Control method and system of jaundice treatment device
By dynamically adjusting the phototherapy parameters of the jaundice treatment device using a predictive model and an MPC optimizer, the problem of inaccurate dosage caused by manual adjustment is solved, thus achieving safe and efficient phototherapy.
Patent Information
- Application Number
- CN202510967530.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-14
- Publication Date
- 2025-11-18
AI Technical Summary
In existing technologies, the adjustment of phototherapy parameters in jaundice treatment devices relies on manual intervention, which can easily lead to overdosing or underdosing, resulting in misoperation.
By acquiring user evaluation data, predictive models and MPC optimizers are used to dynamically adjust phototherapy parameters, including light source wavelength, intensity, and irradiation time, and optimization goals and constraints are set to ensure that the parameters are within a safe range.
It reduces human intervention, avoids overtreatment or undertreatment, improves the safety and efficiency of treatment, and ensures the accuracy and safety of phototherapy parameters.
Smart Images
Figure CN120960651A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of medical devices, and in particular to a control method and system of a jaundice treatment device. BACKGROUND
[0002] Newborn jaundice is caused by abnormal metabolism of bilirubin in the body, resulting in elevated bilirubin concentration in the blood, leading to yellowing of the skin, mucous membranes and sclera. About 80% of full-term infants and most premature infants will have temporary elevated bilirubin concentration, most of which are physiological jaundice, and a small part is pathological jaundice. Severe pathological jaundice can cause hyperbilirubinemic encephalopathy, causing damage to the brain of the newborn (such as hyperkinesia, hearing loss, and dull eyes), and even endangering the life of the newborn.
[0003] In the prior art, when treating newborns with jaundice, medical staff usually adjusts the phototherapy parameters output by the jaundice treatment device according to the current bilirubin concentration. However, the above method relies on manual adjustment, and manual errors or lack of experience can easily result in overdosing or underdosing.
[0004] Therefore, how to reduce manual intervention during treatment to avoid overdosing or underdosing caused by manual errors or lack of experience is a technical problem that needs to be solved at present. SUMMARY
[0005] To solve the technical problem that manual errors or lack of experience can easily result in overdosing or underdosing, the present application provides solutions in the following aspects.
[0006] In a first aspect, the present application provides a control method of a jaundice treatment device, comprising: obtaining evaluation data of a user, the evaluation data comprising spectral reflectance; inputting the evaluation data into a preset prediction model to obtain a bilirubin concentration risk score and a bilirubin concentration prediction value at a next prediction time; in response to the bilirubin concentration risk score being greater than a risk threshold, determining a candidate interval corresponding to a phototherapy parameter based on the bilirubin concentration risk score, the phototherapy parameter comprising a light source wavelength, the lower limit of the candidate interval corresponding to the light source wavelength being a first preset value, and the upper limit being negatively correlated with the bilirubin concentration risk score; determining and outputting an optimal phototherapy parameter from the candidate interval based on an MPC optimizer, so that the jaundice treatment device outputs a corresponding treatment light source.
[0007] Further, the calculation expression of the upper limit of the candidate interval corresponding to the light source wavelength is:
[0008] λ=470-10(risk_score1-0.5);
[0009] In the formula, λ is the upper limit of the candidate interval corresponding to the wavelength of the light source, and risk_score1 is the risk score of bilirubin concentration.
[0010] Furthermore, the phototherapy parameters include light source intensity, and the lower limit of the candidate interval corresponding to the light source intensity is a second preset value, while the upper limit is positively correlated with the bilirubin concentration risk score.
[0011] Furthermore, the light source intensity includes the light source intensity of the limbs and the light source intensity of the torso, and the phototherapy parameters also include the irradiation time.
[0012] Further, determining and outputting optimal phototherapy parameters from the candidate interval based on the MPC optimizer includes: setting the optimization objective and constraints of the MPC optimizer; the optimization objective includes: minimizing the difference between the predicted bilirubin concentration and the target bilirubin concentration; the constraints include: the optimal phototherapy parameters belong to the corresponding candidate interval; and inputting the predicted bilirubin concentration into the MPC optimizer to obtain the optimal phototherapy parameters.
