Smart ring and blood pressure measurement method

Abstract

Embodiments of the present application provide a smart ring and a blood pressure measurement method. The smart ring is internally configured with a soft sac assembly, a control assembly, a PPG sensor group, and a central control unit. When a user wears the smart ring on the middle finger, the central control unit can cause the soft sac assembly to apply a compression force to the first and second arteries through the control assembly, achieving the effect of cutting off the flow of the first and second arteries. The waveform features and / or phase features of the pulse waveforms collected by the PPG sensor group before and after cutting off the flow are taken as blood flow features, and the blood pressure measurement result of the user is determined according to the blood flow feature data. It can be applied to continuous dynamic detection of blood pressure and can ensure the accuracy of the blood pressure measurement result.

Description

Technical Field

[0001] This application relates to the field of blood pressure monitoring technology, and in particular to smart rings and methods for measuring blood pressure. Background Technology

[0002] If hypertension is not detected and treated early, it can easily lead to heart attacks, strokes, myocardial infarctions, and other serious health problems. Therefore, early screening and daily monitoring of hypertension are crucial. However, current blood pressure measurement methods are primarily based on cuff-type blood pressure monitors. These devices only provide dynamic snapshots of blood pressure and are not suitable for continuously tracking changes in a user's blood pressure. Furthermore, they are not very comfortable to wear. Summary of the Invention

[0003] The purpose of this application is to provide a smart ring and a blood pressure measurement method to achieve continuous dynamic detection of blood pressure and ensure the accuracy of the blood pressure measurement results. The specific technical solution is as follows:

[0004] In a first aspect, this application provides a smart ring for wearing on a user's middle finger, the smart ring including a central control unit, a sensor assembly, a soft capsule assembly, and a control assembly;

[0005] The sensor assembly includes a PPG sensor group; the PPG sensor group includes a first PPG sensor, a second PPG sensor, a third PPG sensor, and a fourth PPG sensor; the first PPG sensor and the second PPG sensor are arranged along the inner ring of the smart ring and opposite each other, and are respectively used to measure the pulse wave waveform of the first artery on the side of the middle finger closer to the index finger, and the pulse wave waveform of the second artery on the side of the middle finger closer to the ring finger; the third PPG sensor is arranged along the outer ring of the smart ring and opposite the first PPG sensor, and is used to measure the pulse wave waveform of the third artery on the side of the index finger closer to the middle finger; the fourth PPG sensor is arranged along the outer ring of the smart ring and opposite the second PPG sensor, and is used to measure the pulse wave waveform of the fourth artery on the side of the ring finger closer to the middle finger.

[0006] The soft capsule assembly includes a first soft capsule structure and a second soft capsule structure; the first soft capsule structure is arranged around the first PPG sensor, and the second soft capsule structure is arranged around the second PPG sensor.

[0007] The control component is connected to the soft capsule component and is used to control the first soft capsule structure to apply pressure to the first artery and to control the second soft capsule structure to apply pressure to the second artery.

[0008] The central control unit is configured to acquire first data collected by the sensor assembly when the first artery and the second artery are in a non-compression state, the first data including: first pulse wave waveforms collected by each PPG sensor in the PPG sensor group; control the soft balloon assembly through the control component to apply a first pressure force to the first artery and the second artery; acquire second data collected by the sensor assembly when the first artery and the second artery are in a compression state, the second data including: second pulse wave waveforms collected by each PPG sensor in the PPG sensor group; determine blood flow characteristic data based on the first data and the second data, the blood flow characteristic data including: first waveform characteristics of each first pulse wave waveform and second waveform characteristics of each second pulse wave waveform, and / or, first pulse conduction time difference between first pulse wave waveforms of different phases and second pulse conduction time difference between second pulse wave waveforms of different phases; and determine the user's blood pressure measurement result based on the blood flow characteristic data.

[0009] This application analyzes the blood flow in the arteries of the fingers using Bernoulli's principle, combines the analysis results with existing medical research experience, and proposes a novel approach to blood pressure measurement. Based on this approach, a smart ring for measuring a user's blood pressure is provided. The smart ring includes a soft capsule component that applies pressure to the first and second arteries of the user's middle finger, a control component for controlling the capsule component, a PPG sensor group for measuring the pulse wave waveform of blood in the user's middle, index, and ring fingers, and a central control unit. When the user wears the smart ring on their middle finger, the central control unit applies pressure to the first and second arteries through the control component, achieving a flow restriction effect. Based on this, the waveform characteristics and / or phase characteristics of the pulse wave waveforms collected by the PPG sensor group before and after the flow restriction are used as blood flow characteristic data, and the user's blood pressure measurement result is determined based on this blood flow characteristic data.

[0010] Compared to cuff-type blood pressure measuring devices, blood pressure measurement using the smart ring provided in this application embodiment eliminates the need for inflating and deflating a cuff, making the measurement process simpler and more convenient. Furthermore, because the smart ring itself is easy to wear, it can be used for continuous dynamic blood pressure measurement in daily life to track changes in blood pressure. Moreover, during the measurement of a user's blood pressure using the smart ring, since the first and second arteries within the middle finger are very small, a large amount of pressure is not required to achieve a blockage effect on the first and second arteries, resulting in minimal discomfort for the user. Therefore, it is also suitable for nighttime blood pressure measurement.

[0011] Compared with existing blood pressure measurement technologies based on single-channel pulse wave waveforms, the smart ring provided in this application comprehensively considers the multiple pulse wave waveforms collected by each PPG sensor in the PPG sensor group before and after the first and second arteries are blocked. Blood pressure prediction is based on the waveform characteristics and / or phase characteristics of these multiple pulse wave waveforms, which helps to ensure the accuracy of the obtained blood pressure measurement results.

[0012] In one possible implementation, the PPG sensor group further includes a fifth PPG sensor and a sixth PPG sensor; the fifth PPG sensor and the sixth PPG sensor are arranged along the inner circle of the smart ring and opposite to each other, and a first straight line connecting the fifth PPG sensor and the sixth PPG sensor is perpendicular to a second straight line connecting the first PPG sensor and the second PPG sensor.

[0013] In this embodiment, the smart ring collects pulse waveforms at different positions of the middle finger through a first PPG sensor, a second PPG sensor, a fifth PPG sensor, and a sixth PPG sensor. The pulse waveforms complement each other, which can more accurately reflect the blood flow in the middle finger artery, thereby helping to improve the accuracy of the final predicted blood pressure measurement result.

[0014] In one possible implementation, the control components include: stranded wire, winch, and winch motor;

[0015] One end of the stranded wire is connected to the winch, and is arranged in a clockwise or counterclockwise direction inside the smart ring and wrapped around the soft pouch assembly. The other end of the stranded wire is connected to a fixing member disposed inside the smart ring.

[0016] The winch motor is connected to the winch and is used to drive the winch to rotate to tighten or loosen the strand. The soft capsule assembly applies pressure to the first artery and the second artery as the strand is tightened, and relaxes the pressure applied to the first artery and the second artery as the strand is loosened.

[0017] In this embodiment, the soft capsule assembly is controlled by a winch motor, a winch, and a strand, enabling the soft capsule structure to apply pressure to the first artery and the second artery or to relax the applied pressure, and the overall structure is relatively simple.

[0018] In one possible implementation, the sensor assembly further includes a pressure sensor serving as the fixation element; the first data further includes: a first pressure value collected by the pressure sensor when the first artery and the second artery are in a non-compression state; the second data further includes: a second pressure value collected by the pressure sensor when the first artery and the second artery are in a compression state; the blood flow characteristic data further includes: the second pressure value and the first pressure value.

[0019] In this embodiment, the first pressure value and the second pressure value can characterize the changes in pressure measured by the pressure sensor before and after the first artery and the second artery are blocked by the soft balloon assembly. By using the first pressure value and the second pressure value as part of the blood flow characteristic data for predicting the user's blood pressure, it helps to further improve the accuracy of the predicted blood pressure measurement results.

[0020] In one possible implementation, the sensor assembly further includes an accelerometer, and the first data further includes: first acceleration information collected by the accelerometer when the first artery and the second artery are in a non-compression state; the second data further includes: second acceleration information collected by the accelerometer when the first artery and the second artery are in a compression state;

[0021] The central control unit is further configured to, after acquiring the first data and before controlling the soft balloon assembly through the control component, determine, based on the first acceleration information, whether the user is in motion during the information acquisition process of the sensor assembly; if so, determine that the acquired first data is invalid and terminate the current blood pressure measurement; if not, control the soft balloon assembly through the control component to make the soft balloon assembly apply pressure to the first artery and the second artery.

[0022] The central control unit is further configured to, after acquiring the second data and before inputting the blood flow feature data into the pre-trained blood pressure prediction model, determine, based on the second acceleration information, whether the user is in motion during the information acquisition process of the sensor component; if so, determine that the acquired second data is invalid and terminate the current blood pressure measurement; if not, input the blood flow feature data into the pre-trained blood pressure prediction model to obtain the user's blood pressure measurement result.

[0023] In this embodiment, by utilizing the acceleration information collected by the accelerometer, the smart ring can avoid measuring blood pressure when the user is in motion, which helps to ensure the accuracy of the blood pressure measurement results.

[0024] In one possible implementation, the central control unit is further configured to control the soft balloon assembly via the control component, causing the soft balloon assembly to apply a second pressure force to the user's first and second arteries, and to monitor the peak magnitude of the third pulse wave waveform collected by the first PPG sensor and the second PPG sensor during the application of the second pressure force; the magnitude of the pressure force that causes the peak magnitude of the third pulse wave waveform to be lower than a preset threshold is calibrated as the magnitude of the first pressure force.

[0025] In this embodiment, the first pressure force is calibrated based on the actual situation of each user, and the first and second arteries are compressed by applying the calibrated first pressure force during the actual blood pressure measurement process. Blood pressure prediction is performed based on the first data collected by the sensor component under the compression state, which helps to further ensure the accuracy of the blood pressure measurement results.

[0026] In one possible implementation, the first pulse conduction time difference includes: the pulse conduction time difference between the first pulse waveform acquired by the first PPG sensor and the first pulse waveform acquired by the third PPG sensor, and the pulse conduction time difference between the first pulse waveform acquired by the second PPG sensor and the first pulse waveform acquired by the fourth PPG sensor; the second pulse conduction time difference includes: the pulse conduction time difference between the second pulse waveform acquired by the third PPG sensor and the second pulse waveform acquired by the fourth PPG sensor.

