Anemia blood transfusion risk management system based on real-time monitoring
By monitoring the multi-dimensional physiological parameters of anemia patients in real time, building a dynamic risk index model and performing waveform comparison, the problem of lack of dynamic evaluation in traditional blood transfusion decisions is solved, and precise blood transfusion risk management for anemia patients is achieved.
Patent Information
- Application Number
- CN202510489752.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-04-18
- Publication Date
- 2025-08-01
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Existing blood transfusion decisions rely mostly on static indicators, lack continuous assessment of the dynamic physiological status of anemia patients, difficult to reflect the true tissue oxygen supply level and circulating compensation ability, and lack individualized risk assessment technology that integrates multidimensional physiological parameters.
A risk management system for transfusion based on real-time monitoring is adopted, and data such as hemoglobin concentration, blood oxygen saturation waveform and cardiovascular waveform are obtained through multiple sensors. A dynamic risk index model is constructed using the calculation module, combining the morphological similarity comparison between the real-time waveform and the reference waveform, dynamically correct the risk score, and output the blood transfusion risk level.
It has achieved a comprehensive reflection of the real-time circulatory status and hypoxia risk of anemia patients, improved the accuracy of blood transfusion risk assessment and individualized strategy support, and reduced the system construction and maintenance costs.
Smart Images

Figure CN120413018A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of medical data processing, and more particularly, relates to a risk management system for anemia blood transfusion based on real-time monitoring. Background Art
[0002] Anemia is a common clinical syndrome, often caused by chronic bleeding, malnutrition, neoplastic diseases, or organ dysfunction. During the treatment of patients with moderate to severe anemia, blood transfusion is a commonly used and crucial intervention. However, inappropriate blood transfusion decisions may trigger transfusion-related adverse reactions, immunosuppression, or even increased blood volume load, increasing the potential risks for patients. Traditional blood transfusion judgments mostly rely on static indicators, such as the single measurement result of hemoglobin concentration, lacking continuous assessment of the patient's dynamic physiological state and being difficult to reflect the true tissue oxygen supply level and circulatory compensatory ability. In addition, there is currently a lack of a technical means that can comprehensively integrate multi-dimensional physiological parameters, combine individualized waveform analysis, and dynamically output a risk score. Therefore, there is an urgent need for a risk management system for anemia blood transfusion that has real-time data collection, intelligent comparison, and auxiliary decision-making functions to improve the scientificity and timeliness of clinical blood transfusion decisions.
[0003] Referring to the relevant disclosed technologies, the technical solution with the publication number WO2010048380A1 proposes a method for measuring and evaluating the utilization of blood products, and evaluates the risks of the blood transfusion system by analyzing the blood transfusion environment and storage conditions. The technical solution with the publication number CN116030990B proposes a method and system for generating a perioperative blood transfusion plan based on a cascade model, and evaluates the blood transfusion timing and risks existing during the perioperative period by analyzing a large number of historical cases. The technical solution with the publication number US20140095209A1 proposes a system and method for reducing the risk of hemodialysis reactions, and analyzes the hemolysis risk of the blood recipient through big data.
[0004] The above technical solutions all propose to use various methods or systems to evaluate the risks of blood recipients, but there are currently few technical solutions for the risk management of blood transfusion for anemia patients.
[0005] The foregoing discussion of the background art is only intended to facilitate the understanding of the present invention. This discussion does not recognize or admit that any of the materials mentioned is a part of common general knowledge. Summary of the Invention
[0006] The object of the present invention is to provide a risk management system for anemia blood transfusion based on real-time monitoring, belonging to the technical field of medical data processing. The management system includes a plurality of sensors for obtaining a plurality of monitoring data such as the hemoglobin concentration, blood oxygen saturation waveform, and cardiovascular waveform of a patient, and a calculation module performs real-time analysis. The calculation module constructs a dynamic risk index model to quantify key indicators such as blood volume change and hemoglobin concentration, and at the same time compares the morphological similarity of the real-time waveform with a plurality of established reference waveform templates. The management system corrects the risk level based on the similarity analysis result and outputs a comprehensive score. This score can be used as a reference standard for whether to take blood transfusion intervention clinically.
[0007] The present invention adopts the following technical solutions: A risk management system for anemia blood transfusion based on real-time monitoring, the management system includes:
[0008] One or more sensors for obtaining a plurality of monitoring data from a patient; and
[0009] A calculation module in communication with one or more of the sensors, the calculation module being configured to perform the following steps:
[0010] Obtain the monitoring data of the patient at a plurality of time series;
[0011] Analyze the monitoring data and calculate the risk index;
[0012] Perform a morphological similarity comparison between the actual waveform represented by the current monitoring data and a plurality of reference waveform data included in the pre-established anemia risk model;
[0013] Based on the matching degree between the real-time waveform and the reference waveform, correct or adjust the risk index score;
[0014] Wherein, the risk score is used to indicate the blood transfusion risk level and is used to assist in decision-making or control the treatment response.
[0015] Preferably, the monitoring data includes one or more of the following data: hemoglobin concentration, blood oxygen saturation waveform, and cardiovascular waveform data.