[0013] Furthermore, the optimization objective also includes minimizing the difference between skin temperature and a temperature threshold, and the constraint also includes ensuring that the increase in light source intensity per hour is less than a change threshold.
[0014] Furthermore, when the optimization objective is to minimize the difference between the predicted bilirubin concentration and the target bilirubin concentration, and to minimize the difference between skin temperature and the temperature threshold, the optimization function corresponding to the optimization objective is:
[0015] min[α·(B pred -B target ) 2 +β·(T skin -T0) 2 ];
[0016] In the formula, min() is the minimum value function, B pred B is the predicted value of bilirubin concentration. target T represents the target value for bilirubin concentration. skin T0 is the skin temperature, α is the weighting coefficient of the difference between the predicted bilirubin concentration and the target bilirubin concentration, and β is the weighting coefficient of the difference between the skin temperature and the temperature threshold.
[0017] Furthermore, the method further includes: determining a target value for bilirubin concentration based on the bilirubin concentration risk score, wherein the target value for bilirubin concentration is positively correlated with the bilirubin concentration risk score.
[0018] Further, the spectral reflectance includes a first spectral reflectance, a second spectral reflectance and a third spectral reflectance, and the evaluation data further includes a heart rate, a blood oxygen saturation, a portal vein flow rate and an ambient light intensity; inputting the evaluation data into a preset prediction model includes: calculating a hemoglobin correction factor according to the blood oxygen saturation, and determining skin maturity according to a corresponding gestational age of the user, the hemoglobin correction factor is positively correlated with the blood oxygen saturation, and the skin maturity is positively correlated with the gestational age; normalizing the hemoglobin correction factor, the skin maturity, the spectral reflectance, the heart rate, the blood oxygen saturation, the portal vein flow rate and the ambient light intensity to generate a set of nine-dimensional state vectors; and inputting the nine-dimensional state vectors into the preset prediction model.
[0019] In a second aspect, the present application provides a control system of a jaundice treatment device, comprising a processor and a memory, the memory storing computer program instructions which, when executed by the processor, implement the control method of the jaundice treatment device according to the first aspect.
[0020] The present application has the beneficial effects that: by selecting the optimal phototherapy parameters from the candidate interval determined according to the bilirubin concentration risk, the present application can dynamically adjust the phototherapy parameters according to the current situation, while avoiding the situation of over-treatment, thereby ensuring the safety and efficiency of jaundice treatment; by setting the optimization target to include minimizing the difference between the skin temperature and the temperature threshold, the present application can avoid the situation of scalding during treatment, thereby ensuring the safety of treatment; by inputting the nine-dimensional state vectors composed of nine-dimensional data, the present application can improve the accuracy of the prediction model in predicting the bilirubin concentration, thereby improving the accuracy of setting the phototherapy parameters. BRIEF DESCRIPTION OF DRAWINGS
[0021] Figure 1 is a flow chart schematically showing a control method of a jaundice treatment device according to an embodiment of the present application;
[0022] Figure 2 is a structural block diagram schematically showing a control system of a jaundice treatment device according to an embodiment of the present application. DETAILED DESCRIPTION
[0023] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.
[0024] The specific embodiments of the present application will be described in detail below with reference to the drawings.
[0025] Figure 1 is a flow chart schematically showing a control method of a jaundice treatment device according to an embodiment of the present application.
[0026] In a first aspect, the present application provides a control method of a jaundice treatment device. Wherein the jaundice treatment device capable of implementing the method of the present application is at least provided with a multi-spectral sensor array and a matrix adjustable wavelength LED light source (for example, the adjustable wavelength range is 425-475nm). In optional other embodiments, other sensors (for example, heart rate sensors, blood oxygen probes, etc.) and micro-flow heat dissipation systems (for example, micro-flow heat dissipation systems composed of graphene heat conduction films and micro eddy current fans) can also be installed. As shown in the figure, the method of the present application comprises the following steps. Figure 1
[0027] S101, obtaining evaluation data of a user.
