion artifacts, the inherently low signal-to-noise ratio (SNR), and relatively high static and dynamic background magnetic noise [9] .

[0005] In the development of sensors for detecting minute biomagnetic fields, the initial attempts were to explore the use of superconducting quantum interference devices (SQUIDs)

[10] and optically pumped magnetometers (OPMs) [5] . However, both of these methods have limitations because they require the use of large magnetic shielding rooms. These methods are bulky, costly, energy-consuming, and require a temperature-controlled environment.

[0006] Recently, spintronic sensors based on the magnetoresistance (MR) effect have revolutionized biomagnetic sensing methods due to their full compatibility with traditional silicon technologies. These sensors can be integrated with readout circuits into a standard CMOS process on a substrate with a diameter of sub-millimeters, ultimately enabling on-chip signal conditioning, including amplification, filtering, noise, and drift cancellation

[11] This phenomenon has prompted the development of MR sensors with ultra-high sensitivity, gradually replacing traditional thin-film magnetotransport devices such as Hall sensors.

[12] It has the potential to detect magnetic fields in the pico-Tesla range and is suitable for the bio-magnetic signal level.

[0007] In addition, the miniaturized MR sensor area can improve the resolution of fields with relatively small distance changes. The closer the sensor is to the nerve source, the stronger the signal. Therefore, MR sensors are suitable for array applications with lower power requirements. Recently, giant magnetoresistance (GMR) sensors have been used to record weak bio-magnetic signals. However, the sensitivity of GMR sensors is in the nano-Tesla range, so averaging is required to improve the signal-to-noise ratio. In the past decade, using tunnel magnetoresistance (TMR) sensors, it has been possible to achieve sensing of magnetic field strengths at the pico-Tesla / √Hz level. Such sensors are highly miniaturized and can be operated at room temperature through sensor arrays.

[0008] Various potential applications of bio-magnetic sensors have been confirmed, from clinical diagnosis to human-computer interaction. [1] However, to detect weak bio-magnetic fields from human active organs and tissues, such as magnetocardiogram (MCG) [2][3] , magnetoencephalogram (MEG) [4][5] , magnetomyogram (MMG) [6][7] , magnetoneurogram (MNG) [8] , effective methods with both high spatial resolution and high temporal resolution are required. SUMMARY OF THE INVENTION

[0009] Most generally, the present invention provides a bio-magnetic sensor that includes an integrated active noise cancellation unit that uses a gradiometer to cancel environmental magnetic noise from a detection signal obtained from a tunnel magnetoresistance (TMR) sensor unit, the detection signal indicating a magnetic field adjacent to biological tissue. Thus, environmental magnetic noise can be cancelled in real time at the front end of the sensor (e.g., as part of the circuit of the sensor body itself). The active noise cancellation technique proposed herein may be effective enough such that the bio-magnetic sensor can be used in a non-shielded environment (e.g., an environment affected by the Earth's magnetic field), thereby significantly expanding the potential uses of the sensor.

[0010] In a first aspect, the present invention provides a biomagnetic sensor module, comprising: a tunneling magnetoresistance (TMR) sensor unit configured to output a detection signal indicative of a magnetic field adjacent to biological tissue; a gradiometer unit configured to output a background signal indicative of environmental magnetic noise; a shielding element located between the TMR sensor unit and the gradiometer unit; an active noise cancellation unit configured to remove the background signal from the detection signal to generate a biomagnetic signal; an analog readout circuit configured to receive the biomagnetic signal and perform signal conditioning to generate an analog output; and an analog-to-digital converter (ADC) configured to generate a digital output signal from the analog output.

[0011] The term "biomagnetic" herein refers to a magnetic field generated by electrical activity in biological tissue (such as neural tissue or muscle tissue). Generally, such a magnetic field is very weak, and thus it can be understood that a biomagnetic sensor is a device capable of detecting such a weak magnetic field, for example, having a sensitivity in the pico-Tesla range, such as a sensitivity equal to or less than 50 pT / √Hz, preferably equal to or less than 20 pT / √Hz.

[0012] The TMR sensor unit may include a plurality of TMR sensors, each TMR sensor having an array of magnetic tunnel junctions fabricated on a substrate. The magnetic tunnel junctions can be fabricated using known nano-scale techniques such that the array has a small footprint, such as equal to or less than 12 mm 2 .

[0013] The plurality of TMR sensors in the TMR sensor unit may include four TMR sensors arranged in a Wheatstone bridge configuration.

[0014] The gradiometer unit may include a triaxial gradiometer. The advantage of this arrangement is that, regardless of the orientation of the sensor module, the background signal can provide a consistent indication of the environmental magnetic noise.

[0015] The gradiometer unit may use the same sensing mode as the TMR sensor unit. That is, the gradiometer unit may include a plurality of TMR sensors configured to detect environmental magnetic noise. The plurality of TMR sensors may be configured in the same manner as the TMR sensors in the TMR sensor unit. For example, the gradiometer unit may include four TMR sensors arranged in a Wheatstone bridge configuration. Providing the same sensor configuration in the gradiometer unit and the TMR sensor unit allows the background signal to be directly compared with the detection signal.

[0016] The biomagnetic sensor module may include a plurality of TMR sensor units, wherein each TMR sensor unit is arranged to detect a magnetic field adjacent to biological tissue and output a biomagnetic signal on a corresponding channel, and wherein the biomagnetic sensor module further includes a multiplexer configured to selectively couple each corresponding channel to the analog readout circuit.

[0017] The biomagnetic sensor module may include a controller configured to control the operation of the module, such as a microprocessor unit or the like.

[0018] The active noise cancellation unit may include a comparator, such as a differential amplifier or an equivalent device, which is configured to receive the background signal and the detection signal as inputs, so as to subtract the background signal from the detection signal to generate a biomagnetic signal. Thus, the active noise cancellation unit can directly act on the outputs of the TMR sensor unit and the gradiometer unit, which helps to perform real-time cancellation before further signal processing.