[0027] By using the above three pulse conduction time differences as blood flow characteristic data, the phase relationship of blood flow time at different sites before and after the first and second arteries are cut off can be effectively extracted, which helps to ensure the accuracy of blood pressure measurement results and effectively avoids data redundancy.

[0028] In one possible implementation, the first waveform feature includes the rising area of ​​the first pulse wave waveform acquired by each PPG sensor in the PPG sensor group; the second waveform feature includes the rising area of ​​the second pulse wave waveform acquired by each PPG sensor in the PPG sensor group.

[0029] By using the rising area of ​​each first pulse wave waveform acquired by the PPG sensor group as the first waveform feature and the rising area of ​​each second pulse wave waveform acquired by the PPG sensor group as the second waveform feature, the obtained first and second waveform features can reflect the changes in blood volume flowing through the index and ring fingers after the first and second arteries are blocked. Therefore, when blood pressure prediction is performed based on the first and second waveform features, it helps to ensure the accuracy of the obtained blood pressure measurement results.

[0030] In one possible implementation, the first waveform feature further includes one or more of the 1 / LASI value, 1 / LASI_2 value, PIR value, and RIPV value of the first pulse wave waveform acquired by each PPG sensor in the PPG sensor group.

[0031] Existing medical research has shown a correlation between the 1 / LASI, 1 / LASI_2, PIR, and RIPV values ​​of the pulse wave waveform and blood pressure at the brachial artery. Therefore, in this embodiment, by extracting the 1 / LASI, 1 / LASI_2, PIR, and RIPV values ​​of the first pulse wave waveform collected by each PPG sensor in the PPG sensor group as the first waveform feature, the accuracy of the blood pressure measurement results can be further ensured when combining the first waveform feature with the second waveform feature for subsequent blood pressure prediction.

[0032] Secondly, this application provides a method for measuring blood pressure, including:

[0033] The system acquires first data collected from the user when the first artery on the side of the middle finger near the index finger and the second artery on the side of the middle finger near the ring finger are in a non-compression state. The first data includes: first pulse wave waveforms collected at multiple preset collection points on the user's hand. The preset collection points include: the first artery, the second artery, the third artery on the side of the index finger near the middle finger, and the fourth artery on the side of the ring finger near the middle finger.

[0034] Acquire second data from the user when the first artery and the second artery of the user are under compression; the second data includes: second pulse wave waveforms collected at each of the preset collection sites on the user's hand;

[0035] Based on the first data and the second data, blood flow characteristic data is determined; the blood flow characteristic data includes: the first waveform characteristics of each first pulse wave waveform and the second waveform characteristics of each second pulse wave waveform, and / or, the first pulse conduction time difference between first pulse wave waveforms of different phases and the second pulse conduction time difference between second pulse wave waveforms of different phases;

[0036] Based on the blood flow characteristic data, the user's blood pressure measurement result is determined.

[0037] In one possible implementation, determining the user's blood pressure measurement result based on the blood flow characteristic data includes:

[0038] The blood flow feature data is input into a pre-trained blood pressure prediction model to obtain the blood pressure measurement results;

[0039] The blood pressure prediction model was trained in the following manner:

[0040] The sample first data is collected from the sample user when the first artery on the side of the middle finger near the index finger and the second artery on the side of the middle finger near the ring finger are in a non-compression state; the sample first data includes: the sample first pulse wave waveform collected at the multiple preset collection sites on the sample user's hand respectively;

[0041] Acquire sample second data from the sample user when the first artery and the second artery of the sample user are under compression; the sample second data includes sample second pulse wave waveforms collected at the plurality of preset collection sites on the sample user's hand;

[0042] Based on the first sample data and the second sample data, the blood flow characteristic data of the sample user is determined; the blood flow characteristic data includes: the first waveform characteristics of each sample's first pulse wave waveform and the second waveform characteristics of each sample's second pulse wave waveform, and / or, the first pulse conduction time difference between sample first pulse wave waveforms of different phases and the second pulse conduction time difference between sample second pulse wave waveforms of different phases;

[0043] The blood flow feature data of the sample is input into the blood pressure prediction model to be trained to obtain the predicted blood pressure measurement results of the sample users;

[0044] Based on the predicted blood pressure measurement results of the sample users and the actual blood pressure measurement results of the sample users, the blood pressure prediction model to be trained is trained to obtain the trained blood pressure prediction model.

[0045] The beneficial effects of the second aspect can be referred to the first aspect, and will not be elaborated here. Attached Figure Description

[0046] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other embodiments can be obtained based on these drawings.

[0047] Figure 1 This is a schematic diagram showing the distribution of arteries in the hand;

[0048] Figure 2This is a first hardware structure block diagram of a smart ring provided in an embodiment of this application;

[0049] Figure 3 This is a schematic diagram of a first structure of a smart ring provided in an embodiment of this application;

[0050] Figure 4 For users Figure 3 A diagram illustrating the smart ring worn on the middle finger;

[0051] Figure 5 A schematic diagram of the blood pressure measurement process performed by the central control unit within the smart ring provided in this application embodiment;

[0052] Figure 6 This is a schematic diagram illustrating the process of binding a smart ring to a mobile phone, as provided in an embodiment of this application.

[0053] Figure 7 A schematic diagram illustrating the process of querying blood pressure measurement results via mobile phone, as provided in an embodiment of this application;

[0054] Figure 8 A schematic diagram of pulse conduction time difference provided in an embodiment of this application;

[0055] Figure 9 This is a second hardware structure block diagram of the smart ring provided in the embodiments of this application;

[0056] Figure 10 This is a schematic diagram of a second structure of the smart ring provided in an embodiment of this application;

[0057] Figure 11 A third hardware structure block diagram of the smart ring provided in the embodiments of this application;

[0058] Figure 12 A schematic diagram illustrating the configuration of the control components within the smart ring provided in this application embodiment;

[0059] Figure 13 A fourth hardware structure block diagram of the smart ring provided in the embodiments of this application;

[0060] Figure 14 Another schematic diagram illustrating the configuration of the control components within the smart ring provided in this application embodiment;

[0061] Figure 15 A fifth hardware structure block diagram of the smart ring provided in the embodiments of this application;

[0062] Figure 16 A schematic diagram of the rising area provided in an embodiment of this application;

[0063] Figure 17A schematic diagram illustrating the 1 / LASI value and 1 / LASI_2 value provided in an embodiment of this application;

[0064] Figure 18 A schematic diagram of a pulse wave waveform acquired by the PPG sensor group provided in an embodiment of this application;

[0065] Figure 19 A schematic diagram illustrating a single blood pressure measurement process provided in an embodiment of this application;

[0066] Figure 20 A sixth hardware structure block diagram of the smart ring provided in the embodiments of this application;

[0067] Figure 21 A schematic flowchart illustrating the blood pressure measurement method provided in this application embodiment;

[0068] Figure 22 This is a flowchart illustrating the training process of the blood pressure prediction model provided in the embodiments of this application. Detailed Implementation

[0069] To better understand the technical solution of this application, the embodiments of this application will be described in detail below with reference to the accompanying drawings.

[0070] In the description of the embodiments of this application, "and / or" is merely a way of describing the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Furthermore, in the description of the embodiments of this application, "multiple" refers to two or more.

[0071] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature.

[0072] If hypertension is not detected and treated early, it can easily lead to heart attacks, strokes, myocardial infarctions, and other serious health problems. Therefore, early screening and daily monitoring of hypertension are crucial. However, current blood pressure measurement methods are primarily based on cuff-type blood pressure monitors. These devices only provide dynamic snapshots of blood pressure and are not suitable for continuously tracking changes in a user's blood pressure. Furthermore, they are not very comfortable to wear.

[0073] Currently, some blood pressure measurement methods can achieve cuffless blood pressure measurement, such as the PWV (Pulse Wave Velocity) method and the PPG (Photo Plethysmo Graphy) pulse waveform method. However, these blood pressure measurement methods rely on single-channel pulse waveforms to predict blood pressure, and single-channel pulse waveforms generally have poor accuracy, resulting in poor accuracy of the blood pressure measurement results obtained from them, making them unsuitable for practical use.

[0074] In view of this, this application provides a smart ring that, when worn on the middle finger, can predict blood pressure by collecting pulse wave waveforms from the user's middle, ring, and index fingers, thus obtaining the user's blood pressure measurement results. For ease of understanding, the following will first combine... Figures 1-4 The basic principle of blood pressure measurement in this application and its application scenarios are illustrated by example:

[0075] See Figure 1 The illustration shows the superficial palmar arch of the human hand, formed by the anastomosis of the terminal branch of the ulnar artery and the superficial palmar branch of the radial artery. The superficial palmar arch branches into the downstream palmar digital arteries. Figure 1 The two palmar digital arteries, marked E and F, further branch into the proper palmar digital arteries, and branches of the same palmar digital artery flow to different fingers. Figure 1 The palmar digital arteries shown include: the first artery of the middle finger on the side closer to the index finger, marked A in the figure; the second artery of the middle finger on the side closer to the ring finger, marked B in the figure; the third artery of the index finger on the side closer to the middle finger, marked C in the figure; and the fourth artery of the ring finger on the side closer to the middle finger, marked D in the figure.

[0076] Consider a scenario where pressure is applied to the side of the user's middle finger closest to the index finger and the side closest to the ring finger, causing obstruction of the first and second arteries in the middle finger. Since the superficial palmar arch has many arterial branches, the impact of the obstruction at the first and second arteries on the pressure P0 at the superficial palmar arch is negligible. Therefore, Bernoulli's principle can be used to analyze this situation. Figure 1 The illustration shows the blood flow in each artery in this scenario. Bernoulli's principle states that, under ideal conditions, at any cross-section of the same flow tube, the sum of the kinetic energy, potential energy, and pressure potential energy of a unit volume of fluid is a constant.