[0028] In this embodiment, the evaluation data at least includes spectral reflectance, and can also include heart rate, blood oxygen saturation, portal vein flow rate, ambient light intensity and gestational age (i.e. the time of the newborn developing in the mother's body in weeks).
[0029] In this embodiment, the bilirubin concentration is collected by non-invasive method, that is, the light of a specific wavelength is shot into the skin, and then the light of this wavelength is absorbed by bilirubin in the human body, so that the outflow light signal detected outside the body decreases; the higher the bilirubin concentration, the smaller the signal detected, so that the concentration of bilirubin in the body can be quantitatively analyzed.
[0030] In this embodiment, the multi-spectral sensor array is used to collect the spectral reflectance of the newborn to be treated. Specifically, a three-wavelength micro-spectral sensor (capable of emitting light sources of 460nm, 500nm and 550nm, collected every five seconds) can be used, and the corresponding spectral reflectance includes first spectral reflectance (corresponding to 460nm), second spectral reflectance (corresponding to 500nm) and third spectral reflectance (corresponding to 550nm).
[0031] The heart rate is collected by a heart rate sensor; the blood oxygen saturation is collected by a blood oxygen probe; the portal vein flow rate is collected by a micro-doppler ultrasound module (integrating an ultrasonic transducer and a signal processing unit); the ambient light intensity is collected by a light intensity meter; and the gestational age is input by medical staff.
[0032] S102, inputting the evaluation data into a preset prediction model to obtain a bilirubin concentration risk score and a bilirubin concentration prediction value at the next prediction time.
[0033] Specifically, a hemoglobin correction factor is calculated according to the blood oxygen saturation, and a skin maturity is matched according to the gestational age of the user (i.e., the neonate), wherein the hemoglobin correction factor is positively correlated with the blood oxygen saturation. In one embodiment, the calculation expression of the hemoglobin correction factor is:
[0034] H = 0.82 * sPO2 + 0.18;
[0035] wherein H is the hemoglobin correction factor, and sPO2 is the blood oxygen saturation.
[0036] In the present embodiment, when the gestational age is between 28 and 32 weeks, the skin maturity can be between 0.3 and 0.5 (e.g., set to 0.4); when the gestational age is between 33 and 36 weeks, the skin maturity is between 0.6 and 0.8 (e.g., set to 0.7); and when the gestational age is more than 37 weeks (including 37 weeks), the skin maturity is between 0.9 and 1 (e.g., set to 0.95); wherein the neonate with a gestational age less than 37 weeks is a premature infant.
[0037] In another embodiment, the skin maturity can also be calculated according to the gestational age. Specifically, the calculation expression of the skin maturity is:
[0038]
[0039] wherein S is the skin maturity, and GA is the gestational age.
[0040] Further, the hemoglobin correction factor, the skin maturity, the spectral reflectance, the heart rate, the blood oxygen saturation, the portal flow rate, and the ambient light intensity are normalized to generate a set of nine-dimensional state vectors (one set every five seconds); and the set of nine-dimensional state vectors is input into a preset prediction model to obtain the bilirubin concentration risk score, the bilirubin concentration prediction value, and the bilirubin concentration change curve in the next 6 hours at the next predicted time (the interval between two predicted times is 5 seconds).
[0041] By inputting the hemoglobin correction factor (for compensating the interference of blood absorbance on the spectral reflectance measurement), the skin maturity (low skin maturity represents thin stratum corneum and high light transmittance, and thus the measured value of the bilirubin concentration is high), the heart rate, the blood oxygen saturation, the portal flow rate, and the ambient light intensity in addition to the spectral reflectance, the interference of the skin thickness, the blood oxygen saturation, and the like on the measurement of the bilirubin concentration can be reduced, thereby improving the accuracy of the prediction of the bilirubin concentration.