[0019] The biomagnetic sensor module may be configured to bring the TMR sensor unit close to the measurement position on a human or animal body. For example, the biomagnetic sensor module may include a skin contact interface made of a biocompatible material. The skin contact interface may form the outer surface of the biomagnetic sensor module, such as the bottom. The TMR sensor unit may be arranged on the skin contact interface as close as possible to the outer surface. The skin contact interface may include a heat insulation layer arranged between the TMR sensor unit and the outer surface. The heat insulation layer may inhibit heat conduction, so that the internal components of the biomagnetic sensor module (especially the TMR sensor unit and the gradiometer unit) maintain a stable temperature.

[0020] The gradiometer unit may be located at a position farther from the skin contact interface than the TMR sensor unit. This configuration can avoid the background signal of the gradiometer unit being affected by the magnetic field generated in biological tissues. In one embodiment, the gradiometer unit is more than 5 mm, preferably more than 10 mm, farther from the outer surface than the TMR sensor unit.

[0021] The function of the shielding element located between the TMR sensor unit and the gradiometer unit is to further reduce the influence of the magnetic field generated in biological tissues on the background signal. The shielding element may be a magnetic conductive material, such as permalloy. The shielding element may be in the form of a foil or other sheet material, such as a layer separating the TMR sensor unit and the gradiometer unit. The shielding element can minimize the physical separation between the TMR sensor unit and the gradiometer unit, such as being equal to or less than 10 mm, without affecting the measurement accuracy or sensitivity.

[0022] The biomagnetic sensor module may further include a compensation unit configured to minimize a baseline mismatch between the TMR sensor unit and the gradiometer unit, which may be caused jointly by an inherent mismatch between sensors of the TMR sensor unit and the gradiometer unit and the influence of the shielding element. The "baseline" mentioned here refers to the measured value obtained in the absence of a target signal, i.e., when the TMR sensor unit is not placed on the skin. The compensation unit may include a feedback circuit configured to receive the baseline outputs from the TMR sensor and the gradiometer unit and adjust the gradiometer unit to minimize the baseline output. In one embodiment, the compensation unit may include an adjustable current source configured to generate a bias signal for the gradiometer unit, wherein the bias signal is adjustable to minimize the baseline output. In another embodiment, the compensation unit may include a variable gain amplifier connected between the gradiometer unit and the active noise cancellation unit, wherein the gain of the variable gain amplifier is adjustable to minimize the baseline output.

[0023] In one embodiment, the biomagnetic sensor module may have a layered structure including a plurality of stacked functional layers, such as having a skin contact interface layer at its bottom. The TMR sensor unit may be arranged in the main sensing layer, while the gradiometer unit may be arranged in a secondary sensing layer parallel and spaced apart from the main sensing layer. In this context, the above-mentioned shielding element may be provided in the layer between the main sensing layer and the secondary sensing layer. The active noise cancellation unit and the analog readout circuit may be arranged in one or more layers parallel and spaced apart from the secondary sensing layer. Other components (such as a controller, an ADC, and a power management unit) may be arranged in one or more other layers. All layers may be enclosed by a housing or a casing or held together in other ways.

[0024] The analog readout circuit may be configured to eliminate input offset and low-frequency flicker noise. For example, the analog readout circuit may include a band-pass filter and a common-mode feedback circuit.

[0025] The biomagnetic sensor module may further include a wireless communication unit configured to transmit the digital output signal to a remote device. The remote device may be a smartphone, a tablet computer, a laptop computer, or a desktop device. The wireless communication unit may be configured to use any suitable wireless protocol (such as etc.). The wireless communication unit may also receive information from the remote device, for example, to operate the control module or update its firmware.

[0026] On the other hand, the present invention provides a wearable biomagnetic sensing device, which includes the above-mentioned biomagnetic sensor module. The wearable biomagnetic sensing device can operate in a non-shielded environment. The wearable biomagnetic sensing device can form part of a measurement system, which also includes a remote computing device that communicates with the biomagnetic sensor module, for example, for receiving digital output signals.

[0027] In one embodiment, the wearable biomagnetic sensing device can include a plurality of the above-mentioned biomagnetic sensor modules, which are fixed to a holding element that can be mounted on the human body. The fixing element can be an elastic band so that the device can be fixed on the user's arm or leg.

[0028] The present invention includes combinations of the aspects and preferred features described herein, unless such combinations are clearly not permitted or should be explicitly avoided. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Embodiments and experiments illustrating the principles of the present invention will now be discussed with reference to the accompanying drawings, in which:

[0030] Figure 1 is a schematic diagram showing the hierarchical structure of a biomagnetic sensor module according to an embodiment of the present invention;

[0031] Figure 2 is a schematic diagram showing the components of a sensor array of a biomagnetic sensor module that can be used for Figure 1 ;

[0032] Figure 3 is a perspective view of a wearable biomagnetic sensing device according to an embodiment of the present invention;

[0033] Figure 4 is Figure 1 a schematic diagram of the processing flow between the functional components of a biomagnetic sensor module;

[0034] Figure 5 is a schematic diagram showing Figure 1 the configuration of a gradiometer in a biomagnetic sensor module;

[0035] Figure 6 is a graph showing the effect of a gradiometer compensation technique;

[0036] Figure 7 is a schematic diagram showing Figure 1 the baseline compensation technique of a biomagnetic sensor module that can be used for

[0037] Figure 8 is a schematic diagram showing Figure 1 the readout circuit architecture of a biomagnetic sensor module that can be used for

[0038] Figure 9A andFigure 9B is an analog graph showing the relative amplitudes of the MMG and the noise magnetic signal processed by the biomagnetic sensor module;

[0039] Figure 10 shows a cross-section of a muscle tissue diagram used in a system for simulating biomagnetic potential;