[0016] Preferably, the management system further includes a decision-making module configured to prompt a blood transfusion risk response strategy based on the risk score.
[0017] Preferably, the risk index RI is calculated by the following calculation formula:
[0018]
[0019] In the above formula, Hb(t) is the real-time hemoglobin concentration; Hb st is the blood transfusion trigger threshold of the hemoglobin concentration, Hbcrit is the lowest safety threshold of hemoglobin concentration; ΔV(t) is the change in dynamic blood volume, which is the maximum allowable loss of blood volume; Hb st 、Hb crit and V max are specifically set by relevant technical personnel according to the patient's basic condition.
[0020] Preferably, the waveform similarity analysis of the real-time waveform and the reference waveform includes the following steps:
[0021] S100: Extract the period and perform normalization processing on the actual waveform and the reference waveform;
[0022] S200: Calculate the morphological similarity index between the normalized real-time waveform and the reference waveform to obtain the similarity index;
[0023] S300: According to the similarity index, determine whether the current period matches and eliminate abnormal waveform periods;
[0024] S400: Aggregate the waveforms of all valid periods to update the final risk score.
[0025] Furthermore, a risk management method for anemia transfusion based on real-time monitoring is proposed. The management method is applied to the risk management system for anemia transfusion based on real-time monitoring; the management method includes the following steps:
[0026] E100: Obtain the monitoring data of the patient under multiple time series;
[0027] E200: Analyze the monitoring data and calculate the risk index;
[0028] E300: Compare the morphological similarity between the actual waveform represented by the current monitoring data and multiple reference waveform data included in the pre-established anemia risk model;
[0029] E400: Based on the matching degree between the real-time waveform and the reference waveform, correct or adjust the risk index score.
[0030] The beneficial effects obtained by the present invention are:
[0031] 1. The management system of the present technical solution not only collects hemoglobin concentration, but also combines multiple continuous physiological parameters such as blood oxygen saturation waveform and cardiovascular waveform to construct a dynamic risk index model, which can comprehensively reflect the patient's real-time circulatory state and hypoxia risk, and avoid misjudgment caused by single-index decision-making.
[0032] 2. The management system of this technical solution aligns and matches the morphology of the real-time collected waveform with the reference waveform library through the dynamic time warping algorithm, improving the recognition ability of abnormal physiological states, enabling risk assessment to not only be based on numerical changes but also consider waveform trends and structural features, and enhancing the physiological perception ability of the system.
[0033] 3. The management method of this technical solution can dynamically correct the risk score according to the matching degree between the real-time waveform and the reference model, and output blood transfusion suggestions, effectively improving the judgment accuracy of clinical marginal cases, and providing intelligent support for individualized blood transfusion strategies at the same time.
[0034] 4. The software and hardware parts of the management system of this technical solution adopt a modular design. Each working module, component of the hardware part in the system, as well as instructions, parameters, and algorithms of the software part can be conveniently replaced and / or upgraded later, thereby reducing the construction cost and maintenance cost of this system. BRIEF DESCRIPTION OF THE DRAWINGS
[0035] The present invention can be further understood from the following description in conjunction with the drawings. The components in the drawings are not necessarily drawn to scale, but the emphasis is placed on showing the principles of the embodiments. In different views, the same reference numerals designate corresponding parts.
[0036] Description of the reference numerals in the drawings: 100 - calculation module; 150 - sensor; 200 - patient; 500 - computer system; 502 - bus; 504 - processor; 506 - main memory; 508 - read-only memory; 510 - storage device; 512 - display; 514 - input device; 516 - cursor control device; 518 - network device;
[0037] Figure 1 It is a schematic diagram of the setting of the management system according to the present invention;
[0038] Figure 2 It is a step diagram of performing waveform comparison in an embodiment of the present invention;
[0039] Figure 3 It is a step diagram of performing the management method in an embodiment of the present invention;
[0040] Figure 4 It is a schematic diagram of comparing the actual waveform with the reference waveform in an embodiment of the present invention;
[0041] Figure 5 It is a schematic framework diagram of the computer system adopted by the calculation module in an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS
[0042] In order to make the object, technical solutions and advantages of the present invention more clear and understandable, the present invention will be further described in detail below in conjunction with its embodiments; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention. For those skilled in the art, after referring to the following detailed description, other systems, methods, and / or features of this embodiment will become obvious. It is intended that all such additional systems, methods, features, and advantages be included within this specification, be included within the scope of the present invention, and be protected by the appended claims. Additional features of the disclosed embodiments are described in the following detailed description and will be obvious from such detailed description.
[0043] In the drawings of the embodiments of the present invention, the same or similar reference numerals correspond to the same or similar components; in the description of the present invention, it should be understood that if there are terms such as "upper", "lower", "left", "right", etc. indicating the orientation or positional relationship, they are based on the orientation or positional relationship shown in the drawings. This is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or component referred to must have a specific orientation, be constructed and operated in a specific orientation. Therefore, the terms describing the positional relationship in the drawings are only for illustrative purposes and cannot be understood as a limitation of this patent. For those of ordinary skill in the art, the specific meanings of the above terms can be understood according to specific circumstances.