[0042] In the embodiment, the prediction model is an LSTM model. In an optional embodiment, a space-time dual-path attention module can be added to the LSTM output layer to automatically focus on the 30-minute key window period after phototherapy and reduce the interference of limb movement on the prediction of bilirubin concentration, thereby improving the accuracy of the prediction of bilirubin concentration. When training the prediction model, a basic model can be trained using 100,000 jaundice cases (the basic model can learn the general rules of jaundice cases), and then the basic model can be fine-tuned using 5,000 jaundice cases of the target hospital to adapt to the specific factors (such as equipment calibration error and regional high-risk causes) of the target hospital. Finally, online individual calibration can be performed according to the data in the use process.
[0043] S103, in response to the bilirubin concentration risk score being greater than the risk threshold, inputting the bilirubin concentration prediction value into a preset MPC optimizer to obtain optimal phototherapy parameters.
[0044] In the embodiment, the phototherapy parameters at least include the wavelength of the light source, and can also include the intensity of the light source and the irradiation time. The MPC (Model Predictive Control Optimizer) optimizer is used to solve the optimal control sequence according to the optimization function (corresponding to the optimization target) and the constraint condition at each prediction time. The bilirubin concentration risk score is obtained by the prediction model according to the change trend of the bilirubin concentration, and the value range is 0 to 10. When the bilirubin concentration risk score is greater than 3 (i.e., the risk threshold), the phototherapy is automatically started, and the treatment light source is output according to the optimal phototherapy parameters obtained by the MPC optimizer.
[0045] Specifically, the optimization target and the constraint condition of the MPC optimizer are set. In the embodiment, the optimization target at least includes minimizing the difference between the bilirubin concentration prediction value and the bilirubin concentration target value; and can also include minimizing the difference between the skin temperature and the temperature threshold.
[0046] When the optimization target is to minimize the difference between the bilirubin concentration prediction value and the bilirubin concentration target value, and to minimize the difference between the skin temperature of the user and the temperature threshold, the optimization function corresponding to the optimization target can be:
[0047] min [a · (B pred -B taeget ) 2 + β · (T skin -T0) 2 ];
[0048] In the formula, min() is the minimum value function, B pred is the bilirubin concentration prediction value, B taeget is the bilirubin concentration target value, T skinT is the skin temperature, T0 is the temperature threshold, a is a weight coefficient of the difference between the predicted value of the bilirubin concentration and the target value of the bilirubin concentration, and b is a weight coefficient of the difference between the skin temperature and the temperature threshold.
[0049] In this embodiment, the temperature threshold is set to 37℃, the skin temperature of the newborn is collected by the infrared thermal imager, a can be set to 0.6, and b can be set to 0.4.
[0050] By setting the optimization target, the problem of burns caused by the treatment light source to cause the skin temperature to be too high can be avoided while ensuring the jaundice treatment effect, thereby improving the safety and efficiency of the treatment.
[0051] In other optional embodiments, the optimization function corresponding to minimizing the difference between the skin temperature and the temperature threshold can also be:
[0052]
[0053] In the formula, min() is a minimum function, T skin T is the skin temperature, T0 is the temperature threshold, T is the skin temperature, T0 is the temperature threshold,
[0054] By additionally increasing the skin temperature change rate, the discomfort caused to the newborn by the sudden change in temperature can be prevented.
[0055] In one embodiment, the target value of the bilirubin concentration can be determined according to the bilirubin concentration risk score, wherein the target value of the bilirubin concentration is positively correlated with the bilirubin concentration risk score. Specifically,
[0056]
[0057] In the formula, B taeget T is the target value of the bilirubin concentration, risk_score1 is the bilirubin concentration risk score, GA is the gestational age, k1 is the weight coefficient, and k2 is the gestational age coefficient.
[0058] In one embodiment, when the weight of the newborn is less than 2.5 kg, the weight coefficient can be set to 1.2, and when the weight is greater than or equal to 2.5 kg, the weight coefficient can be set to 1. Further, the gestational age of the newborn can be divided by 40 to obtain the gestational age coefficient.
[0059] By setting different adjustment coefficients (weight coefficient, gestational age coefficient) according to premature infants and full-term infants, the obtained bilirubin concentration target value can be ensured to be more accurate and more suitable for the actual situation of the user; and by determining the bilirubin concentration target value according to the current bilirubin concentration risk score, the corresponding bilirubin concentration target value can be set according to the actual situation of the newborn, so that the optimal phototherapy parameter can be accurately output according to the bilirubin concentration target value in the subsequent process, thereby improving the efficiency of treatment.