[0040] Figure 11A shows a cross-section of a muscle tissue diagram, showing the simulated electric field potential in the muscle tissue in the absence of a fat layer;

[0041] Figure 11B shows Figure 11B a cross-section of the muscle tissue diagram in, showing the simulated magnetic potential;

[0042] Figure 11C shows a cross-section of a muscle tissue diagram, showing the simulated electric field potential in the muscle tissue and a layer of fat around the muscle bundle;

[0043] Figure 11D shows Figure 11C a cross-section of the muscle and fat tissue diagram in, showing the simulated magnetic potential;

[0044] Figure 12 is a graph showing the change in the simulated electric field potential along lines A-A and C-C in Figure 11A and Figure 11C ;

[0045] Figure 13 is a graph showing the change in the simulated magnetic potential along lines B-B and D-D in Figure 11B and Figure 11D ; and

[0046] Figure 14 is a schematic diagram of the change in the biomagnetic potential in the muscle tissue with its distance from the source. DETAILED DESCRIPTION

[0047] Aspects and embodiments of the present invention will now be discussed with reference to the accompanying drawings. Further aspects and embodiments will be apparent to those skilled in the art. All documents mentioned herein are incorporated herein by reference.

[0048] The present invention relates to a biomagnetic measurement system, and more particularly to a wearable biomagnetic sensing device and a biomagnetic sensing module used therein. The purpose of the biomagnetic measurement system is to detect or estimate the biomagnetic potential related to muscle activity in a non-invasive manner at room temperature. The muscle conditions in the areas controlling limb movement can be better understood through such a system.

[0049] As described in detail below, the biomagnetic sensing module provides on-chip, real-time noise cancellation functionality, which can reduce or eliminate noise from the environment (such as the Earth's magnetic field, thermal noise, and 1 / f noise) by combining multiple device and circuit technologies. Using these technologies, the biomagnetic sensing module discussed in this article can achieve non-invasive recording of magnetobiomedical signals from neurons, nerves, and muscles at room temperature for the first time without external shielding.

[0050] Generally speaking, the biomagnetic sensing devices discussed in this article include one or more biomagnetic sensing modules, each of which includes multiple sensors, and each sensor in turn has multiple tunnel magnetoresistance (TMR) elements configured in an array, for example, fabricated on a silicon substrate using conventional CMOS microfabrication techniques. Each sensing module may also include an integrated signal processing package, which includes analog readout circuitry configured to generate an analog output that is transmitted to an analog-to-digital converter (ADC). The ADC is connected to a microcontroller, which includes a wireless communication module for transmitting the digital output signal to a remote device for analysis. The sensors in each module can be configured to output signals on each of multiple channels. The analog readout circuitry can be configured into multiple module units (one for each channel), and these module units can be connected to the ADC through a multiplexer. As described below, the integrated signal processing package provides a noise cancellation function, which can ensure that the detected biomagnetic potential is correctly displayed in the output signal.

[0051] Figure 1 FIG. 7 is a schematic diagram showing the hierarchical structure of a biomagnetic sensor module 100 according to an embodiment of the present invention. The biomagnetic sensor module 100 is configured to be used with a wearable biomagnetic sensing device, which is designed to be worn on the area of the user's skin where measurements are needed.

[0052] The biomagnetic sensor module 100 in this example is configured with a hierarchical structure, and each layer is parallel to the skin surface when worn. This hierarchical structure includes a skin interface 102, which may include a biocompatible material or coating on the bottom surface of the sensor module 100. The terms "top" and "bottom" related to the biomagnetic sensor module 100 in this article can be understood in relation to the expected orientation of the module relative to the skin surface when in use. That is, when in use, the bottom of the module is closer to the skin surface than the top of the module. The skin interface 102 can facilitate the easy installation and removal of the device at various positions without applying or removing gels and adhesives. In addition, by using biocompatible materials, a comfortable fit can be ensured. Examples of suitable biocompatible materials include nanocellulose, biocompatible polymers, or e-skin.

[0053] The skin interface 102 may also include a thermal insulation layer, such as made of polyurethane, etc., for isolating the components of the sensor module 100 from the heat generated by the body when the user wears the device. This can ensure that the sensor array inside the device operates at a consistent temperature, thus avoiding introducing errors caused by thermal variations.

[0054] Adjacent to the skin interface 102 is the magnetic induction sub-structure 103, which includes three layers: the main sensing layer 104, the shielding layer 106, and the secondary sensing layer 108. Both the main sensing layer 104 and the secondary sensing layer 108 include a plurality of tunnel magnetoresistance (TMR) sensors, as described below. The shielding layer 106 is used to magnetically isolate the main sensing layer 104 from the secondary sensing layer 108. The shielding layer 106 is preferably formed of a magnetically permeable material, such as permalloy. In one embodiment, the shielding layer 106 may be formed of formed, with a composition of 80% nickel, 4.5% molybdenum, and the balance iron, to provide a high magnetic susceptibility. As discussed in more detail below, the secondary sensing layer 108 is configured as a gradiometer to measure the background magnetic potential, which can be subtracted from the signal detected by the main sensing layer 104 as part of the noise cancellation function provided by the sensor module 100. The role of the shielding layer 106 is to shield the biological magnetic field of the secondary sensing layer 108. According to the inverse square law, the intensity of the biological magnetic field will naturally weaken as the distance from the skin surface increases. Since the secondary sensing layer 108 is farther from the skin interface 102 than the main sensing layer 104, the biological magnetic field signal it receives during use will be weaker. However, the shielding layer 106 enhances this isolation and can ensure that part of the biological magnetic signal is not accidentally cancelled when subtracting the background magnetic potential measured by the gradiometer from the signal detected by the main sensing layer 104.