[0044] Embodiment 1: Exemplarily, a risk management system for anemia blood transfusion based on real-time monitoring is proposed, as shown in the attached Figure 1 figure, the management system includes:
[0045] One or more sensors for obtaining multiple monitoring data from a patient; and
[0046] A computing module communicating with one or more of the sensors, the computing module being configured to perform the following steps:
[0047] Obtain the monitoring data of the patient at multiple time series;
[0048] Analyze the monitoring data and calculate a risk index;
[0049] Compare the actual waveform represented by the current monitoring data with multiple reference waveform data included in a pre-established anemia risk model for morphological similarity;
[0050] Based on the matching degree between the real-time waveform and the reference waveform, correct or adjust the risk index score;
[0051] Wherein, the risk score is used to prompt the blood transfusion risk level and is used to assist in decision-making or control the treatment response.
[0052] Preferably, the monitoring data includes one or more of the following data: hemoglobin concentration, oxygen saturation waveform, and cardiovascular waveform data.
[0053] Preferably, the management system further includes a decision-making module configured to prompt a blood transfusion risk response strategy based on the risk score.
[0054] Preferably, the risk index RI is calculated by the following formula:
[0055]
[0056] In the above formula, Hb(t) is the real-time hemoglobin concentration; Hb st is the blood transfusion trigger threshold of the hemoglobin concentration, and Hb crit is the lowest safety threshold of the hemoglobin concentration; ΔV(t) is the change in dynamic blood volume, and V max is the maximum allowable blood volume loss; Hb st , Hb crit and V max are specifically set by relevant technical personnel according to the patient's basic condition.
[0057] Among them, for ordinary adults, the following set values can be referred to: Hb st = 7, Hb crit = 5;
[0058] For patients with coronary heart disease, it may be Hb st = 9, Hb crit = 7;
[0059] In addition, for the value of V max , specific values need to be set for adults, children, patients with chronic anemia, postoperative patients, etc.
[0060] Preferably, as shown in the appendix Figure 2 , the waveform similarity analysis of the execution real-time waveform and the reference waveform includes the following steps:
[0061] S100: Extract the period and perform normalization processing on the actual waveform and the reference waveform;
[0062] S200: Calculate the morphological similarity index of the normalized real-time waveform and the reference waveform to obtain the similarity index;
[0063] S300: Determine whether the current period matches according to the similarity index, and eliminate abnormal waveform periods;
[0064] S400: Aggregate the waveforms of all valid periods for updating the final risk score.
[0065] Further, a risk management method for anemia blood transfusion based on real-time monitoring is proposed, and the management method is applied to the risk management system for anemia blood transfusion based on real-time monitoring; as shown in the appendix Figure 3 As shown, the management method includes the following steps:
[0066] E100: Obtain the monitoring data of the patient at multiple time series;
[0067] E200: Analyze the monitoring data and calculate the risk index;
[0068] E300: Compare the actual waveform represented by the current monitoring data with multiple reference waveform data included in the pre-established anemia risk model in terms of morphological similarity;
[0069] E400: Based on the matching degree between the real-time waveform and the reference waveform, correct or adjust the risk index score.
[0070] Specifically, as shown in the appendix Figure 1 As shown, an exemplary implementation mode shows the overall setting of the management system. The management system 100 includes a computing module 100, and the computing module 100 can be a computer system with computing capabilities. The computing module 100 communicates with one or more sensors 105 (schematically marked as 105a, 105b, 105c in the figure), and these sensors are configured to obtain monitoring data from the monitored patient 200. In a specific implementation mode, the computing module 100 can read the values from the physiological sensors 105, perform data processing and data analysis based on these data according to the technical solution of the present invention, and use the analysis results for individualized estimation, prediction or other diagnoses, and display the analysis results, and propose and / or implement corresponding treatment suggestions.
[0071] Exemplarily, the sensor 105 can be various types of electronic sensors, but it should be clear that the types of sensors mentioned in this article are only for example and not for technical limitation; multiple sensors are used to obtain physiological data from the subject. An exemplary sensor combination can include a sensor for non-invasive continuous blood pressure waveform acquisition, a pulse oximeter sensor, and an analog-to-digital converter for connecting the sensor to the computing module 100. More generally, in one implementation mode, one or more sensors 105 can perform the continuous acquisition operations required by the technical solution of the present invention, so as to continuously obtain physiological waveform data from the patient 200, such as continuous blood pressure signals, and the data obtained by the sensor 105 can be sent to the computing module 100 through a communication line.
[0072] In an exemplary embodiment, the management system may further include a treatment device 115, which can be controlled by the computing module 100 to implement a treatment plan according to the suggestions obtained from the analysis of the patient's physiological data. In a specific embodiment, the treatment device may include a hemodialysis device, and the computing module 100 can control the treatment device 115 to operate based on the risk score evaluated for the patient. Exemplarily, the treatment device 11 may include a cardiac assist device, a ventilator, an automatic implantable cardioverter defibrillator (AICD), a pacemaker, an extracorporeal membrane oxygenation (ECMO) system, a positive pressure ventilation device, an anesthetic machine, an integrated intensive care system, a medical robot, a venous or arterial pump for infusing and / or delivering therapeutic drugs (such as providing drugs through intravenous injection), a heating / cooling blanket, etc.