[0060] In one embodiment, the constraint condition at least includes that the output optimal phototherapy parameter needs to be within the corresponding candidate region (i.e., the safe region), and can also include that the increase of the light source intensity per hour is less than a change threshold (in this embodiment, set to 10), that is, the difference between the light source intensity of the last hour and the light source intensity (optimal treatment parameter) output at the current time is less than 10.
[0061] In one embodiment, the candidate interval of the phototherapy parameter can be determined according to the bilirubin concentration risk score. In this embodiment, the optional interval of the light source wavelength can be set to 425-475 nm, the optional interval of the light source intensity can be set to 20-50 μW / cm 2 / nm, and the optional interval of the irradiation time can be set to 15-150 min.
[0062] Further, the candidate interval of the light source wavelength is determined according to the bilirubin concentration risk score, the lower limit of the candidate interval is the minimum value of the optional interval (i.e., 425 nm), and the upper limit is determined according to the bilirubin concentration risk score. Specifically, the calculation expression of the upper limit of the candidate interval of the light source wavelength is:
[0063] λ = 470 - 10 (risk_score1 - 0.5);
[0064] In the formula, λ is the upper limit of the candidate interval of the light source wavelength, and risk_score1 is the bilirubin concentration risk score.
[0065] In this embodiment, the light source intensity includes the trunk light source intensity and the limb light source intensity. The lower limit of the candidate interval corresponding to the trunk light source intensity is the minimum value of the optional interval (i.e., 20 μW / cm 2 / nm), and the upper limit is determined according to the bilirubin concentration risk score. Specifically, the calculation expression of the upper limit of the candidate interval of the trunk light source wavelength is:
[0066] I = 35 + 0.5 × risk_score1;
[0067] In the formula, I is the upper limit of the candidate interval corresponding to the trunk light source intensity, and risk_score1 is the bilirubin concentration risk score.
[0068] The lower limit of the candidate range corresponding to the intensity of the limb light source is the minimum value of the selectable range (i.e., 20 μW / cm). 2 / nm), with an upper limit of 40% or 50% of the upper limit of the candidate range corresponding to the intensity of the torso light source.
[0069] Therefore, the safe range of light source intensity in phototherapy parameters can be obtained.
[0070] Furthermore, candidate intervals corresponding to irradiation time are determined based on the bilirubin concentration risk score. The lower limit of these candidate intervals is the minimum value of the selectable intervals (i.e., 15 minutes), and the upper limit is determined based on the bilirubin concentration risk score. Specifically, the formula for calculating the upper limit of the candidate interval corresponding to the irradiation time is as follows:
[0071]
[0072] `time` represents the upper limit of the candidate interval corresponding to the irradiation time, and `risk_score1` represents the bilirubin concentration risk score.
[0073] Therefore, the safe range of irradiation time in phototherapy parameters can be obtained.
[0074] Furthermore, the predicted bilirubin concentration is input into the MPC optimizer with the optimization objectives and constraints set to obtain the optimal phototherapy parameters.
[0075] By setting a safety range, the output phototherapy parameters can be kept within a safe range, thus avoiding overtreatment. By using the MPC optimizer to select the optimal phototherapy parameters, the output phototherapy parameters can be made to match the actual situation of the newborn, avoiding undertreatment and thus improving the efficiency of treatment.
[0076] S104. Output the optimal phototherapy parameters so that the jaundice treatment device outputs the corresponding therapeutic light source.
[0077] By automatically generating optimal phototherapy parameters from the jaundice treatment device and outputting the corresponding treatment light source based on these optimal parameters, manual intervention can be reduced. It eliminates the need to rely on the experience of medical staff for setting up the device, thereby reducing the waste of human resources and operation time, improving treatment efficiency, and avoiding overtreatment.