[0055] Figure 2 is a schematic diagram showing the components of the sensor unit 204 that can be used for the main sensing layer 104 and the secondary sensing layer 108. In this example, the sensor unit 204 is configured with four sensors 206 in a Wheatstone bridge structure to provide a signal on the first channel. The main sensing layer 104 and the secondary sensing layer 108 of the sensor module 100 may be configured with multiple channels (e.g., 8 or more), and the output of each channel comes from an independent sensor unit 204 installed in the relevant layer. Therefore, both the main sensing layer 104 and the secondary sensing layer 108 may include the components of the sensor unit 204.

[0056] Each sensor 206 in the sensor unit 204 includes an array 208 of TMR sensor elements 210. Each TMR sensor element 210 in the array is a magnetic tunnel junction, including two layers of ferromagnetic materials 212, 216, separated by a very thin insulating layer 214. The top layer 212 is defined as the free layer because its magnetization direction can be freely changed, while the bottom layer 216 is called the fixed layer because its magnetization direction is fixed during the manufacture of the sensor. The sensor is configured to allow the tunneling effect, where electrons can cross the insulating layer 214 under specific conditions, which in turn causes the structure to exhibit spin-dependent magnetoresistive properties at room temperature.

[0057] The response of the TMR sensor corresponds to the change in the resistance across the device as the magnetic field changes. For biomagnetic measurements, it is desirable for the response to be linear and hysteresis-free. Generally, the best noise performance can be obtained using large arrays of large-area sensors. In the example discussed herein, 1102 TMR sensor elements 210 are connected in series to form 38 rows and 29 columns to minimize the 1 / f noise of the sensors. Each TMR sensor element 210 consists of the following stack (nm): 5Ta / 25CuN / 5Ta / 5Ru / 20IrMn / 2CoFe 30 / 0.85Ru / 2.6CoFe 40 B 20 / 1MgO[9kΩ·μm 2 / 2CoFe 40 B 20 / 0.21Ta / 4NiFe / 0.20Ru / 6IrMn / 2Ru / 5Ta / 10Ru. The size of each TMR element is 100x100μm. The size of the array 208 is 6x4mm, which means that the footprint of the sensor unit 204 can be controlled to not exceed 20mm 2 . Each array 210 on the sensor unit 204 is electrically connected to a corresponding electrode pad 218, through which it is connected to the rest of the electronics.

[0058] A Wheatstone bridge structure is employed to minimize temperature drift and nullify the output signal in the absence of any applied magnetic field. In the example discussed herein, four sensors 206 of the above type are arranged in a full Wheatstone bridge configuration. When the bias current is 20mA, the measurement linear range of the sensor is approximately -1Oe to 1Oe. With a full-bridge setup, the measured resistance change of each TMR sensor is 280Ω·μm 2 / Oe. Thus, for 1102 elements with an area of 100x100μm 2 , its sensitivity is calculated to be approximately 0.617V / Oe.

[0059] Back to Figure 1, Module 100 further includes a signal processing circuit 109 disposed above the magnetic induction sub-structure 103. The signal processing circuit 109 is configured to perform on-device processing on the signals received from the sensor unit. The signal processing circuit 109 includes an active noise cancellation unit 111, an analog readout circuit 110, and associated control circuits 112, which may be arranged in respective layers or may be combined in one or more layers. These layer functions will be discussed in more detail with reference to Figures 5 to 7 below.

[0060] Module 100 further includes a wireless communication unit 114, which may be configured to communicate with a remote computer to send or receive data from the module. For example, module 100 may be configured to transmit the post-processed output signal from the sensor unit to an external device for further analysis.

[0061] Module 100 further includes a power management unit 116, which may include a battery (e.g., a lithium-ion battery) and associated control circuits.

[0062] Figure 3 An example of a wearable biomagnetic sensing device 300 is shown, which includes a plurality of the above-described biomagnetic sensor modules 100. Each module 100 is encapsulated in a corresponding protective housing that has a biocompatible bottom surface through which the sensor unit 204 can detect biomagnetic signals. In this embodiment, eight modules 100 are interconnected into a ring by an elastic band 302. This configuration may be suitable for wearing on a user's wrist. The elastic band 302 allows the device 300 to be easily installed, for example, by sliding it up and down the wrist, while also ensuring that the device remains in place during use, thereby ensuring data collection with minimal motion artifacts. In other embodiments, the device 300 may be used in combination with a motion sensor, thereby allowing the measured biomagnetic data to be combined with motion information.

[0063] Figure 4It is a schematic diagram showing the processing flow among the functional components of the above-mentioned biomagnetic sensor module 100 and the communication between the module 100 and the external computing system 150. In this embodiment, the control circuit 112 of the module 100 includes a microprocessor unit, which, under the control of the power and amplitude calculation modules 136, 138, interacts with the magnetic induction sub-structure 103 and the analog readout circuit through the corresponding voltage regulators 120, 122, so as to ensure that they receive a stable power supply. As described above, the magnetic induction sub-structure 103 includes a main sensing layer, which provides signals from one or more sensor units, and each sensor unit has four TMR sensors arranged in a Wheatstone bridge. The control circuit allows real-time reading of the signals from the sensor units. The Wheatstone bridge preferably operates in voltage mode, but the module can also be configured to selectively operate in voltage mode or current mode, for example, by using a selector actuator (such as a toggle switch). For a given resistance change, the sensitivity in current mode is twice that in voltage mode. In addition, the output of the sensor preferably exhibits stability within a temperature range commensurate with the expected operating conditions (such as -20 °C to 50 °C). Integrating the TMR sensors into a complete Wheatstone bridge can provide a zero voltage output in the absence of an external excitation field, while ensuring that each device in the arrangement outputs a complete signal that can be used in a differential amplifier.