[0073] Further, in a preferred embodiment, the hemoglobin concentration Hb(t) can be continuously measured and recorded in value through various real-time measurement methods;
[0074] Preferably, it includes using a sensor based on multi-wavelength spectroscopy. Based on the difference in the absorption characteristics of hemoglobin for light of different wavelengths and the different absorption peaks of reduced hemoglobin at 660 nm and 940 nm, the transmitted / reflected light signals are collected through a photoelectric sensor, such as PPG or NIRS, and the hemoglobin concentration is deduced using the following calculation formula:
[0075]
[0076] In the above formula, α i is the absorption coefficient at wavelength λ i , I is the light intensity, d is the optical path, and ∈ Hb is the extinction coefficient of hemoglobin. Preferably, 8 LED light sources with wavelengths of 500 - 1000 nm can be used, combined with an adaptive noise cancellation algorithm, to continuously monitor Hb; or a ring PPG sensor is used to compress the finger to temporarily block venous blood flow, enhancing the signal-to-noise ratio of the arterial signal, and optionally, the measurement time interval is about 30 seconds to 60 seconds.
[0077] In a preferred embodiment, it includes using a minimally invasive intravascular fiber optic sensor or a microneedle array-based transcutaneous hemoglobin monitoring to perform real-time monitoring of the hemoglobin concentration; this technical solution does not limit the specific monitoring method adopted.
[0078] Further, the blood transfusion trigger threshold Hb st is the dynamic hemoglobin concentration critical value used in clinical practice to determine whether a patient needs a blood transfusion. Its setting is based on the patient's pathophysiological characteristics, organ function status, and real-time metabolic requirements, aiming to balance tissue oxygen supply and blood transfusion risk. For different patient groups, Hb stThere are significant differences in the settings. Preferably, for example, anemic patients with coronary heart disease have limited myocardial oxygen supply due to coronary artery stenosis and need to maintain a relatively high hemoglobin level to ensure sufficient oxygenation of myocardial cells. Therefore, Hb st can be set to 9 g / dL; while healthy adults, due to their strong cardiovascular compensatory ability, their Hb st can be safely set to 7 g / dL.
[0079] Additionally preferably, Hb st is not a fixed value and can be dynamically adjusted according to real-time monitoring parameters. For example, when the patient's cardiac function declines, such as when the left ventricular ejection fraction LVEF < 40%, for every 10% decrease, Hb st is increased by 0.5 g / dL; or in a hypermetabolic state, such as during a fever, for every 1°C increase in body temperature, Hb is increased by 0.3 g / dL. The management system can automatically correct the threshold Hb st to match the current physiological state. Meanwhile, the dynamic setting can be achieved by integrating the patient's electronic medical record data (such as the diagnosis of underlying diseases) with dynamic physiological signals, such as changes in cardiac output analyzed from arterial blood pressure waveforms, so as to set the final value of Hb st .
[0080] In a preferred embodiment, photoplethysmography (PPG) is used to obtain the patient's blood oxygen saturation waveform and cardiovascular waveform in real time. By emitting red light (660 nm) and infrared light (940 nm) from the detection device to the epidermis at the pulse position of the human body, the skin tissue can be penetrated to measure the absorption difference of hemoglobin and oxyhemoglobin to light. The absorption of light by hemoglobin in the blood changes periodically with the change in vascular volume caused by cardiac pulsation, thereby generating a volume pulse waveform. The blood oxygen saturation waveform can reflect the periodic change in blood volume caused by peripheral vascular pulsation, and its fluctuation frequency is synchronized with the heart rate.
[0081] Meanwhile, a detection device with the same principle can be used to collect and analyze the cardiovascular waveform of the patient's artery. The analysis content includes:
[0082] Pulse wave morphology analysis. By analyzing the geometric characteristics of the PPG waveform, the cardiovascular function state can be quantified. The detectable parameters include:
[0083] (1) The main wave rise time, which refers to the time required for the pulse wave to rise from the baseline to the peak of the main wave. The normal range is 80 - 12 ms; this parameter directly reflects the efficiency of cardiac output and arterial elasticity. When myocardial contractility increases or arterial vascular elasticity is good, the peripheral blood vessels are rapidly filled with blood and the rise time is shortened; conversely, if arteriosclerosis or cardiac function declines, the main wave rise time will be significantly prolonged, for example, greater than 150 ms.
[0084] (2) Dicrotic wave amplitude ratio analysis, that is, calculating the ratio of the amplitude of the dicrotic wave trough to the amplitude of the main wave peak in the PPG waveform, with a normal range of 0.2 - 0.4. This ratio is negatively correlated with peripheral vascular resistance. When peripheral resistance increases, such as in the early compensatory stage of shock, the blood return resistance increases, the dicrotic wave amplitude is suppressed, resulting in a decrease in the ratio; while when the blood vessels dilate, the ratio increases, making the measured value greater than 0.5.