[0077] for Figure 1In the illustrated scenario, the superficial palmar arch, the palmar digital arteries downstream of the superficial palmar arch, and the proper palmar digital arteries further downstream of the palmar digital arteries can be considered as a flow tube for transporting blood. The superficial palmar arch, the palmar digital arteries, and the proper palmar digital arteries are located at different cross-sections of this flow tube. Analysis based on Bernoulli's principle shows that after the first and second arteries of the middle finger are blocked, the downstream resistance of the palmar digital arteries E and F increases. Therefore, the blood volume and pressure flowing through the index and ring fingers will change slightly, and this change has a non-linear relationship with the pressure P0 at the superficial palmar arch. Existing medical research results indicate a strong correlation between the pressure P0 at the superficial palmar arch and the blood pressure at the brachial artery. Therefore, combining the analysis based on Bernoulli's principle and medical research experience, it can be concluded that the changes in blood volume and pressure flowing through the user's index and ring fingers before and after the blocking of the first and second arteries of the user's middle finger are related to the blood pressure at the user's brachial artery.

[0078] Based on the above analysis, this application proposes a novel approach to blood pressure measurement. According to this approach, pulse waveforms need to be collected at the first and second arteries of the user's middle finger, the third artery of the index finger, and the fourth artery of the ring finger, under two different conditions: uncompressed and compressed. For ease of description, the pulse waveform collected when the first and second arteries are uncompressed will be referred to as the first pulse waveform, and the pulse waveform collected when the first and second arteries are compressed will be referred to as the second pulse waveform. Then, blood flow characteristic data reflecting the blood volume and / or pressure within the blood vessels are extracted from the first and second pulse waveforms. These blood flow characteristic data reflect the changes in blood volume and / or pressure flowing through the user's index and ring fingers before and after the first and second arteries of the middle finger are blocked. Therefore, these blood flow characteristic data can be considered correlated with the user's blood pressure. Based on this, in this application, the user's blood pressure measurement results are determined based on these blood flow characteristic data.

[0079] Specifically, in actual measurement scenarios, when applying pressure to the side of the user's middle finger closest to the index finger and the side of the user's middle finger closest to the ring finger to achieve the closure of the first and second arteries, it is impossible to completely stop the blood flow in the first and second arteries. Instead, it can only reduce the blood volume flowing through the first and second arteries to a certain extent. Depending on the degree of reduction in the blood volume flowing through the first and second arteries, the changes in blood volume and pressure flowing through the index and ring fingers will also differ. Therefore, in this application, feature data will be extracted from the first and second pulse wave waveforms collected from the first and second arteries of the middle finger, and combined with feature data extracted from the first and second pulse wave waveforms collected from the third artery of the index finger and the fourth artery of the ring finger to predict the user's blood pressure, thereby obtaining a more accurate blood pressure measurement result.

[0080] Based on the aforementioned blood pressure measurement approach, this application provides a smart ring. See also... Figure 2 and Figure 3 As illustrated, the smart ring 100 provided in this application embodiment includes a central control unit 140, a sensor assembly 120, a soft capsule assembly 130, and a control assembly 150.

[0081] The sensor assembly 120 includes a PPG sensor group, which contains multiple PPG sensors. The capsule assembly 130 includes two capsule structures. The PPG sensors are used to detect pulse wave waveforms. Their basic detection principle is as follows: during each heartbeat, the contraction and expansion of blood vessels affect the absorption of light by the blood in the arteries. The PPG sensor illuminates the skin surface with a light beam of a specific wavelength and detects the intensity of the reflected light. Based on the changes in the reflected light intensity, the pulse wave waveform within the cardiac cycle is derived. This pulse wave waveform reflects the flow of blood in the arteries during the cardiac cycle.

[0082] The specific deployment details of the soft capsule assembly 130 and the PPG sensors within the PPG sensor group are as follows: Figure 3 As shown, in Figure 3 In the structure, the smart ring 100 has four PPG sensors, and the four PPG sensors are on the same straight line. Two PPG sensors are arranged along the inner circle of the ring body 110, and the other two PPG sensors are arranged along the outer circle of the ring body 110. The soft capsule assembly includes two sets of soft capsule structures, which are respectively arranged around the two PPG sensors arranged along the ring body 110.

[0083] During the use of the smart ring 100, the user can wear it on the middle finger, aligning the two PPG sensors along the inner ring of the ring body 110 with the side of the middle finger closest to the index finger (i.e., the location of the first artery) and the side of the middle finger closest to the ring finger (i.e., the location of the second artery), respectively. In practical application, the user can wear the smart ring 100 on either the left or right middle finger; the subsequent blood pressure prediction process remains completely consistent. Furthermore, the two PPG sensors along the inner ring of the ring body 110 only need to be aligned with the first and second arteries, respectively; whether each PPG sensor is specifically aligned with the first or second artery is irrelevant, and the subsequent blood pressure prediction process remains completely consistent.

[0084] For ease of explanation, in the following text, the PPG sensor located along the inner ring of the ring 110 and aligned with the middle finger near the index finger will be designated as the first PPG sensor 121; the PPG sensor located along the inner ring of the ring 110 and aligned with the middle finger near the ring finger will be designated as the second PPG sensor 122; the PPG sensor located along the outer ring of the ring 110 and aligned with the index finger near the middle finger will be designated as the third PPG sensor 123; and the PPG sensor located along the outer ring of the ring 110 and aligned with the ring finger near the middle finger will be designated as the fourth PPG sensor 124. The soft capsule structure surrounding the first PPG sensor 121 will be designated as the first soft capsule structure 131, and the soft capsule structure surrounding the second PPG sensor 122 will be designated as the second soft capsule structure 132.

[0085] See Figure 4 (a) shows that Figure 3 The diagram shows the smart ring 100 worn on the user's middle finger, and (b) shows an enlarged view of the smart ring 100 in (a). Figure 4 (b) further illustrates the various PPG sensors and Figure 1 The positional relationship between the first to fourth arteries, labeled AD, is shown. It can be seen that, based on the user's... Figure 4 When the smart ring 100 is worn as shown, the first PPG sensor 121 measures the pulse waveform of the first artery (artery A in the diagram) on the side of the middle finger closer to the index finger; the second PPG sensor 122 measures the pulse waveform of the first artery (artery B in the diagram) on the side of the middle finger closer to the ring finger; the third PPG sensor 123 measures the pulse waveform of the third artery (artery C in the diagram) on the side of the index finger closer to the middle finger; and the fourth PPG sensor 124 measures the pulse waveform of the fourth artery (artery D in the diagram) on the side of the ring finger closer to the middle finger. The first soft pouch structure 131 can be used to apply pressure to the first artery, and the second soft pouch structure 132 can be used to apply pressure to the second artery.

[0086] In practical applications, the specific shapes of the first soft capsule structure 131 and the second soft capsule structure 132 can be set according to actual needs, meeting the requirements for applying pressure to the first and second arteries. Each soft capsule structure can contain one or more soft capsules. Figure 3 In the schematic structure, the first soft capsule structure 131 and the second soft capsule structure 132 each contain two soft capsules. Each soft capsule can be made of an elastic material that can deform under stress. Furthermore, to facilitate the application of pressure from the first soft capsule structure 131 and the second soft capsule structure 132 to the user's finger, elastic materials can be used for the corresponding areas of the inner ring of the smart ring 110 where the first soft capsule structure 131 and the second soft capsule structure 132 are located, while rigid materials can be used for other areas of the ring 110.

[0087] The control component 150 is connected to the soft capsule component 130 and can be used to control the soft capsule component 130 such that the first soft capsule structure 131 of the soft capsule component 130 applies pressure to the first artery of the user's middle finger, and controls the second soft capsule structure 132 to apply pressure to the second artery of the user's middle finger.

[0088] The central control unit 140 can be used to control the control component 150, thereby enabling the soft balloon component to apply pressure to the first and second arteries of the user's middle finger. The central control unit 140 can also acquire data collected by the sensor component 120 and process the acquired data to obtain the user's blood pressure measurement result.

[0089] The central processing unit 140 can be understood as a processor integrated inside the smart ring 100. The central processing unit 140 may have pre-installed algorithms for blood pressure measurement. When the user wears the smart ring 100 on their middle finger, the central processing unit 140 can execute the corresponding blood pressure measurement process based on these algorithms (the specific blood pressure measurement process will be described later). Figure 5 (This is explained) to enable blood pressure measurement for users.

[0090] To achieve blood pressure measurement for the user, when the user wears the smart ring 100 on their middle finger, the central control unit 140 can execute... Figure 5 The following process demonstrates how to measure a user's blood pressure:

[0091] Step S101: Acquire the first data collected by the sensor assembly 120 when the first artery and the second artery are in a non-compression state. The first data includes the first pulse wave waveform collected by each PPG sensor in the PPG sensor group.

[0092] Step S102: Control the soft balloon assembly 130 through the control component 150 to apply a first pressure force to the first artery and the second artery.

[0093] Step S103: Acquire second data collected by sensor assembly 120 when the first artery and the second artery are under compression. The second data includes the second pulse wave waveform collected by each PPG sensor in the PPG sensor group.

[0094] Regarding steps S101-S103 above, it should be understood that after the user wears the smart ring 100 on their middle finger, the physical structure of the smart ring 100 may have a certain degree of influence on the side of the user's middle finger closest to the index finger and the side of the middle finger closest to the ring finger. Therefore, the uncompressed state and the compressed state of the first and second arteries mentioned in this application embodiment can be specifically understood as two different relative states. Compared to the uncompressed state, the first and second arteries in the compressed state are subjected to greater pressure by the soft capsule component 130, and the blood volume flowing through the first and second arteries also decreases compared to the uncompressed state.

[0095] Specifically, the control component 150 can be used to adjust the position of the soft capsule assembly 130, causing the first soft capsule structure 131 and the second soft capsule structure 132 to move closer towards the center of the annulus 110, thereby applying greater pressure to the first and second arteries. In practical applications, the first soft capsule structure 131 and the second soft capsule structure 132 and their initial positions can be calibrated firstly, and when the first soft capsule structure 131 and the second soft capsule structure 132 are in their initial positions, the first and second arteries are considered to be in a non-compressed state. The central control unit 140 can obtain the first data by reading the data collected by the sensor assembly 120 in this state.