[0064] As described above, the magnetic induction sub-structure 103 includes a gradiometer for measuring the dynamic magnetic background noise, which can be subtracted from the signals measured by the main sensing layer. Conventionally, in order to reduce noise sources (such as acoustic noise from the earth and surrounding devices and interference from magnetic and electric fields), the magnetic sensing system must be operated in a magnetic shielding environment. In the present invention, using the gradiometer to compensate for the dynamic magnetic background noise provides the device with a sensitivity (~20 pT) comparable to previous designs, but a higher dynamic range (target ~50 pT), without the need for magnetic shielding.

[0065] Figure 5It is a schematic diagram showing the configuration of the gradiometers in the active noise cancellation unit 111. The secondary sensing layer 106 includes a plurality of TMR sensor units of the same type as the primary sensing layer. However, the TMR sensor units in the secondary sensing layer 106 are not configured to detect the magnetic field on the skin surface, but are configured as one or more gradiometers to record the dynamic magnetic background noise. In a preferred embodiment, three gradiometers are arranged in a three-axis configuration, where a plurality of TMR sensors are integrated into the micro Wheatstone bridge configuration of each axis of the gradiometer. The signals of these gradiometers are combined at the pickup sensor to provide a background signal independent of the device orientation. The background signal is provided to the inverting input of the comparator 160 (differential amplifier), where it is subtracted from the signal provided by the primary sensing layer 104. The output of the comparator 160 is sent to the analog readout circuit 110. Thus, real-time compensation of the dynamic magnetic background noise occurs at the front end of module 100. This greatly improves the system's tolerance to large DC geomagnetic fields and AC environmental noise.

[0066] Figure 6 The experimental result graph of using the three-axis gradiometer in the above manner is shown. The proposed active noise cancellation technique is based on the phase cancellation principle, where the background magnetic noise 164 is recorded, inverted to generate "anti-noise", and then added to the output signal 162 of the primary sensing layer, which includes the desired biomagnetic signal. The anti-noise signal cancels the actual background magnetic noise, thus providing a compensated signal 166 output to the analog readout circuit.

[0067] The environmental magnetic noise can be sampled and its phase aligned with the measurement signal as accurately as possible to provide maximum attenuation. Although 100% noise cancellation cannot be achieved in a practical system, the gradiometer compensation technique disclosed herein can achieve a noise reduction of 20 - 40 dB, thus reducing the background noise level to between one quarter and one sixteenth of the original level.

[0068] Back to Figure 4, the output of the magnetic induction sub-structure 103 is received by the analog readout circuit 110. In this example, the analog readout circuit 110 includes a transimpedance amplifier (not shown), an instrumentation amplifier 124, a band-pass filter 126, a programmable gain amplifier 128, and an analog multiplexer 130, which selectively connects the signals of each channel in the magnetic induction sub-structure 103 to an analog-to-digital converter (ADC) 132, and this converter can be part of a microprocessor unit. The transimpedance amplifier is used to sense the current signal generated by the TMR sensor and convert it into a voltage reading with the maximum signal-to-noise ratio for subsequent signal processing. The band-pass filter 126 can be a high-order filter using the Sallen-key topology, with a cut-off frequency of 300 to 500 Hz. In other embodiments, the band-pass filter 126 can include a 20th-order Butterworth filter with a band-pass range of 30 to 300 Hz. The filtered signal is converted into digital data by the ADC 132. In this example, the ADC 132 is an 18-bit successive approximation register ADC, featuring high speed, high precision, low power consumption, and low cost.

[0069] The converted digital signal can be transmitted to the external computing system 150 through the wireless communication module 114. In this example, the external computing system 150 can include a computer (such as a laptop or desktop device) 152 and a smartphone 154, which allows for the quick display of data in a convenient manner. Data can be transmitted between the module 100 and the smartphone 154 through the protective case 140 of the module using a conventional wireless communication protocol (such as etc.). The protective case 140 can be configured for magnetic shielding to reduce external magnetic noise in the module 100. The smartphone 154 can be connected to communicate with the computer 152, for example, through a USB interface or a wireless connection, so that the digital signal can be extracted, classified, and displayed in the LabVIEW interface 156 on the computer.

[0070] Figure 7 is a schematic diagram showing the baseline compensation arrangement that can be used for the above-mentioned biomagnetic sensor module. The baseline compensation arrangement is used to compensate for the inherent mismatch between the TMR sensor units in the main sensing layer 104 and the secondary sensing layer 108, as well as the mismatch introduced by the shielding layer 106. This compensation can be achieved in the following ways: by controlling the bias of the reference TMR sensor unit in the secondary sensing layer 108 (i.e., the gradiometer unit) through an adjustable current source, or by modifying the output signal from the secondary sensing layer 108 using a variable gain low-noise amplifier, and then being received by a noise cancellation unit (such as a subtraction circuit, such as the comparator 160 discussed above).

[0071] The baseline compensation arrangement is initialized by performing a preliminary step of measuring the baseline output voltages from the primary sensing layer 104 and the secondary sensing layer 108 without the target magnetic field (i.e., with the biomagnetic sensor module away from the measurement location). This baseline output is due to the inherent mismatch between the TMR sensor units in the primary sensing layer 104 and the secondary sensing layer 108 and the mismatch introduced by the shielding layer 106. This mismatch can then be eliminated by adjusting the settings of the secondary sensing layer 108. In an example where an adjustable current source is used to bias the secondary sensing layer 108, this can be achieved by modifying the bias signal received by the TMR sensor units in the secondary sensing layer 108. In this case, the secondary sensing layer 108 typically receives a different bias signal from the primary sensing layer 104. Alternatively, in an example where a variable gain amplifier is used at the output of the secondary sensing layer 108, this can be achieved by modifying the gain of the amplifier. By using the output voltage of the readout circuit, a feedback loop is established to set the level of the bias signal or gain to minimize the baseline output voltage difference. Once the optimal bias level or gain is found, it can be applied to the TMR sensor units or the low-noise amplifier of the secondary sensing layer 108 when measuring the target magnetic field. This reduces the mismatch caused by the sensor and the magnetic shielding, improving the overall sensitivity and accuracy of the gradiometer configuration.