[0085] Furthermore, it also includes measuring the pulse wave conduction time. By deploying dual-channel PPG sensors at different body surface locations, such as the radial artery and fingertips, the time difference (PTT) of the pulse wave propagation can be measured. Based on the positive correlation between the pulse wave velocity and arterial wall stiffness, clinical studies have shown that for every 10-millisecond shortening of the PTT, the systolic blood pressure increases by an average of 8 - 12 mmHg.
[0086] The combined analysis of the above parameters can evaluate cardiovascular function. For example, an extended main wave rise time and a decreased combined dicrotic wave amplitude ratio indicate heart insufficiency combined with peripheral vasoconstriction; while the real-time dynamic monitoring of the PTT provides a continuous quantitative tool for evaluating the drug efficacy in hypertensive patients.
[0087] When processing the recorded PPG waveform, first extract the waveform features of the PPG waveform. Preferably, first preprocess the original signal x raw (t). Preferably, a second-order Butterworth low-pass filter with a cut-off frequency f c [[ID=1 \ 3]]= 8 Hz and an order n = 4 is used to remove high-frequency noise and retain the low-frequency components related to cardiac pulsation. The filtered signal x filter (t) can be expressed as:
[0088] x filter (t)= Butterworth(x raw (t), fc, n);
[0089] In the above formula, Butterworth() is the Butterworth low-pass filtering function.
[0090] Subsequently, identify the main wave peak point on the smoothed waveform, that is, the highest amplitude point in each cardiac cycle. Define the main wave peak time point t peak and amplitude A peak as:
[0091]
[0092] A perk = x filter (t perk );
[0093] Among them, T is the preset heartbeat cycle window, and the optional values are, for example, 1 s or 1.2 s. At the same time, the starting time t0 of the time count is updated by the sliding window method.
[0094] Next, in a similar way, the dicrotic notch is located in the descending segment of the waveform. This characteristic point corresponds to the reflected wave of the peripheral blood vessels. The corresponding value can be obtained by retrieving the first local minimum point after the main peak. The time point and amplitude of the dicrotic notch are denoted as t valley and A valley .
[0095] Then, characteristic parameters are calculated, including:
[0096] The time difference Δt, which reflects the pulse wave conduction velocity: Δt = t valley -t peak ;
[0097] The amplitude ratio ΔA, which evaluates the peripheral vascular resistance: ΔA = A valley / A peak .
[0098] Furthermore, to improve the robustness, an amplitude threshold mechanism is introduced. Only when the amplitude of the main peak exceeds 1.5 times the baseline amplitude A baseline is it determined as an effective characteristic point. The baseline amplitude A baseline is:
[0099]
[0100] where N is the number of main peaks in the most recent monitoring period (for example, 10 s or 20 s).
[0101] After calculating the above parameters, the waveform comparison process is started, which specifically includes the following steps:
[0102] S100: Extract the period and perform standardization processing on the actual waveform and the reference waveform;
[0103] Specifically, it includes the following sub-steps:
[0104] S110, Extract each complete heartbeat cycle from the real-time collected cardiovascular waveform (such as PPG). It can be completed by detecting the main peak and expanding a fixed time window before and after, for example, from 200 milliseconds before the main peak to 500 milliseconds after the main peak;
[0105] S120, Perform standardization processing on the time axis of each cycle, and interpolate and resample it to the same number of sampling points, such as 500 or 1000 points, to eliminate the cycle length difference caused by the heart rate change;
[0106] S130, perform amplitude normalization, that is, normalize the amplitude of each period to the interval [0, 1]; the specific method is: take the baseline amplitude A baseline as equivalent to 0, and the amplitude A peak of the main wave peak in the corresponding period as equivalent to 1, and the remaining points are scaled proportionally; this step can eliminate the difference in individual signal intensity and highlight the relative shape of the waveform.
[0107] S200: Calculate the morphological similarity between the actual waveform and the reference waveform. Preferably, each of the reference waveforms has been subjected to normalization processing. Preferably, use dynamic time warping (DTW) to calculate the degree of alignment of the time axes of the actual waveform and the reference waveform, and calculate the correlation coefficient between the two waveforms to evaluate their morphological trend similarity. Specifically, it may include one or more of the following comparison steps:
[0108] Preferably, perform dynamic time warping on the waveforms, and by elastically stretching or compressing one or both of the two waveforms, find the most matching feature points of the two, so as to measure the similarity of the overall shape. More simply, first compare the differences between the two waveforms point by point (such as the absolute value of the amplitude difference); then, from the start point to the end point of the waveform, automatically find an alignment path so that the cumulative difference of all points on the path is the smallest. This path determines the elastic matching method of the two waveforms, such as aligning the wave peaks or aligning the wave valleys; further, divide the cumulative difference by the path length to obtain the final similarity index, that is, the dynamic difference value. The smaller the dynamic difference value, the higher the similarity between the two waveforms. Through the analysis of the dynamic difference value of dynamic time warping, it can be judged whether the wave shapes of different periods are consistent, and abnormal waveforms caused by motion artifacts or noise are excluded. For example, if the DTW distance between a certain period and the template waveform exceeds the threshold, it is considered that this period is unreliable and does not participate in subsequent parameter calculations.