[0078] In an optional embodiment, the method of the present invention may further include: calculating and outputting a reference bilirubin concentration risk score, so that medical personnel can adjust optimal treatment parameters based on the reference bilirubin concentration risk score. Specifically, the calculation expression for the reference bilirubin concentration risk score is:
[0079]
[0080] In the formula, risk_score2 is the reference bilirubin concentration risk score, B is the current bilirubin concentration, Bt is the target bilirubin threshold (0.8 x gestational age - 2), AH R is the heart rate variability coefficient (the ratio of the standard deviation of heart rate to the average heart rate), w1 is the first weight (which can be set to 4), w2 is the second weight (which can be set to 3), and w3 is the third weight (which can be set to 3).
[0081] It can be understood that whether the optimal phototherapy parameter output by the reference bilirubin concentration risk confirmation is reasonable can be determined by experienced medical staff, and if not, the parameter can be adjusted to further ensure the safety of the treatment, and the prediction model and the MPC optimizer are optimized according to the data before and after the adjustment.
[0082] In one embodiment, the method of the present application further comprises detecting whether the eye mask of the neonate is off through pressure sensor and camera data, and if so, triggering phototherapy suspension (response time less than 50 ms) to ensure the safety of the treatment.
[0083] In one embodiment, the method of the present application further comprises synchronizing the relevant data to the mobile terminal through Bluetooth, and displaying the real-time bilirubin concentration, phototherapy parameter, bilirubin concentration risk score and reference bilirubin concentration risk score on the mobile terminal to understand the treatment condition and effect at any time.
[0084] In one embodiment, the method of the present application further comprises triggering the micro-channel heat dissipation system to cool down when the skin temperature is detected to be higher than 37℃ to ensure the safety of the treatment.
[0085] In an optional embodiment, the method of the present application further comprises feeding back the actual irradiance through a 16-channel light intensity sensor, and correcting the output in combination with a PID controller to ensure the treatment accuracy.
[0086] Figure 2 is a structural block diagram schematically showing a control system of the jaundice treatment device according to the present embodiment.
[0087] In a second aspect, the present application further provides a control system of a jaundice treatment device. As shown in Figure 2 the control system comprises a processor and a memory, and the memory stores computer program instructions which, when executed by the processor, implement the control method of the jaundice treatment device according to the first aspect of the present application.
[0088] The control system further comprises a communication interface and other components well known to those skilled in the art, and the settings and functions thereof are known in the art, and thus will not be described here.
[0089] In this invention, the aforementioned memory can be any tangible medium containing or storing a program that can be used or combined with an instruction execution system, apparatus, or device. For example, a computer-readable storage medium can be any suitable magnetic or magneto-optical storage medium, such as Resistive Random Access Memory (RRAM), Dynamic Random Access Memory (DRAM), Static Random Access Memory (SRAM), Enhanced Dynamic Random Access Memory (EDRAM), High-Bandwidth Memory (HBM), Hybrid Memory Cube (HMC), etc., or any other medium that can be used to store desired information and can be accessed by an application, module, or both. Any such computer storage medium can be part of a device or accessible to or connected to a device. Any application or module described in this invention can be implemented using computer-readable / executable instructions that can be stored or otherwise maintained by such a computer-readable medium.
[0090] In the description of this specification, "multiple" means at least two, such as two, three or more, unless otherwise explicitly specified. Furthermore, the steps described above are for clarity only; in implementation, they can be combined into one step or some steps can be broken down into multiple steps, as long as they include the same logical relationships.
[0091] While this specification has shown and described numerous embodiments of the invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Many modifications, alterations, and alternatives will occur to those skilled in the art without departing from the spirit and essence of the invention. It should be understood that various alternatives to the embodiments of the invention described herein may be employed in the practice of this invention.