[0096] In this embodiment, the first data includes four first pulse wave waveforms collected by the first PPG sensor 121 to the fourth PPG sensor 124 in the PPG sensor group. Specifically, to obtain valid pulse wave waveforms, the central control unit 140 should read the data collected by the sensor assembly 120 during a preset duration while the first and second arteries are in a non-compression state, to ensure that each read first pulse wave waveform contains waveform data from at least one complete cardiac cycle. The specific value of the preset duration can be set according to actual needs; in one example, the preset duration can be set to 8 seconds.

[0097] After obtaining the first data, the central control unit 140 can control the control component 150 to move the first soft capsule structure 131 and the second soft capsule structure 132 closer to the center of the ring body 110 to reach the target position, thereby applying a first compressive force to the first and second arteries of the middle finger. The magnitude of the first compressive force can be pre-calibrated, or the target position of the soft capsule structure 131 and the second soft capsule structure 132 can be considered pre-calibrated. When the first soft capsule structure 131 and the second soft capsule structure 132 are at the target position, the first and second arteries are considered to be under compression. The central control unit 140 obtains the second data by reading the data collected by the sensor component 120 in this state.

[0098] In this embodiment, the second data includes four second pulse wave waveforms collected by the first PPG sensor 121 to the fourth PPG sensor 124 in the PPG sensor group. Similar to the description above, in order to obtain an effective pulse wave waveform, the central control unit 140 should read the data collected by the sensor assembly 120 within a preset duration during the period when the first and second arteries are under compression.

[0099] Step S104: Based on the first data and the second data, determine the blood flow characteristic data, which includes: the first waveform characteristics of each first pulse wave waveform and the second waveform characteristics of each second pulse wave waveform, and / or, the first pulse conduction time difference between first pulse wave waveforms of different phases and the second pulse conduction time difference between second pulse wave waveforms of different phases.

[0100] Among them, waveform features are features related to key parameters such as the shape and peak value of the pulse wave. In this embodiment of the application, waveform features that are related to blood flow volume and / or pressure can be extracted from the obtained first pulse wave waveform and second pulse wave waveform as blood flow feature data.

[0101] Furthermore, due to the different vascular structures in different parts of the human body, the time required for blood to travel from upstream to downstream varies, resulting in phase differences between pulse wave waveforms simultaneously collected from different parts of the body. In this embodiment, the time difference between the in-phase points of pulse wave waveforms from different locations is called the pulse conduction time difference, denoted as ΔPTT. Figure 8 An example is given, showing two pulse wave waveforms with different phases. The horizontal axis of the figure represents time, and the vertical axis represents the amplitude of the pulse wave waveform. The time axes of the two pulse wave waveforms are aligned. It can be seen that the time difference between the peak points of the two pulse wave waveforms is the pulse conduction time difference ΔPTT between the two pulse wave waveforms.

[0102] against Figure 4 In the illustrated scenario, the pulse wave waveforms of the first and second arteries of the middle finger are in phase, while the pulse wave waveforms of the first artery of the middle finger, the third artery of the index finger, and the fourth artery of the ring finger are out of phase with each pair. Current medical research indicates a correlation between the pulse conduction time difference between pulse wave waveforms at different downstream locations and the blood pressure within the arteries. Therefore, in this embodiment, the first pulse conduction time difference between first pulse wave waveforms of different phases and the second pulse conduction time difference between second pulse wave waveforms of different phases can be extracted as blood flow characteristic data.

[0103] Step S105: Determine the user's blood pressure measurement result based on blood flow characteristic data.

[0104] As mentioned earlier, based on Bernoulli's principle and existing medical research experience, it can be concluded that the changes in blood volume and pressure in the middle, index, and ring fingers before and after the first and second arteries of the middle finger are blocked are correlated. Therefore, this application uses the waveform characteristics and / or phase characteristics (pulse conduction time difference between pulse waveforms of different phases) of the pulse wave collected by the PPG sensor array before and after the blockage as blood flow characteristic data. This blood flow characteristic data can characterize the aforementioned changes, thereby enabling the prediction of the user's blood pressure measurement results based on this blood flow characteristic data.

[0105] For example, the blood pressure measurement results to be obtained may specifically include two items: systolic blood pressure (SBP) and diastolic blood pressure (DBP).

[0106] In the initial state of use of the smart ring 100, the first and second arteries of the user's middle finger are in an uncompressed state. The central control unit 140 can first read the first data collected by the sensor component 120 within a preset time period in this state; then, through the control component 150, it controls the soft balloon component 130 to apply pressure to the first and second arteries, so that the first and second arteries are in a compressed state, and then read the second data collected by the sensor component 120 within a preset time period in this state; finally, based on the first and second data, it determines the blood flow characteristic data, and uses a pre-trained blood pressure prediction model to obtain the user's blood pressure measurement result. Based on this process, a single blood pressure measurement for the user can be completed. After completing the current blood pressure measurement, the central control unit 140 can control the soft balloon component 130 to return to the initial state through the control component 150, so that the first and second arteries return to an uncompressed state, thereby enabling the next blood pressure measurement to be performed based on the same process.

[0107] In practical applications, blood pressure measurement intervals can be set within the algorithm threshold of the central control unit 140 to achieve continuous dynamic tracking of blood pressure. For example, blood pressure can be measured every 20 minutes. Alternatively, communication functionality can be configured for the smart ring 100, allowing users to bind the smart ring 100 to electronic devices such as mobile phones and set the blood pressure measurement intervals themselves through these devices.

[0108] As can be seen from the above description, this application utilizes Bernoulli's principle to analyze the blood flow within the finger arteries, combines the analysis results with existing medical research experience, and proposes a novel approach to blood pressure measurement. Based on this approach, a smart ring for measuring a user's blood pressure is provided. The smart ring includes a soft capsule component for applying pressure to the first and second arteries of the user's middle finger, a control component for controlling the soft capsule component, a PPG sensor group for measuring the pulse wave waveform of blood within the user's middle, index, and ring fingers, and a central control unit. When the user wears the smart ring on their middle finger, the central control unit can apply pressure to the first and second arteries through the control component, achieving a flow throttling effect. Based on this, the user's blood pressure measurement result is determined by using the waveform characteristics and / or phase characteristics of the pulse wave waveform collected by the PPG sensor group before and after the flow throttling as blood flow characteristic data.

[0109] Compared to cuff-type blood pressure measuring devices, blood pressure measurement using the smart ring provided in this application embodiment eliminates the need for inflating and deflating a cuff, making the measurement process simpler and more convenient. Furthermore, because the smart ring itself is easy to wear, it can be used for continuous dynamic blood pressure measurement in daily life to track changes in blood pressure. Moreover, during the measurement of a user's blood pressure using the smart ring, since the first and second arteries within the middle finger are very small, a large amount of pressure is not required to achieve a blockage effect on the first and second arteries, resulting in minimal discomfort for the user. Therefore, it is also suitable for nighttime blood pressure measurement.

[0110] Compared with existing blood pressure measurement technologies based on single-channel pulse wave waveforms, the smart ring provided in this application comprehensively considers the multiple pulse wave waveforms collected by each PPG sensor in the PPG sensor group before and after the first and second arteries are blocked. Blood pressure prediction is based on the waveform characteristics and / or phase characteristics of these multiple pulse wave waveforms, which helps to ensure the accuracy of the obtained blood pressure measurement results.

[0111] In practical applications, the smart ring provided in this application embodiment can be applied to daily blood pressure measurement, such as daily single or continuous measurements, and nighttime blood pressure measurement, depending on specific needs. Furthermore, this smart ring can track blood pressure changes through continuous dynamic measurement, which helps in screening for hard hypertension and can be applied to hypertension screening in the general population. This smart ring can also be used to track both shaped and non-shaped blood pressure, helping people with hypertension manage their blood pressure.

[0112] In one possible application scenario of this application, the smart ring 100 can establish a communication connection with electronic devices such as mobile phones, laptops, and tablets via wireless means, and bind the smart ring 100 to the electronic device. For example, the selected wireless communication connection method can be Bluetooth (BT), Wireless Fidelity (WIFI), Narrow Band Internet of Things (NB IoT), cellular communication networks, etc.

[0113] In practical applications, a dedicated application for managing the smart ring 100 can be developed and installed on electronic devices such as mobile phones. Once the smart ring 100 is paired with this electronic device, the user can use the application to configure its functions, such as setting the blood pressure measurement interval. Alternatively, the user can use the application to view the blood pressure readings from the smart ring 100 to understand their health status.

[0114] Figures 6-7 An example is given. See also Figure 6 Assuming the electronic device is a smartphone 200, after installing the application for managing the smart ring 100, the user can do as follows: Figure 6 As shown in (a), open the application and enter... Figure 6 (b) in the diagram illustrates the main interface of the application. Figure 6 In illustration (b), the main interface includes two options: "Device Discovery" and "Measurement Record Query." When a user first uses the smart ring 100, they can select the "Device Discovery" option, allowing the smartphone 200 to discover the smart ring 100 through its device discovery function. After the smartphone 200 discovers the smart ring 100, it will display the following... Figure 6 (c) shows the discovery results interface, and displays the name of Smart Ring 100 as "Smart Ring" in the "Discovered Devices" list. Users can click on "Smart Ring" to... Figure 6 As illustrated in (d), a pop-up window will appear in the results interface for the smart ring. The user can select the "Connect" option to bind the smart ring 100 to the smartphone 200. In practical applications, the wireless communication technology mentioned earlier can be used to enable the smartphone 200 to discover the smart ring 100 and to bind the smartphone 200 to the smart ring 100.

[0115] With the smart ring 100 already bound to the smartphone 200, after the smart ring 100 obtains the user's blood pressure measurement result based on the aforementioned steps S101-S105, it can send the blood pressure measurement result to the smartphone 200. The smartphone 200 can then generate a blood pressure measurement record based on the obtained result for the user to query. See [link to relevant documentation] Figure 7 , Figure 7 (a) shows the main interface of the application used to manage the smart ring 100, which users can access by clicking the "Measurement Record Query" option. Figure 7 The blood pressure measurement results are shown in diagram (b). As mentioned earlier, the Smart Ring 100 can measure blood pressure at regular time intervals; therefore, the results can be displayed to the user as a line graph. Figure 7 The example given in (b) shows the systolic and diastolic blood pressure values ​​(BP) measured by the smart ring 100 during the period from 12:20 to 13:40, which allows users to intuitively see their own blood pressure changes.