[0109] Finally, as shown in the appendix Figure 4 After comparing the actual waveform and the reference waveform based on the dynamic time warping method, the difference value between the two waveforms is statistically calculated.
[0110] Preferably, the linear correlation of the waveforms after normalization can also be evaluated using the correlation index. Specifically, set the waveform sequences of the two waveforms after normalization as: X = [x1, x2, …, x N and Y = [y1, y2, …, y N , then the correlation index γ is:
[0111]
[0112] In the above formula, μ(x) and μ(y) respectively represent the means of the waveform sequences X and Y.
[0113] S300: Perform correlation determination and abnormal screening on the actual waveform and the reference waveform;
[0114] In similarity determination, it is necessary to comprehensively determine the degree of correlation between two waveforms. According to historical data or experimental calibration, the upper limit of the distance of the dynamic difference value can be set to 0.2 or less, and the lower limit of the correlation coefficient γ can be set to 0.8 or greater. The waveform period pairs exceeding the threshold are regarded as having inconsistent morphologies to calculate the similarity degree of the two waveforms being compared.
[0115] Exemplarily, if the comparison result of a real-time period with any reference template simultaneously meets the above thresholds, it is determined as a "matching period", and its best-matching template is recorded; if all comparison results of a real-time period do not meet the threshold requirements, then this period is determined as an "abnormal period".
[0116] S400: Extract features and screen periods from the real-time waveform. Among them, for all waveforms determined as "matching periods", key morphological parameters are extracted, including the main wave rise time, dicrotic wave ratio, time difference (Δt) and amplitude difference (ΔA) between the main wave and the dicrotic wave, etc.; meanwhile, the above parameters of multiple effective periods are statistically analyzed to calculate their mean and standard deviation as the feature output of the current real-time period. Preferably, a volatility constraint can also be set. For example, if the standard deviation of a certain characteristic value of the real-time waveform exceeds 20% of the mean within a preset time period, it is prompted that the period fluctuation is too large and there may be undetected abnormalities.
[0117] Finally, determine the reference waveform most similar to the current real-time waveform and output multiple characteristic parameters of the real-time waveform.
[0118] Furthermore, each waveform in each group of multiple groups of reference waveforms included in the anemia risk model corresponds to a typical physiological feature of a historical clinical scenario, such as compensatory shock, chronic anemia, etc.; and preferably, each group of reference waveforms has already corresponded to a reference risk index RI ref . After calculating the real-time risk index RI based on the real-time monitoring data, set the real-time risk index as RI Pre , and compare RI Pre with the reference risk index RI ref to verify the credibility of the real-time risk index RI Pre and identify special risk patterns. For example, if the real-time RI Pre = 0.4, but the similarity between the real-time waveform and the reference waveform of "occult shock" reaches 80%, and the corresponding reference RI ref = 0.8, it is prompted that the actual risk may be higher than the RI calculated value and urgent intervention is required.
[0119] Embodiment 2: This embodiment should be understood as at least including all the features of any one of the foregoing embodiments and being further improved on this basis.
[0120] In a preferred embodiment, the transfusion risk of the monitored anemic patients is finally reflected by a comprehensive evaluation index RI. total The comprehensive evaluation index RI total is calculated by an expression including dynamic weights, specifically:
[0121] RI total = α · RI pre + β · RI ref ;
[0122] where α and β are the dynamic weights corresponding to the real-time risk index RI Pre and the reference risk index RI ref respectively, and α + β = 1.
[0123] Furthermore, the specific values of α and β are adjusted dynamically to achieve an intelligent trade-off between the reliability of real-time data and the specificity of pathological features. The dynamic adjustment of the two weights takes into account physiological bases.
[0124] Among them, the weight α reflects the credibility of the real-time risk index RI Pre calculated from the hemoglobin-volume parameter, and the conditions for its increase include considerations of:
[0125] The reliability of the dynamic blood volume change ΔV(t): When the blood volume loss is verified by gold standards such as ultrasonic detection and invasive hemodynamic monitoring, the measurement error of ΔV(t) ≤ 10%, and at this time, a higher weight should be given to α. For example, the initial value of α can be 0.7 - 0.8.
[0126] The stability of the real-time hemoglobin concentration Hb(t): If the fluctuation of Hb(t) is less than 0.5 g / dL within 5 consecutive monitoring periods, it indicates that the optical signal quality is stable, and the weight α can be increased by 0.05 - 0.1 subsequently.
[0127] On the other hand, the weight β can affect the comprehensive evaluation index RI total based on pathological sensitivity. β reflects the pathological specificity of the waveform matching result. Preferably, for example, the weight β can be increased in the following situations:
[0128] Highly consistent characteristic waveforms: When waveforms with clear pathological significance such as paradoxical pulse waveforms and pulsus alternans are matched, that is, the similarity between the actual waveform and the reference waveform is greater than 90%, β is set to 0.7 - 0.8.
[0129] Microcirculation disorder markers: The disappearance of the d wave (ΔA < 0.1) and the DTW distance < 0.1 in the second derivative of PPG indicate capillary leakage, and β is up-regulated to 0.7.