Claims
1. A control method for a jaundice treatment device, characterized in that, include: Obtain user evaluation data, including spectral reflectance; The evaluation data is input into a preset prediction model to obtain the bilirubin concentration risk score and bilirubin concentration prediction value at the next prediction time. In response to the bilirubin concentration risk score being greater than a risk threshold, a candidate interval corresponding to the phototherapy parameters is determined based on the bilirubin concentration risk score. The phototherapy parameters include the wavelength of the light source. The lower limit of the candidate interval corresponding to the wavelength of the light source is a first preset value, and the upper limit is negatively correlated with the bilirubin concentration risk score. The predicted bilirubin concentration is input into the MPC optimizer so that the MPC optimizer selects the optimal phototherapy parameters from the candidate interval; The optimal phototherapy parameters are output so that the jaundice treatment device outputs the corresponding therapeutic light source.
2. The control method for the jaundice treatment device according to claim 1, characterized in that, The formula for calculating the upper limit of the candidate interval corresponding to the wavelength of the light source is: λ=470-10(risk_score1-0.5); In the formula, λ is the upper limit of the candidate interval corresponding to the wavelength of the light source, and risk_score1 is the risk score of bilirubin concentration.
3. The control method for the jaundice treatment device according to claim 2, characterized in that, The phototherapy parameters include light source intensity, and the lower limit of the candidate interval corresponding to the light source intensity is a second preset value, while the upper limit is positively correlated with the bilirubin concentration risk score.
4. The control method for the jaundice treatment device according to claim 3, characterized in that, The light source intensity includes the light source intensity of the limbs and the light source intensity of the torso, and the phototherapy parameters also include the irradiation time.
5. The control method for the jaundice treatment device according to claim 1, characterized in that, The optimal phototherapy parameters are determined and output from the candidate intervals based on the MPC optimizer, including: Set the optimization objective and constraints for the MPC optimizer; the optimization objective includes minimizing the difference between the predicted bilirubin concentration and the target bilirubin concentration; the constraints include ensuring that the optimal phototherapy parameters belong to the corresponding candidate interval; The predicted bilirubin concentration is input into the MPC optimizer to obtain the optimal phototherapy parameters.
6. The control method for the jaundice treatment device according to claim 5, characterized in that, The optimization objective also includes minimizing the difference between skin temperature and a temperature threshold, and the constraint also includes ensuring that the increase in light source intensity per hour is less than a change threshold.
7. The control method for the jaundice treatment device according to claim 6, characterized in that, When the optimization objective is to minimize the difference between the predicted bilirubin concentration and the target bilirubin concentration, and to minimize the difference between skin temperature and the temperature threshold, the optimization function corresponding to the optimization objective is: min[α·(B pred -B taeget ) 2 +β·(T skin -T0) 2 ]; In the formula, min() is the minimum value function, B pred B is the predicted value of bilirubin concentration. taeget T represents the target value for bilirubin concentration. skin T0 is the skin temperature, α is the weighting coefficient of the difference between the predicted bilirubin concentration and the target bilirubin concentration, and β is the weighting coefficient of the difference between the skin temperature and the temperature threshold.
8. The control method for the jaundice treatment device according to claim 6, characterized in that, Also includes: The target value of bilirubin concentration is determined based on the bilirubin concentration risk score, and the target value of bilirubin concentration is positively correlated with the bilirubin concentration risk score.
9. The control method for the jaundice treatment device according to claim 1, characterized in that, The spectral reflectance includes a first spectral reflectance, a second spectral reflectance, and a third spectral reflectance. The evaluation data also includes heart rate, blood oxygen saturation, portal vein flow velocity, and ambient light intensity. Inputting the evaluation data into a preset prediction model includes: The hemoglobin correction factor is calculated based on the blood oxygen saturation, and the skin maturity is determined based on the user's corresponding gestational age. The hemoglobin correction factor is positively correlated with the blood oxygen saturation, and the skin maturity is positively correlated with the gestational age. The hemoglobin correction factor, skin maturity, spectral reflectance, heart rate, blood oxygen saturation, portal vein flow velocity, and ambient light intensity are normalized to generate a set of nine-dimensional state vectors; The nine-dimensional state vector is input into a preset prediction model.
10. A control system for a jaundice treatment device, characterized in that, It includes a processor and a memory, the memory storing computer program instructions that, when executed by the processor, implement the control method of the jaundice treatment device according to any one of claims 1-9.