[0116] In some embodiments of this application, the relationship between blood flow feature data and a user's blood pressure can be learned using machine learning methods, resulting in a blood pressure prediction model that can predict blood pressure measurement results based on the blood flow feature data. Therefore, for the aforementioned step S105, the blood flow feature data obtained in step S104 can be processed using the trained blood pressure prediction model to obtain the user's blood pressure measurement results.

[0117] The blood pressure prediction model takes the user's blood flow characteristic data as input and predicts the user's blood pressure based on this data. In practical applications, steps S101-S104 described above can be used to collect sample blood flow characteristic data from sample users and obtain their actual blood pressure measurements. A training set is then established based on the collected sample blood flow characteristic data and the actual blood pressure measurements. The neural network model is trained using this training set to obtain the desired blood pressure prediction model. Further details regarding the training process will be provided later.

[0118] Based on the pre-trained blood pressure prediction model, in step S105, the blood flow characteristic data obtained in step S104 is input into the blood pressure prediction model to obtain the user's blood pressure measurement result.

[0119] Specifically, the training process for the blood pressure prediction model can be executed on a server. In one example, after training the blood pressure prediction model, it can be copied to the central control unit 140. Then, after the central control unit 140 obtains the user's blood flow characteristic data based on steps S101-S104, it can input this data into the blood pressure prediction model to obtain the user's blood pressure measurement result. In another example, the trained blood pressure prediction model can be deployed on a server, and a communication connection can be established between the smart ring 100 and the server via wireless communication methods such as Bluetooth or Wi-Fi. After the central control unit 140 obtains the user's blood flow characteristic data based on steps S101-S104, it can report this data to the server. The server then inputs the blood flow characteristic data into the blood pressure prediction model, and subsequently, the server can return the obtained blood pressure measurement result to an electronic device such as a mobile phone bound to the smart ring 100, allowing the user to query the blood pressure measurement result.

[0120] In some embodiments of this application, see Figures 9-10The PPG sensor group of sensor assembly 120 also includes a fifth PPG sensor 125 and a sixth PPG sensor 126. The fifth PPG sensor 125 and the sixth PPG sensor 126 are both arranged along the inner ring of the ring body 110 and are opposite to each other. Specifically, the fifth PPG sensor 125 and the sixth PPG sensor 126 are located in the normal direction of the line from the first PPG sensor 121 to the fourth PPG sensor 124. (The last sentence appears to be incomplete and possibly refers to a user-based system.) Figure 4 When the smart ring 100 is configured as shown in (a), the fifth PPG sensor 125 is used to measure the pulse wave waveform at the back of the middle finger, and the sixth PPG sensor 126 is used to measure the pulse wave waveform at the pad of the middle finger.

[0121] When the PPG sensor group contains six PPG sensors, the first data acquired by the central control unit 140 includes: the first pulse wave waveforms collected by the first PPG sensor 121 to the sixth PPG sensor 126 respectively when the first artery and the second artery are in a non-compression state; the second data includes: the second pulse wave waveforms collected by the first PPG sensor 121 to the sixth PPG sensor 126 respectively when the first artery and the second artery are in a compression state.

[0122] Similarly, after acquiring the first and second data, the central control unit 140 can determine blood flow characteristic data for predicting blood pressure based on the first and second data. The blood flow characteristic data includes: the first waveform characteristics of six first pulse wave waveforms collected by the first PPG sensor 121 to the sixth PPG sensor 126 respectively, and the second waveform characteristics of six second pulse wave waveforms collected by the first PPG sensor 121 to the sixth PPG sensor 126 respectively, and / or, the first pulse conduction time difference between first pulse wave waveforms of different phases and the second pulse conduction time difference between second pulse wave waveforms of different phases. Specifically, when the user... Figure 10 When the illustrated smart ring is worn on the middle finger, the pulse wave waveforms collected by the first PPG sensor 121, the second PPG sensor 122, the fifth PPG sensor 125, and the sixth PPG sensor 126 are in phase.

[0123] In reality, the palmar digital artery is not located on the back or pad of the middle finger. However, because the fifth PPG sensor 125 and the sixth PPG sensor 126 are in close contact with the skin of the middle finger, they can still detect the blood flow in the middle finger artery to some extent. By collecting pulse wave waveforms from different locations on the middle finger using the first PPG sensor 121, the second PPG sensor 122, the fifth PPG sensor 125, and the sixth PPG sensor 126, the pulse wave waveforms complement each other, which can more accurately reflect the blood flow in the middle finger artery, thus helping to improve the accuracy of the final predicted blood pressure measurement result.

[0124] In some embodiments of this application, see Figures 11-12 The control component 150 includes a winch motor 151, a winch 152, and a stranded wire 153.

[0125] One end of the stranded wire 153 is connected to the winch 152 and is arranged in a clockwise or counterclockwise direction inside the smart ring 100 and wrapped around the soft pouch assembly 130. The soft pouch assembly 130 includes a first soft pouch structure 131 and a second soft pouch structure 132. The other end of the stranded wire 153 is connected to a fixing member (marked as O in the figure) disposed inside the smart ring 100.

[0126] The winch motor 151 is connected to the winch 152 and is used to drive the winch 152 to rotate to tighten or loosen the strand 153. The soft capsule assembly 130 applies pressure to the first artery and the second artery as the strand 153 is tightened, and relaxes the pressure applied to the first artery and the second artery as the strand 153 is loosened.

[0127] against Figure 12 The schematic diagram shows that one end of the stranded wire 153 is connected to the winch 152, arranged clockwise inside the smart ring 100, and sequentially wound around the second soft pouch structure 132 and the first soft pouch structure 131. The other end of the stranded wire is connected to a fixing member. When the winch 152 rotates in the first direction, the stranded wire 153 is further wound around the winch 152, causing the stranded wire 153 to be tightened. As the stranded wire 153 tightens, the second soft pouch structure 132 and the first soft pouch structure 131 will shift further towards the user's middle finger, generating pressure on the first and second arteries. When the winch 152 rotates in the second direction, a portion of the wound stranded wire 153 will be released, causing the stranded wire 153 to loosen. As the stranded wire 153 loosens, the pressure on the first and second arteries will also be relieved. In practical applications, the first direction can be clockwise and the second direction can be counterclockwise, or the first direction can be counterclockwise and the second direction can be clockwise, depending on the specific winding method of the stranded wire 153.

[0128] In some embodiments of this application, to ensure the smooth tightening and loosening process of the twisted wire 153 within the smart ring 100, such as... Figure 12 As illustrated, a fixed pulley structure 154 can be set inside the smart ring 100.

[0129] against Figures 11-12 The smart ring shown, after the central control unit 140 reads the first data collected by the sensor assembly 120 when the first and second arteries are in a non-compressed state, can send a first control message to the winch motor 151, causing the winch motor 151 to rotate a specific amount of rotation in a first direction, so that the soft pouch assembly 130 can apply a first pressure force to the first and second arteries as the strand 153 tightens. Similarly, after the central control unit 140 reads the first data collected by the sensor assembly 120 when the first and second arteries are in a compressed state, it can send a second control message to the winch motor 151, causing the winch motor 151 to rotate a specific amount of rotation in a second direction, so that the soft pouch assembly 130 can relax the pressure applied to the first and second arteries as the strand 153 relaxes.

[0130] In some embodiments of this application, see Figures 13-14 The sensor assembly 120 also includes a pressure sensor 127, which also serves as a fastener O for securing the stranded wire 153. The tension in the stranded wire 153 generates pressure applied to the pressure sensor 127, which can be used to measure this pressure.

[0131] In this embodiment, the tension of the stranded wire differs between the uncompressed state and the compressed state of the first and second arteries, resulting in different pressure readings from the pressure sensor 127. Furthermore, there is a correlation between the pressure at the first and second arteries and the pressure measured by the pressure sensor 127. For example, if a user experiences higher pressure at the first and second arteries, greater pressure may be required through the soft capsule assembly 130 to achieve the desired flow obstruction, thus potentially leading to a greater variation in pressure measured by the pressure sensor 127 between the two states.

[0132] Therefore, in this embodiment, the first data read by the central control unit 140 further includes: a first pressure value collected by the pressure sensor 127 when the first and second arteries are in a non-compressed state; the second data further includes: a second pressure value collected by the pressure sensor 127 when the first and second arteries are in a non-compressed state; the blood flow characteristic data further includes: the second pressure value and the first pressure value. The first pressure value and the second pressure value can characterize the change in pressure measured by the pressure sensor 127 before and after the first and second arteries are blocked by the soft balloon assembly 130. By using the first pressure value and the second pressure value as part of the blood flow characteristic data for predicting the user's blood pressure, it helps to further improve the accuracy of the predicted blood pressure measurement results.

[0133] In some embodiments of this application, see Figure 15 The sensor assembly 120 also includes an acceleration sensor 128.

[0134] The accelerometer 128 can detect the acceleration information of the smart ring 100 in various directions (generally the x-axis, y-axis, and z-axis), and can be used to identify the user's motion state. Generally, when the acceleration values ​​collected by the accelerometer 128 in all directions are less than a threshold, the smart ring 100 is considered to be in a stationary state, that is, the user wearing the smart ring is in a stationary state; otherwise, the user is considered to be in motion.

[0135] Specifically, during human movement, blood pressure changes due to the movement. Therefore, to ensure the accuracy of blood pressure measurement results, it is necessary to ensure that the sensor data used to predict blood pressure results, such as the pulse wave waveform collected by the PPG sensor, are collected when the user is at rest.

[0136] Therefore, in this embodiment, in step S101, the first data to be acquired by the central control unit 140 further includes: first acceleration information collected by the accelerometer 128 when the first artery and the second artery are in a non-compressed state. After acquiring the first data, the central control unit 140 can determine whether the user is in motion during the information acquisition process of the sensor component 120 based on the first acceleration information. If so, the acquired first data is determined to be invalid, and the current blood pressure measurement is terminated; if not, the central control unit 140 can continue to execute the subsequent steps S102-S105.

[0137] Similarly, in step S103, the second data to be acquired by the central control unit 140 also includes: second acceleration information collected by the accelerometer 128 when the first and second arteries are under compression. After acquiring the second data, the central control unit 140 can determine whether the user was in motion during the information acquisition process of the sensor assembly 120 based on the second acceleration information. If so, the acquired second data is determined to be invalid, and the current blood pressure measurement is terminated; otherwise, the central control unit 140 can continue to execute the subsequent steps S104-S105.