[0072] Figure 8 A three operational amplifier architecture 170 is shown, which is located between the Wheatstone bridge arrangement 172 and the comparator 160 of each channel of the primary sensing layer 104 and the secondary sensing layer 106. Each Wheatstone bridge arrangement 172 includes a plurality of TMR sensor arrays for providing differential input signals V in+ and V in- . The architecture 170 includes an input buffer stage 172, followed by a differential amplifier stage 174, after which the signal is transmitted to the comparator 160 to remove background noise from the measurement signal before providing the measurement signal to the analog readout circuit 110.

[0073] The input buffer stage 172 and the differential amplifier stage 174 are implemented using a three operational amplifier architecture. Two low-noise input amplifiers A1, A2 are used in the input buffer stage 172. A third amplifier A3 is used in the differential amplifier stage 174. As described in more detail below, this architecture can be configured to achieve a high input impedance and excellent linearity and to extend the input range by using a rail-to-rail input stage.

[0074] The control circuit 112 (e.g., a microprocessor unit) is arranged to generate control signals for the architecture 170. One of the control signals can be used to set the gain of the input buffer stage 172 by adjusting the variable resistor R1, and the input buffer stage 172 is connected to the control circuit 112 via a digital-to-analog converter (DAC) 176. Similarly, another control signal can be used to set the gain of the differential amplifier stage 174 by adjusting the variable resistor R3. In the latter case, a digital-to-analog converter (DAC) connected to the input of the fully differential amplifier A3 can be used to adjust the output offset of the three operational amplifiers. Thus, the DAC 176 can set the currents I P and I N to adjust the resistor R3. The control circuit 112 can also generate a common-mode feedback (CMFB) control signal to set the bias current of the differential amplifier stage 174.

[0075] The two input amplifiers A1, A2 can utilize chopper switches to implement modulation and demodulation techniques, thereby eliminating input offset and low-frequency flicker noise. For example, the two amplifiers A1, A2 in the input buffer stage 172 can have input chopper switches (not shown), which modulate the differential input signals V in+ and V in- to the chopping frequency, thus helping to eliminate the upshift offset and low-frequency flicker noise. The amplifier A3 in the differential amplifier stage 174 can have a chopping output stage configured to reintegrate the signal. The chopping output stage can include chopper switches for synchronously demodulating the signal back to its original frequency while modulating the offset and 1 / f noise of the amplifier input stage to the chopping frequency. The chopper switches are driven by corresponding control signals issued by the microcontroller, thereby implementing a suitable modulation / demodulation process. The chopping frequency is typically selected between a few hundred hertz and a few thousand hertz. The chopping frequency should be chosen to be greater than (at least twice) the sampling frequency of the ADC 132 to prevent errors due to aliasing.

[0076] A common-mode feedback (CMFB) circuit can be employed to maintain the DC voltage output. The CMFB circuit stabilizes the common-mode voltage by regulating the common-mode output current. In this example, the CMFB circuit is configured to detect the common-mode voltage by obtaining the average value of the differential output voltage from the amplifier A3, compare the obtained average value with a reference voltage, and return the differential voltage between the average value and the reference voltage to the bias network of the differential amplifier stage 174. Here, the differential voltage is converted into a common-mode output current to adjust the common-mode voltage. Thus, it eliminates the output common-mode current component and fixes the DC output to the desired level. Typically, the reference voltage can be set to half of the rail voltage.

[0077] The transfer function of the proposed three operational amplifier structure including DAC operation is expressed as:

[0078] Figure 9A Shows the simulated MMG signal, which is used to simulate the effect of the noise cancellation technique discussed above. The figure shows two simulated MMG signals 200, 202 with different amplitudes. The first signal 200 represents the MMG signal near its source muscle. The first signal 200 has a frequency of 100 Hz and an amplitude of 322 pT. The second signal 202 represents the same MMG signal, but its distance from the muscle is increased by 20 mm compared to the first signal 200. The second signal 202 has a frequency of 100 Hz and an amplitude of 177 pT to illustrate the way the signal attenuates as the distance from the muscle increases. Providing a shielding layer between the TMR sensors in the primary sensing layer 104 and the secondary sensing layer 108 can further attenuate the MMG signal such that it is substantially negligible on the TMR sensors in the secondary sensing layer 108.

[0079] Figure 9B Is a graph showing the relative amplitudes of the noise signal 204 experienced by the TMR sensors in the primary sensing layer 104 and the secondary sensing layer 108 and the MMG signal 200 near its source muscle. The noise signal 204 in this example is simulated as a 10 Hz signal with an amplitude of 2.5 μT, which is one order of magnitude higher than the MMG signal 200. As referred to above Figure 6 As described, running the simulation to invert the simulated noise signal on the TMR sensors in the secondary sensing layer 108 with the simulated noise and MMG signals on the TMR sensors in the primary sensing layer 104 shows that the MMG signal can be effectively extracted despite the relatively high amplitude of the noise signal.

[0080] In a further embodiment, the noise cancellation technique discussed above can be further enhanced by combining the biomagnetic sensor module described herein with existing bioelectric sensors (such as electrocardiogram sensors, etc.). The bioelectric sensors provide complementary sensing outputs, and the noise in the spectrally similar signals can be eliminated by using on-chip signal processing of the biomagnetic signals. In fact, this is achieved using reservoir computing (RC) techniques, in which the outputs from the EMG and MMG sensors are both analyzed to identify and separate 1 / f noise from the MMG signal. Since the sensors operate as physical RC model units, implementing using physical reservoir computing is beneficial for greatly reducing the training time and memory requirements

[13] ,

[0081] To explore the practicality of the above biomagnetic sensor module, a model simulating biomagnetic potential was established. Figure 10 Shows a cross-section of the illustration 400 of the muscle tissue 402 used in the simulation model. Figure 10The image in shows a cross-section (orthogonal to the fiber direction) of muscle tissue 402, which consists of multiple muscle fascicles 404, each muscle fascicle having multiple fibers 406, 408, 410 belonging to different motor units. Each muscle fiber carries an electric current (propagating action potential). The model simulates the combined effect of the magnetic fields generated by the time-varying action potential propagation in a population of skeletal muscle cells in the time and space domains based on the different electric currents in each muscle fiber.