[0130] In a preferred embodiment, several rules can be set by relevant technical personnel to clarify the increase and decrease rules of α and β; for example, refer to the exemplary adjustment rules shown in the following table:
[0131]
[0132] In a preferred embodiment, the initial values and adjustment rules of the weights α and β are derived from multi-center clinical data-driven modeling. Preferably, the modeling process includes:
[0133] Incorporate a large amount of patient data, covering scenarios such as trauma, infection, and cardiogenic shock, and label the real clinical treatment methods and corresponding patient condition feedback.
[0134] Establish a feature matrix, which includes calculated risk index RI pre values, waveform matching degree, source of ΔV(t) (invasive / non-invasive), pathological type and other multi-dimensional parameters.
[0135] Through machine learning modeling, for example, using random forests to rank the feature importance, and discover the correlation degree between the weights α and β and various features; for example: when ΔV(t) is invasive verification, the contribution degree of the ARI value to the blood transfusion decision is increased to 75%; when specific waveform matching occurs, the contribution degree of the waveform feature is as high as 82%.
[0136] Quantify the marginal effect of each feature on the prediction result through SHAP value analysis, and the initial ratio of α / β can be deduced, such as 0.6:0.4.
[0137] Example 3: This example should be understood as including at least all the features of any one of the foregoing examples, and further improved on this basis.
[0138] Exemplarily, as shown in the appendix Figure 5 illustrate the implementation manner of the computer system 500 adopted by the computing module 100; the computer system 500 can be applied to the data storage, operation and result output processes of each working module in the recognition and judgment system.
[0139] Exemplarily, the computer system 500 includes a bus 502 or other communication mechanisms for transmitting information, and one or more processors 504 coupled to the bus 502 for processing information; the processor 504 can be, for example, one or more general-purpose microprocessors.
[0140] The computer system 500 also includes a main memory 506, such as a random access memory (RAM), a cache, and / or other dynamic storage devices, which are coupled to the bus 502 for storing information and instructions to be executed by the processor 504; the main memory 506 can also be used to store temporary variables or other intermediate information during the execution of instructions executed by the processor 504; when these instructions are stored in a storage medium accessible by the processor 504, the computer system 500 is presented as a special-purpose machine customized to execute the operations specified in the instructions.
[0141] The computer system 500 may also include a read-only memory (ROM) 508 or other static storage devices coupled to the bus 502 for storing static information and instructions of the processor 504; a storage device 510 such as a magnetic disk, an optical disk, or a USB drive (flash drive) etc. will be coupled to the bus 502 for storing information and instructions.
[0142] Furthermore, coupled to the bus 502 may also include a display 512 for displaying various information, data, media, etc., and an input device 514 for allowing a user of the computer system 500 to control, manipulate, and / or interact with the computer system 500.
[0143] A preferred way to interact with the management system may be through a cursor control device 516, such as a computer mouse or a similar control / navigation mechanism.
[0144] Furthermore, the computer system 500 may also include a network device 518 coupled to the bus 502; the network device 518 may include components such as a wired network card, a wireless network card, a switching chip, a router, a switch, etc.
[0145] Generally speaking, terms such as "engine", "component", "system", "database", etc. as used herein may refer to logic embodied in hardware or firmware, or refer to a set of software instructions, which may have entries and exit points and are written in a programming language such as Java, C, or C++; software components can be compiled and linked into an executable program, installed in a dynamic link library, or can be written in an interpreted programming language (such as BASIC, Perl, or Python); it should be understood that software components can call other components or themselves, and / or can be called in response to detected events or interrupts.
[0146] Software components configured to execute on a computing device may be provided on a computer-readable medium, such as a compact disc, digital video disc, flash drive, magnetic disk, or any other tangible medium, or as a digital download (and may initially be stored) in a compressed or installable format that requires installation, decompression, or decryption before execution); such software code may be stored, in whole or in part, on the memory device of the executing computing device for execution by the computing device; the software instructions may be embedded in firmware, such as an EPROM; it should also be understood that hardware components may consist of connected logic units (such as gates and flip-flops), and / or may consist of programmable units (such as programmable gate arrays or processors).
[0147] Computer system 500 may include custom hardwired logic, one or more ASICs or FPGAs, firmware, and / or program logic to implement the techniques described herein, the program logic being combined with the computer system to cause computer system 500 to be a specialized computing device.
[0148] In accordance with one or more embodiments, the techniques herein are performed by computer system 500 in response to one or more sequences of one or more instructions contained in main memory 506 being executed by processor 504; such instructions may be read into main memory 506 from another storage medium, such as storage device 510; execution of the instruction sequences contained in main memory 506 causes processor 504 to perform the processing steps described herein; in alternative embodiments, hardwired circuitry may be used in place of or in combination with software instructions.
[0149] As used herein, the term "non-transitory medium" and like terms refer to any medium that stores data and / or instructions that cause a machine to operate in a particular manner; such non-transitory media may include non-volatile media and / or volatile media; non-volatile media includes, for example, optical or magnetic disks, such as storage device 510; volatile media includes dynamic memory, such as main memory 506.