[0138] In one example, if the central control unit 140 terminates the current blood pressure measurement while executing step S101 or step S103, the central control unit 140 can wait until the next blood pressure measurement opportunity to perform a measurement. For example, if the interval for blood pressure measurement is set to 20 minutes, if the central control unit 140 terminates the current blood pressure measurement due to user movement, the central control unit 140 can perform a new blood pressure measurement 20 minutes later based on the aforementioned process of steps S101-S105.

[0139] In some embodiments of this application, the magnitude of the first pressure applied by the soft balloon assembly 130 to the first and second arteries in step S102 can be obtained through pre-calibration. To calibrate the first pressure, the central control unit 140 is further configured to control the soft balloon assembly 130 via the control component 150, causing the soft balloon assembly 130 to apply a second pressure to the user's first and second arteries, and to monitor the peak value of the third pulse wave waveform collected by the first PPG sensor 121 and the second PPG sensor 122 during the application of the second pressure; the magnitude of the pressure that causes the peak value of the third pulse wave waveform to be lower than a preset threshold is calibrated as the magnitude of the first pressure.

[0140] In this embodiment of the application, when the peak value of the pulse wave waveform at the first artery collected by the first PPG sensor 121 is less than a preset threshold, and the peak value of the pulse wave waveform at the second artery collected by the second PPG sensor 121 is less than a preset threshold, it can be considered that the throttling of the first artery and the second artery has been achieved.

[0141] In practical applications, the preset threshold value and the magnitude of the second pressure applied to the first and second arteries via the soft capsule component 130 can be configured at the factory of the smart ring 100. After the user wears the smart ring 100, the central control unit 140 can perform the first blood pressure measurement on the user according to a process similar to the aforementioned steps S101-S105. The only difference between this process and steps S101-S105 is that the pressure used in step S103 is the second pressure value configured at the factory, rather than the first pressure value to be calibrated. It is understandable that the process of the soft balloon assembly 130 applying pressure to the first and second arteries is actually a dynamic process. That is, the force applied by the soft balloon assembly 130 actually needs a period of time to gradually increase to the second pressure. The central control unit 140 can determine the magnitude of the pressure that can achieve the blocking effect on the first and second arteries by monitoring the peak value of the pulse wave waveform collected by the first PPG sensor 121 and the second PPG sensor 122 during this process, and calibrate the determined pressure magnitude as the magnitude of the first pressure. Therefore, when performing subsequent blood pressure measurements, blood pressure measurements can be performed based on the calibrated first pressure, according to the aforementioned steps S101-S105.

[0142] Furthermore, regarding Figure 11 as well as Figure 12 In the schematic structure, the magnitude of the pressure applied by the soft capsule assembly 130 to the first and second arteries is determined by the rotation amount of the winch 152. Therefore, a preset second rotation amount can be set at the factory, and the pressure achievable by this second rotation amount is considered to be the magnitude of the second pressure. Then, when the central control unit 140 performs the first blood pressure measurement on the user, it calibrates the magnitude of the first rotation amount that can achieve the blocking of the first and second arteries, and considers the pressure achievable by this first rotation amount to be the magnitude of the first pressure. Thus, in subsequent blood pressure measurements, when executing step S102, the central control unit 140 controls the winch motor 151 to drive the winch 152 to achieve the first rotation amount through the first control message.

[0143] In practical applications, the blood flow in the finger arteries may differ among different users, thus the required initial pressure force to achieve throttling of the first and second arteries in the middle finger may also vary. In this embodiment, the initial pressure force is calibrated based on each user's specific situation. During actual blood pressure measurement, the calibrated initial pressure force is applied to the first and second arteries to compress them. Blood pressure prediction is then performed based on the initial data collected by the sensor assembly 120 under this compressed state, which helps to further ensure the accuracy of the blood pressure measurement results.

[0144] In some embodiments of this application, the first waveform feature involved in the aforementioned step S104 includes: the rising area of ​​the first pulse wave waveform collected by each PPG sensor in the PPG sensor group; the second waveform feature includes: the rising area of ​​the second pulse wave waveform collected by each PPG sensor in the PPG sensor group.

[0145] See the concept of rising area. Figure 16 , Figure 16 The diagram shows the pulse wave waveform within a cardiac cycle, with the horizontal axis representing time and the vertical axis representing amplitude. Draw a line segment parallel to the horizontal axis along the starting point of the pulse wave waveform, and then draw a line segment parallel to the vertical axis along the dicrotic notch of the pulse wave waveform. The area enclosed by these two line segments and the pulse wave waveform is the rising area S0.

[0146] Therefore for Figures 2-3 In the case of the smart ring 100 shown, after obtaining the four first pulse wave waveforms collected by the first PPG sensor 121 to the fourth PPG sensor 124 respectively, it is necessary to determine the rising area of ​​the four pulse wave waveforms as the first waveform feature of each first pulse wave waveform; after obtaining the four second pulse wave waveforms collected by the first PPG sensor 121 to the fourth PPG sensor 124 respectively, it is also necessary to determine the rising area of ​​the four pulse wave waveforms as the second waveform feature of each second pulse wave waveform.

[0147] Specifically, the rising area of ​​a pulse wave reflects the blood volume within an artery. As explained earlier in the section on blood pressure measurement principles, after severing the first and second arteries in the middle finger, the change in blood volume flowing through the index and ring fingers is related to the pressure at the superficial palmar arch, which in turn is related to the blood pressure at the user's brachial artery. Therefore, by using the rising area of ​​each first pulse wave waveform acquired by the PPG sensor group as the first waveform feature and the rising area of ​​each second pulse wave waveform acquired by the PPG sensor group as the second waveform feature, the obtained first and second waveform features can reflect the change in blood volume flowing through the index and ring fingers after severing the first and second arteries. This helps ensure the accuracy of blood pressure measurement results when making predictions based on the first and second waveform features.

[0148] In some embodiments of this application, in addition to using the rising area of ​​the first pulse wave waveform collected by each PPG sensor in the PPG sensor group as the first waveform feature, the first waveform feature to be extracted from each first pulse wave waveform also includes one or more of the following: the 1 / LASI value, 1 / LASI_2 value, PIR value, and RIPV value of the first pulse wave waveform collected by each PPG sensor in the PPG sensor group.

[0149] For the concepts of 1 / LASI value, 1 / LASI_2 value, PIR value, and RIPV value, please refer to [link / reference needed]. Figure 17 The graph shows the pulse wave waveform within a cardiac cycle, with time on the horizontal axis and amplitude on the vertical axis. It can be seen that 1 / LASI value is the time difference between the peak point and the dicrotic notch; 1 / LASI_2 value is the time difference between point M and the dicrotic notch. Point M is the point with the steepest slope on the pulse wave waveform, meaning... Figure 17 The first derivative image of the pulse wave waveform shown has a maximum value at point M.

[0150] also, Figure 17 The diagram also shows the starting amplitude S1, ending amplitude E1, and peak amplitude P1 of the pulse wave waveform. The calculation methods for PIR and RIPV values ​​are: PIR = P1 / S1, RIPV = P1 / E1.

[0151] Existing medical research has shown a correlation between the 1 / LASI, 1 / LASI_2, PIR, and RIPV values ​​of the pulse wave waveform and blood pressure at the brachial artery. Therefore, in this embodiment, by extracting the 1 / LASI, 1 / LASI_2, PIR, and RIPV values ​​of the first pulse wave waveform collected by each PPG sensor in the PPG sensor group as the first waveform feature, the accuracy of the blood pressure measurement results can be further ensured when combining the first waveform feature with the second waveform feature for subsequent blood pressure prediction.

[0152] In some embodiments of this application, the first pulse conduction time difference involved in the aforementioned step S104 includes: the pulse conduction time difference between the first pulse waveform collected by the first PPG sensor 121 and the first pulse waveform collected by the third PPG sensor 123, and the pulse conduction time difference between the first pulse waveform collected by the second PPG sensor 122 and the fourth PPG sensor 124; the second pulse conduction time difference includes: the pulse conduction time difference between the second pulse waveform collected by the third PPG sensor 123 and the fourth PPG sensor 124.

[0153] To facilitate understanding, the following will be combined with Figure 18 An illustrative explanation is provided for confirming the first pulse conduction time difference and the second pulse conduction time difference. Figure 18 This illustration shows an example of the pulse wave waveforms collected by the first PPG sensor 121 to the sixth PPG sensor 126 (referred to as PPG1 to PPG6 sequentially) when the user wears the smart ring 100 on their middle finger. Figure 18 (a) shows the first pulse wave waveforms of PPG1 to PPG6 when the first and second arteries were in an uncompressed state. Figure 18 (b) shows the second pulse wave waveforms collected by PPG1 to PPG6 when the first and second arteries were under compression.

[0154] As explained above, the pulse wave waveforms acquired by PPG1, PPG2, PPG5, and PPG6 for measuring the pulse wave waveform of the internal artery of the middle finger are in phase, while the pulse wave waveforms acquired by PPG1, PPG3, and PPG4 are out of phase in each pair. For each first pulse wave waveform, this embodiment specifically extracts the pulse conduction time difference between the first pulse wave waveform acquired by PPG1 and the first pulse wave waveform acquired by PPG3 (which is substantially the same as the pulse conduction time difference between the first pulse wave waveform acquired by PPG2 and the first pulse wave waveform acquired by PPG3), denoted as ΔPTT_13, and the pulse conduction time difference between the first pulse wave waveform acquired by PPG2 and the first pulse wave waveform acquired by PPG4 (which is substantially the same as the pulse conduction time difference between the first pulse wave waveform acquired by PPG1 and the first pulse wave waveform acquired by PPG3), denoted as ΔPTT_24. Combined with... Figure 18 It can be seen that since the pulse wave waveforms acquired by PPG1 and PPG2 are in phase, ΔPTT_13 and ΔPTT_24 can actually reflect the phase relationship between the first pulse wave waveform acquired by PPG3 and the first pulse wave waveform acquired by PPG4. Therefore, there is no need to calculate other pulse conduction time differences to characterize the phase relationship between blood flow in the middle, index and ring fingers, which helps to reduce data redundancy.