[0082] The layer between the muscle and the skin surface is called volume conduction and plays a key role during signal measurement. Finite-difference time-domain simulations are obtained using the simulation model discussed above to study the effects of volume conduction on electrical and magnetic signals. Figure 11A Shows a cross-section of a muscle tissue diagram 500, showing the simulated electric field potential in volume 501 between muscle bundles 505 and the skin surface 507 in the absence of a fat layer. Figure 11B Shows the same cross-section, but shows the simulated magnetic potential.

[0083] Figure 11C Shows a cross-section of another representation of muscle tissue, showing the simulated electric field potential in volume 501 between muscle bundles 505 and the skin surface 507, with a layer of fat 503 around the muscle bundles. As Figure 11D Shows the same cross-section, but shows the simulated magnetic potential.

[0084] Figure 12 Is a graph showing the variation of the simulated electric field potential along Figure 11A And Figure 11C Lines A-A and C-C in . Figure 13 Is a graph showing the variation of the simulated magnetic potential along Figure 11B And 11D Lines B-B and D-D in . As can be seen from the inset in Figure 12 Increasing the fat layer by 1 mm results in a 60% decrease in the electrical signal 502 compared to the electrical signal 504 at the skin surface, while the magnetic potential remains unchanged. Therefore, the isolation layer between the skin and the signal source is transparent to the magnetic field but disturbs the electric field.

[0085] Figure 14It is a schematic diagram of the change of the biomagnetic potential in muscle tissue with its distance from the source. This figure is obtained using the simulation model discussed above. It can be seen that the amplitude of the biomagnetic potential drops rapidly from the source and is already close to 50 pT at the skin surface. However, it can be seen that in the region 25 - 30 mm from the source (10 - 15 mm from the skin surface), the amplitude of the biomagnetic field is in the range of 24 - 28 pT, which belongs to the sensitivity range of the above-mentioned TMR sensor. Within this detection region, the intensity of the biomagnetic field is still decreasing, so the secondary sensing layer will experience a significantly lower biomagnetic field. As discussed regarding Figure 1 providing a shielding layer between the primary sensing layer and the secondary sensing layer can ensure that the biomagnetic field in the secondary sensing layer is negligible.

[0086] As described above, the biomagnetic induction device of the present invention can adopt a combination of different technologies to eliminate noise in the environment (i.e., the earth's magnetic field, thermal noise, and 1 / f noise): 1) Adopting readout circuit technologies such as chopping and auto-zeroing to compensate for offsets; 2) Integrating a gradiometer on the chip and using another TMR sensor to record the earth's magnetic field; 3) Utilizing a thermally stable material layer at the device manufacturing level to stabilize the temperature.

[0087] The features disclosed in the foregoing description, the appended claims, or the drawings, in their specific forms or in the form of means for performing the disclosed functions or in the form of methods or processes for obtaining the disclosed results, can be used alone or in any combination of these features to implement different forms of inventions.

[0088] Although the present invention has been described in connection with the above exemplary embodiments, many equivalent modifications and variations will be apparent to those skilled in the art when the present disclosure is given. Therefore, the above exemplary embodiments of the present invention are illustrative rather than restrictive. Various changes can be made to the described embodiments without departing from the spirit and scope of the present invention.

[0089] To avoid any doubt, any theoretical explanations provided herein are for enhancing the reader's understanding. The inventors do not wish to be bound by any theoretical explanations.

[0090] Any chapter headings used herein are for organizational purposes only and should not be construed as limiting the subject matter described.

[0091] Throughout the specification (including the following claims), unless the context otherwise requires, words such as "comprising", "including", and "containing" and their variants shall be understood to mean including the stated integers or steps or groups of integers or steps, but not excluding any other integers or steps or groups of integers or steps.

[0092] It must be noted that, as used in the specification and the appended claims, the singular forms "a", "an" and "the" include plural referents unless the context clearly dictates otherwise. Ranges may be expressed herein as from "about" one particular value, and / or to "about" another particular value. When such a range is expressed, another embodiment includes from the one particular value and / or to the other particular value. Similarly, when values are expressed as approximations, by use of the antecedent "about", the particular value is understood to form another embodiment. The term "about" associated with a numerical value is optional and is meant to represent, for example, + / - 10%.

[0093] References

[0094] Numerous publications are cited above in order to more fully describe and disclose the present invention and the prior art in the field to which the present invention pertains. The complete citations of these references are shown below. The entire contents of these references are incorporated herein by reference. [1] J.Malmivuo and R.Plonsey, Bioelectromagnetism: principles and applications of bioelectric and biomagnetic fields. Oxford University Press, USA, 1995. [2] D.B.Geselowitz, ‘Magnetocardiography: an overview’, IEEE Transactions on Biomedical Engineering, no.9, pp.497 - 504, 1979. [3] R.Fenici, D.Brisinda, and A.M.Meloni, ‘Clinical application of magnetocardiography’, Expert Review of Molecular Diagnostics, vol.5, no.3, pp.291 - 313, 2005. [4] S.Baillet, ‘Magnetoencephalography for brain electrophysiology and imaging’, Nature Neuroscience, vol.20, no.3, pp.327 - 339, 2017. [5]E.Boto et al.,‘Moving magnetoencephalography towards real-worldapplications with awearable system’,Nature,vol.555,no.7698,p.657,2018. [6]S.Zuo,H.Heidari,D.Farina,and K.Nazarpour,‘Miniaturized magneticsensors forimplantable magnetomyography’,Advanced Materials Technologies,no.2000185,2020. [7]S.Zuo et al.,‘Ultrasensitive Magnetoelectric Sensing System forpico-TeslaMagnetoMyoGraphy’,IEEE Transaction on Biomedical Circuits andSystems,2020. [8]B.-M.Mackert,‘Magnetoneurography:theory and application toperipheral nervedisorders’,Clin.Neurophysiol.,vol.115,no.12,pp.2667-2676,2004. [9]C.-H.Im,S.C.Jun,and K.Sekihara,‘Recent advances in biomagnetismand itsapplications’.Springer,2017.