[0150] Common forms of non-transitory media include, for example, floppy disks, hard disks, solid state drives, magnetic tape, or any other magnetic data storage medium, CD-ROM, any other optical data storage medium, any physical medium with a hole pattern, RAM, PROM, and EPROM, FLASH-EPROM, NVRAM, any other memory chip or cartridge, and network versions thereof.
[0151] Non-transient media is different from transmission media but may be used in combination with transmission media; transmission media participates in the transfer of information between non-transient media; for example, transmission media includes coaxial cables, copper wire, and fiber optics, including the wires that make up bus 502; transmission media may also take the form of acoustic or light waves, such as radio waves and infrared data communications.
[0152] Although the present application has been described above with reference to various embodiments, it should be understood that many changes and modifications can be made without departing from the scope of the present application. That is, the methods, systems, and devices discussed above are examples. Various configurations may appropriately omit, replace, or add various processes or components. For example, in an alternative configuration, the method may be performed in a different order than that described, and / or various components may be added, omitted, and / or combined. Moreover, the features described with respect to certain configurations may be combined in various other configurations, such as different aspects and elements of the configuration may be combined in a similar manner. In addition, as technology develops, the elements therein may be updated, that is, many elements are examples and do not limit the scope of the present disclosure or the claims.
[0153] Specific details are given in the description to provide a thorough understanding of the exemplary configurations including the implementation. However, the configurations may be practiced without these specific details. For example, well-known circuits, processes, algorithms, structures, and technologies have been shown without unnecessary details to avoid obscuring the configurations. The description only provides example configurations and does not limit the scope, applicability, or configuration of the claims. Instead, the foregoing description of the configurations will provide those skilled in the art with an enabling description for implementing the described technology. Various changes can be made to the functions and arrangements of the elements without departing from the spirit or scope of the present disclosure.
[0154] In summary, it is intended that the above detailed description be considered illustrative rather than restrictive, and it should be understood that the above embodiments should be construed as only for illustrating the present invention and not for limiting the protection scope of the present invention. After reading the content recorded in the present invention, those skilled in the art can make various changes or modifications to the present invention, and these equivalent changes and modifications also fall within the scope defined by the claims of the present invention.
Claims
1. A real-time monitoring-based anemia blood transfusion risk management system, characterized in that, The management system includes: One or more sensors for obtaining multiple monitoring data from a patient; and A computing module communicating with one or more of the sensors, the computing module being configured to perform the following steps: Obtain the monitoring data of the patient under multiple time series; Analyze the monitoring data and calculate a risk index; Compare the actual waveform characterized by the current monitoring data with multiple reference waveform data included in a pre-established anemia risk model in terms of morphological similarity; Based on the matching degree between the real-time waveform and the reference waveform, correct or adjust the risk index score; Wherein, the risk score is used to prompt the blood transfusion risk level and is used to assist in decision-making or control the treatment response.
2. The management system according to claim 1, wherein The monitoring data includes one or more of the following data: hemoglobin concentration, blood oxygen saturation waveform, and cardiovascular waveform data.
3. The management system according to claim 1, characterized in that, The management system further includes a decision-making module configured to prompt a blood transfusion risk coping strategy based on the risk score.
4. The management system according to claim 1, wherein The risk index RI is calculated by the following calculation formula: In the above formula, Hb(t) is the real-time hemoglobin concentration; Hb st is the blood transfusion trigger threshold of hemoglobin concentration, and Hb crit is the lowest safety threshold of hemoglobin concentration; ΔV(t) is the dynamic blood volume change, and is the maximum allowable blood volume loss; Hb st 、Hb crit and V max are specifically set by relevant technical personnel according to the patient's basic condition.
5. The management system according to claim 1, wherein Performing the waveform similarity analysis of the real-time waveform and the reference waveform includes the following steps: S100: Perform period extraction and normalization processing on the actual waveform and the reference waveform; S200: Calculate the morphological similarity index between the normalized real-time waveform and the reference waveform to obtain a similarity index; S300: According to the similarity index, determine whether the current period matches and eliminate abnormal waveform periods; S400: Aggregate the waveforms of all valid periods for updating the final risk score.
6. A risk management method for anemia blood transfusion based on real-time monitoring, characterized in that, The management method is applied to an anemia blood transfusion risk management system based on real-time monitoring as described in any one of claims 1 to 5; the management method includes the following steps: E100: Obtain the monitoring data of the patient under multiple time series; E200: Analyze the monitoring data and calculate a risk index; E300: Compare the actual waveform characterized by the current monitoring data with multiple reference waveform data included in a pre-established anemia risk model in terms of morphological similarity; E400: Based on the matching degree between the real-time waveform and the reference waveform, correct or adjust the risk index score.
Citation Information
Patent Citations
A method and system for generating perioperative transfusion protocols based on a cascade model.
CN116030990B
System and method for mitigating the risk of hemolytic transfusion reactions
US20140095209A1
Method for optimizing blood utilization
WO2010048380A1
Cited By
Real-time monitoring method for physiological indexes of patient in blood transfusion process
CN121171656A