[0155] For each second pulse wave waveform, this embodiment of the application specifically extracts the pulse conduction time difference between the second pulse wave waveform acquired by PPG3 and the second pulse wave waveform acquired by PPG4, denoted as ΔPTT_34. Specifically, since the first and second arteries of the middle finger are already in a state of shunting during the acquisition of the second pulse wave waveform, further analysis of the phase relationship between the blood in the middle finger and the blood in the index finger, or the phase relationship between the blood in the middle finger and the blood in the ring finger, is of no reference value. Therefore, only the pulse conduction time difference ΔPTT_34 is extracted.

[0156] To facilitate understanding of the technical solutions of the above embodiments of this application, the following will be provided: Figure 18 Based on the combination Figure 19 The process of measuring a user's blood pressure using the smart ring 100 is further illustrated by an example. Figure 19 The process of a single blood pressure measurement is shown when the smart ring 100 is equipped with a PPG sensor group including a first PPG sensor 121 to a sixth PPG sensor 126, a pressure sensor 127, and an acceleration sensor 128.

[0157] Specifically, before starting blood pressure measurement, the sensors can be initialized first, and then the first stage of blood pressure measurement can begin. During this first stage, the user's first and second arteries are in a non-compressed state. The central control unit 140 reads sensor data collected by the sensor assembly during a certain period (e.g., 8 seconds) in the first stage. The read sensor data includes: first acceleration information collected by the accelerometer 128, first pulse wave waveforms collected by each PPG sensor in the PPG sensor group, and first pressure values ​​collected by the pressure sensor 127. After reading the first data, the central control unit 140 can first determine whether the user is stationary based on the first acceleration information; if so, it further extracts the required blood flow characteristic data from the first data. The blood flow characteristic data extracted from the first data include: the first pressure value, the pulse conduction time difference ΔPTT_13 between the first pulse wave waveform collected by PPG1 and the first pulse wave waveform collected by PPG3, the pulse conduction time difference ΔPTT_24 between the first pulse wave waveform collected by PPG2 and the first pulse wave waveform collected by PPG4, and the rising area, 1 / LASI value, 1 / LASI_2 value, PIR value, and RIPV value extracted from the six first pulse wave waveforms collected by PPG1 to PPG6, respectively.

[0158] After the sensor assembly 120 acquires the sensor data required for the first stage, the central control unit 140 controls the winch 152 to rotate via the winch motor 151, tightening the stranded cable 153. This causes the stranded cable 153 to press against the soft capsule structures (first soft capsule structure 131 and second soft capsule structure 132), applying a first compressive force to the first and second arteries. In practical applications, the process of controlling the soft capsule structures to apply the first compressive force to the first and second arteries requires a certain amount of time; for example, this process may take 2 seconds.

[0159] After the soft balloon structure applies the first compression force to the first and second arteries, the second stage of blood pressure measurement begins, during which the user's first and second arteries are under compression. The central control unit 140 reads sensor data collected by the sensor assembly during a certain duration (e.g., 8 seconds) in the second stage. The read sensor data includes: second acceleration information collected by the accelerometer 128, second pulse wave waveforms collected by each PPG sensor in the PPG sensor group, and second pressure values ​​collected by the pressure sensor 127. After reading the second data, the central control unit 140 can first determine whether the user is stationary based on the second acceleration information. If so, it further extracts the required blood flow characteristic data from the second data. The blood flow characteristic data extracted from the second data includes: the second pressure value, the pulse conduction time difference ΔPTT_34 between the second pulse wave waveforms collected by PPG3 and PPG4, and the area of ​​rise extracted from each of the six second pulse wave waveforms collected by PPG1 to PPG6.

[0160] Finally, the blood flow feature data extracted from the first data and the blood flow feature data extracted from the second data are input into the pre-trained blood pressure prediction model to obtain the user's blood pressure measurement results, which include SBP and DBP.

[0161] Furthermore, based on Figure 8 It can be seen that the pulse wave waveform acquired by the PPG sensor in the first or second stage may contain multiple cardiac cycles. In this case, one cardiac cycle can be selected to determine the waveform characteristics and / or phase characteristics of each pulse wave waveform; alternatively, the waveform characteristics and / or phase characteristics within each cardiac cycle can be determined separately, and the waveform characteristics and / or phase characteristics used for blood pressure prediction can be determined by processing the waveform characteristics and / or phase characteristics within each cardiac cycle, such as by averaging.

[0162] In practical applications, based on the above embodiments, other functional modules can be further integrated into the smart ring 100 according to specific needs. Figure 20 An illustration is provided, based on the above embodiments, Figure 20 It further includes a Bluetooth module 160, a touch control unit 180, a power supply 170, and a sensor assembly 120, which also includes a gyroscope sensor 129.

[0163] The Bluetooth module 160 is used to enable communication between the smart ring 100 and external devices. For example, the smart ring 100 can connect to a mobile phone via the Bluetooth module 160, allowing the smart ring 100 to share blood pressure measurement results to the mobile phone, so that the user can check the blood pressure measurement results through the mobile phone.

[0164] The touch control unit 180 can be used to enable users to control the smart ring 100 via touch. For example, users can be set to turn the smart ring 100 on or off, or instruct the smart ring 100 to perform active blood pressure measurement, by performing specific touch actions (such as single click, double click, etc.).

[0165] The power supply 170 can be used to power the Bluetooth module 160, winch motor 151, central control unit 140, sensor assembly 120 and other components within the smart ring 100.

[0166] Based on the same inventive concept, this application also provides a method for measuring blood pressure, see [link to relevant documentation]. Figure 21 The method includes:

[0167] Step S201: Acquire first data collected from the user when the first artery on the side of the middle finger near the index finger and the second artery on the side of the middle finger near the ring finger are in a non-compression state; the first data includes: first pulse wave waveforms collected at multiple preset collection sites on the user's hand; the preset collection sites include: the first artery, the second artery, the third artery on the side of the index finger near the middle finger, and the fourth artery on the side of the ring finger near the middle finger.

[0168] Step S202: Acquire second data collected from the user when the user's first and second arteries are under compression; the second data includes: second pulse waveforms collected at each preset collection site on the user's hand.

[0169] Step S203: Based on the first data and the second data, determine blood flow characteristic data; the blood flow characteristic data includes: the first waveform characteristics of each first pulse wave waveform and the second waveform characteristics of each second pulse wave waveform, and / or, the first pulse conduction time difference between first pulse wave waveforms of different phases and the second pulse conduction time difference between second pulse wave waveforms of different phases.

[0170] Step S204: Determine the user's blood pressure measurement result based on blood flow characteristic data.

[0171] In one example, this blood pressure measurement method can be applied to the smart ring 100 provided in the aforementioned embodiments.

[0172] In one embodiment of this application, the preset collection sites further include: the back of the middle finger and the pad of the middle finger.

[0173] In one embodiment of this application, when the first artery and the second artery are under compression, the magnitude of the first compressive force on the first artery and the second artery is pre-calibrated. The calibration process is as follows: a second compressive force is applied to the user's first artery and the second artery, and the peak value of the third pulse wave waveform collected at the first artery and the second artery is monitored during the application of the second compressive force. The magnitude of the compressive force that causes the peak value of the third pulse wave waveform to be lower than a preset threshold is calibrated as the magnitude of the first compressive force.

[0174] In one embodiment of this application, the first pulse conduction time difference includes: the pulse conduction time difference between the first pulse waveform acquired at the first artery and the first pulse waveform acquired at the third artery, and the pulse conduction time difference between the first pulse waveform acquired at the second artery and the first pulse waveform acquired at the fourth artery; the second pulse conduction time difference includes: the pulse conduction time difference between the second pulse waveform acquired at the third artery and the second pulse waveform acquired at the fourth artery.

[0175] In one embodiment of this application, the first waveform feature includes the rising area of ​​the first pulse wave waveform collected at each preset acquisition location; the second waveform feature includes the rising area of ​​the second pulse wave waveform collected at each preset acquisition location.

[0176] In one embodiment of this application, the first waveform feature further includes one or more of the following: the 1 / LASI value, 1 / LASI_2 value, PIR value, and RIPV value of the first pulse wave waveform acquired at each preset acquisition location.

[0177] For more details on this blood pressure measurement method and its beneficial effects, please refer to the previous description of the smart ring embodiment, which will not be repeated here.

[0178] In one embodiment of this application, for the aforementioned step S204, the user's blood pressure measurement result can be obtained by inputting blood flow feature data into a pre-trained blood pressure prediction model.

[0179] Among them, blood pressure prediction models can be used Figure 22 The following steps are illustrated in the training process:

[0180] Step S301: Obtain sample first data collected from the sample user when the first artery on the side of the middle finger near the index finger and the second artery on the side of the middle finger near the ring finger are in a non-compression state; the sample first data includes: sample first pulse wave waveforms collected at multiple preset collection sites on the sample user's hand.

[0181] The preset collection location here is the same as the preset collection location involved in the aforementioned step S201.

[0182] Step S302: Obtain sample second data collected from the sample user when the first and second arteries of the sample user are under compression; the sample second data includes sample second pulse wave waveforms collected at multiple preset collection sites on the sample user's hand.

[0183] Step S303: Based on the first sample data and the second sample data, determine the blood flow characteristic data of the sample users; the blood flow characteristic data includes: the first waveform characteristics of the first pulse wave waveform of each sample and the second waveform characteristics of the second pulse wave waveform of each sample, and / or, the first pulse conduction time difference between the first pulse wave waveforms of different phases and the second pulse conduction time difference between the second pulse wave waveforms of different phases.

[0184] Step S304: Input the blood flow feature data of the sample into the blood pressure prediction model to be trained, and obtain the predicted blood pressure measurement results of the sample users.

[0185] Step S305: Based on the predicted blood pressure measurement results of the sample users and the actual blood pressure measurement results of the sample users, train the blood pressure prediction model to be trained to obtain the trained blood pressure prediction model.

[0186] In practical applications, based on steps S301-S302 above, sample first data and sample second data can be collected from a large number of sample users. Then, based on step S303, sample blood flow feature data corresponding to each sample user can be obtained, and the real blood pressure measurement result of each sample user can be used as the ground truth label corresponding to the sample blood flow feature data of that sample user, thus completing the acquisition of training data. In one example, the user's real blood pressure measurement result can be measured using an electronic blood pressure monitor or a mercury sphygmomanometer.