[10] R.Kleiner,D.Koelle,F.Ludwig,and J.Clarke,‘Superconducting quantuminterferencedevices:State of the art and applications’,Proceedings of theIEEE,vol.92,no.10,pp.1534-1548,2004.

[11] S. Zuo, K. Nazarpour, and H. Heidari, ‘Device modelling of MgO-barrier tunnelling magnetoresistors for hybrid spintronic-CMOS’, IEEE Electron Device Letter, vol. 39, no. 11, pp. 1784-1787, 2018.

[12] H. Heidari, S. Zuo, A. Krasoulis, and K. Nazarpour, ‘CMOS Magnetic Sensors for Wearable Magnetomyography’, 2018 40th International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Honolulu, HI, 2018, pp. 2116-2119.

[13] Liang, X., Zhong, Y., Tang, J. et al. Rotating neurons for all-analog implementation of cyclic reservoir computing. Nat Commun 13, 1549 (2022).

Claims

1. A biomagnetic sensor module, comprising: A tunneling magnetoresistance sensor unit configured to output a detection signal indicative of a magnetic field in proximity to biological tissue; A gradiometer unit configured to output a background signal indicative of ambient magnetic noise; A shielding element located between the tunneling magnetoresistance sensor unit and the gradiometer unit; An active noise cancellation unit configured to remove the background signal from the detection signal to generate a biomagnetic signal; An analog readout circuit configured to receive the biomagnetic signal and perform signal conditioning to generate an analog output; And An analog-to-digital converter configured to generate a digital output signal from the analog output.

2. The biomagnetic sensor module according to claim 1, wherein the gradiometer unit comprises a triaxial gradiometer.

3. The biomagnetic sensor module according to claim 1 or 2, wherein the gradiometer unit comprises a plurality of tunneling magnetoresistance sensors configured to detect ambient magnetic noise.

4. The biomagnetic sensor module according to any one of the preceding claims, wherein the tunneling magnetoresistance sensor unit comprises four tunneling magnetoresistance sensors arranged in a Wheatstone bridge configuration.

5. The biomagnetic sensor module according to any one of the preceding claims, wherein the tunneling magnetoresistance sensor unit is one of a plurality of similarly configured tunneling magnetoresistance sensor units, wherein each tunneling magnetoresistance sensor unit is arranged to detect a magnetic field in proximity to biological tissue and output a biomagnetic signal on a corresponding channel, and wherein the biomagnetic sensor module further comprises a multiplexer configured to selectively couple each corresponding channel to the analog readout circuit.

6. The biomagnetic sensor module according to any one of the preceding claims, further comprising a skin contact interface made of a biocompatible material.

7. The biomagnetic sensor module according to claim 6, wherein the gradiometer unit is located at a position farther from the skin contact interface than the tunneling magnetoresistance sensor unit.

8. The biomagnetic sensor module according to any one of the preceding claims, wherein the shielding element comprises a magnetically permeable material.

9. The biomagnetic sensor module according to any one of the preceding claims, further comprising a compensation unit configured to minimize a baseline mismatch between the tunneling magnetoresistance sensor unit and the gradiometer unit.

10. The biomagnetic sensor module according to claim 9, wherein the compensation unit comprises a feedback circuit configured to receive baseline outputs from the tunneling magnetoresistance and gradiometer units and adjust the gradiometer unit to minimize the baseline output.

11. The biomagnetic sensor module according to claim 9 or 10, wherein the compensation unit comprises an adjustable current source configured to generate a bias signal for the gradiometer unit, wherein the bias signal is adjustable to minimize the baseline output.

12. The biomagnetic sensor module according to claim 9 or 10, wherein the compensation unit includes a variable gain amplifier connected between the gradiometer unit and the active noise cancellation unit, and the gain of the variable gain amplifier is adjustable to minimize the baseline output.

13. The biomagnetic sensor module according to any one of the preceding claims, wherein the tunneling magnetoresistance sensor unit is arranged in the main sensing layer, and the gradiometer unit is arranged in a secondary sensing layer parallel to and spaced apart from the main sensing layer.

14. The biomagnetic sensor module according to claim 13, wherein the distance between the main sensing layer and the secondary sensing layer is equal to or less than 10 mm.

15. The biomagnetic sensor module according to claim 13 or 14, wherein the active noise cancellation unit and the analog readout circuit are arranged in one or more layers parallel to and spaced apart from the secondary sensing layer.

16. The biomagnetic sensor module according to any one of the preceding claims, wherein the analog readout circuit includes a bandpass filter and a common mode feedback circuit.

17. The biomagnetic sensor module according to any one of the preceding claims, further comprising a wireless communication unit configured to transmit a digital output signal to a remote device.

18. A wearable biomagnetic sensing device comprising the biomagnetic sensor module according to any one of the preceding claims.

19. The wearable biomagnetic sensing device according to claim 18, comprising a plurality of biomagnetic sensor modules according to any one of claims 1 to 17, the biomagnetic sensor modules being fixed to a holding element that can be mounted on a human body.

20. The wearable biomagnetic sensing device according to claim 19, wherein the holding element is an elastic band.

Images