[0187] After obtaining the training data, the blood flow feature data of the samples in the training data is input into the blood pressure prediction model to be trained, thereby obtaining the blood pressure measurement result predicted by the blood pressure prediction model for the blood flow feature data of the samples. In this embodiment, the model loss of the blood pressure prediction model can be determined based on the blood pressure measurement result predicted by the blood pressure prediction model for the blood flow feature data of the samples, and the actual blood pressure measurement result corresponding to the blood flow feature data of the samples. The model parameters of the blood pressure prediction model are then updated based on the model loss, thereby achieving the training of the blood pressure prediction model.

[0188] The model structure and specific form of the loss function used in the blood pressure prediction model can be selected according to actual needs. For example, the blood pressure prediction model can be a neural network model, a linear regression model, a logistic regression model, etc. The loss function can be the accuracy of the blood pressure measurement result predicted by the blood pressure prediction model, or the mean squared error between the predicted blood pressure measurement result and the actual blood pressure measurement result, etc.

[0189] After obtaining the trained blood pressure prediction model based on the above steps S301-S305, the trained blood pressure prediction model can be applied to the smart ring in the aforementioned embodiment.

[0190] In one example, after the actual production of the smart ring 100 provided in the foregoing embodiments of this application is completed, the smart ring 100 can be used to execute steps S101-S103 mentioned above to obtain the training data required for the model training process.

[0191] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of some embodiments of this application and are not intended to limit it. Although some embodiments of this application have been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of some implementation examples of this application without departing from the spirit and scope of the technical solutions of some embodiments of this application.

Claims

1. A smart ring, characterized in that, The smart ring is worn on the user's middle finger and includes a central control unit, a sensor assembly, a soft capsule assembly, and a control assembly. The sensor assembly includes a PPG sensor group; the PPG sensor group includes a first PPG sensor, a second PPG sensor, a third PPG sensor, and a fourth PPG sensor; the first PPG sensor and the second PPG sensor are arranged along the inner ring of the smart ring and opposite each other, and are respectively used to measure the pulse wave waveform of the first artery on the side of the middle finger closer to the index finger, and the pulse wave waveform of the second artery on the side of the middle finger closer to the ring finger; the third PPG sensor is arranged along the outer ring of the smart ring and opposite the first PPG sensor, and is used to measure the pulse wave waveform of the third artery on the side of the index finger closer to the middle finger; the fourth PPG sensor is arranged along the outer ring of the smart ring and opposite the second PPG sensor, and is used to measure the pulse wave waveform of the fourth artery on the side of the ring finger closer to the middle finger. The soft capsule assembly includes a first soft capsule structure and a second soft capsule structure; the first soft capsule structure is arranged around the first PPG sensor, and the second soft capsule structure is arranged around the second PPG sensor. The control component is connected to the soft capsule component and is used to control the first soft capsule structure to apply pressure to the first artery and to control the second soft capsule structure to apply pressure to the second artery. The central control unit is configured to acquire first data collected by the sensor assembly when the first artery and the second artery are in a non-compression state, the first data including: first pulse wave waveforms collected by each PPG sensor in the PPG sensor group; control the soft balloon assembly through the control component to apply a first pressure force to the first artery and the second artery; acquire second data collected by the sensor assembly when the first artery and the second artery are in a compression state, the second data including: second pulse wave waveforms collected by each PPG sensor in the PPG sensor group; determine blood flow characteristic data based on the first data and the second data, the blood flow characteristic data including: first waveform characteristics of each first pulse wave waveform and second waveform characteristics of each second pulse wave waveform, and / or, first pulse conduction time difference between first pulse wave waveforms of different phases and second pulse conduction time difference between second pulse wave waveforms of different phases; and determine the user's blood pressure measurement result based on the blood flow characteristic data.

2. The smart ring according to claim 1, characterized in that, The PPG sensor group further includes a fifth PPG sensor and a sixth PPG sensor; the fifth PPG sensor and the sixth PPG sensor are arranged along the inner circle of the smart ring and are opposite to each other, and the first straight line formed by the fifth PPG sensor and the sixth PPG sensor is perpendicular to the second straight line formed by the first PPG sensor and the second PPG sensor.

3. The smart ring according to claim 1, characterized in that, The control components include: stranded wire, winch, and winch motor; One end of the stranded wire is connected to the winch, and is arranged in a clockwise or counterclockwise direction inside the smart ring and wrapped around the soft pouch assembly. The other end of the stranded wire is connected to a fixing member disposed inside the smart ring. The winch motor is connected to the winch and is used to drive the winch to rotate to tighten or loosen the strand. The soft capsule assembly applies pressure to the first artery and the second artery as the strand is tightened, and relaxes the pressure applied to the first artery and the second artery as the strand is loosened.

4. The smart ring according to claim 3, characterized in that, The sensor assembly further includes a pressure sensor as the fixing element; the first data further includes: a first pressure value collected by the pressure sensor when the first artery and the second artery are in a non-compression state; the second data further includes: a second pressure value collected by the pressure sensor when the first artery and the second artery are in a compression state; the blood flow characteristic data further includes: the second pressure value and the first pressure value.

5. The smart ring according to claim 1, characterized in that, The sensor assembly further includes an accelerometer, and the first data further includes: first acceleration information collected by the accelerometer when the first artery and the second artery are in a non-compression state; the second data further includes: second acceleration information collected by the accelerometer when the first artery and the second artery are in a compression state; The central control unit is further configured to, after acquiring the first data and before controlling the soft balloon assembly through the control component, determine, based on the first acceleration information, whether the user is in motion during the information acquisition process of the sensor assembly; if so, determine that the acquired first data is invalid and terminate the current blood pressure measurement; if not, control the soft balloon assembly through the control component to make the soft balloon assembly apply pressure to the first artery and the second artery. The central control unit is further configured to, after acquiring the second data and before inputting the blood flow feature data into the pre-trained blood pressure prediction model, determine, based on the second acceleration information, whether the user is in motion during the information acquisition process of the sensor component; if so, determine that the acquired second data is invalid and terminate the current blood pressure measurement; if not, input the blood flow feature data into the pre-trained blood pressure prediction model to obtain the user's blood pressure measurement result.

6. The smart ring according to claim 1, characterized in that, The central control unit is also configured to control the soft balloon assembly through the control component, so that the soft balloon assembly applies a second pressure force to the user's first artery and second artery, and monitor the peak value of the third pulse wave waveform collected by the first PPG sensor and the second PPG sensor during the application of the second pressure force; and calibrate the magnitude of the pressure force that causes the peak value of the third pulse wave waveform to be lower than a preset threshold as the magnitude of the first pressure force.

7. The smart ring according to claim 1, characterized in that, The first pulse conduction time difference includes: the pulse conduction time difference between the first pulse waveform acquired by the first PPG sensor and the first pulse waveform acquired by the third PPG sensor, and the pulse conduction time difference between the first pulse waveform acquired by the second PPG sensor and the first pulse waveform acquired by the fourth PPG sensor; the second pulse conduction time difference includes: the pulse conduction time difference between the second pulse waveform acquired by the third PPG sensor and the second pulse waveform acquired by the fourth PPG sensor.

8. The smart ring according to claim 1, characterized in that, The first waveform feature includes the rising area of ​​the first pulse wave waveform acquired by each PPG sensor in the PPG sensor group; the second waveform feature includes the rising area of ​​the second pulse wave waveform acquired by each PPG sensor in the PPG sensor group.

9. The smart ring according to claim 8, characterized in that, The first waveform feature also includes one or more of the 1 / LASI value, 1 / LASI_2 value, PIR value, and RIPV value of the first pulse wave waveform collected by each PPG sensor in the PPG sensor group.

10. A method for measuring blood pressure, characterized in that, include: The first data collected from the user is obtained when the first artery on the side of the middle finger closer to the index finger and the second artery on the side of the middle finger closer to the ring finger are in a non-compression state. The first data includes: first pulse waveforms collected at multiple preset collection points on the user's hand; the preset collection points include: the first artery, the second artery, the third artery on the side of the index finger near the middle finger, and the fourth artery on the side of the ring finger near the middle finger. Acquire second data from the user when the first artery and the second artery of the user are under compression; the second data includes: second pulse wave waveforms collected at each of the preset collection sites on the user's hand; Based on the first data and the second data, blood flow characteristic data is determined; the blood flow characteristic data includes: the first waveform characteristics of each first pulse wave waveform and the second waveform characteristics of each second pulse wave waveform, and / or, the first pulse conduction time difference between first pulse wave waveforms of different phases and the second pulse conduction time difference between second pulse wave waveforms of different phases; Based on the blood flow characteristic data, the user's blood pressure measurement result is determined.

11. The method according to claim 10, characterized in that, Determining the user's blood pressure measurement result based on the blood flow characteristic data includes: The blood flow feature data is input into a pre-trained blood pressure prediction model to obtain the blood pressure measurement results; The blood pressure prediction model was trained in the following manner: The sample first data is collected from the sample user when the first artery on the side of the middle finger near the index finger and the second artery on the side of the middle finger near the ring finger are in a non-compression state; the sample first data includes: the sample first pulse wave waveform collected at the multiple preset collection sites on the sample user's hand respectively; Acquire sample second data from the sample user when the first artery and the second artery of the sample user are under compression; the sample second data includes sample second pulse wave waveforms collected at the plurality of preset collection sites on the sample user's hand; Based on the first sample data and the second sample data, the blood flow characteristic data of the sample user is determined; the blood flow characteristic data includes: the first waveform characteristics of each sample's first pulse wave waveform and the second waveform characteristics of each sample's second pulse wave waveform, and / or, the first pulse conduction time difference between sample first pulse wave waveforms of different phases and the second pulse conduction time difference between sample second pulse wave waveforms of different phases; The blood flow feature data of the sample is input into the blood pressure prediction model to be trained to obtain the predicted blood pressure measurement results of the sample users; Based on the predicted blood pressure measurement results of the sample users and the actual blood pressure measurement results of the sample users, the blood pressure prediction model to be trained is trained to obtain the trained blood pressure prediction model.

Images