Disclosure of Invention
In view of the problems in the background art, the invention provides an intelligent nursing system and method for a patient after breast cancer operation.
The problem to be solved by the present invention is therefore how to combine multidimensional patient data with machine learning algorithms and to make personalized dynamic adjustments for the actual situation of the patient.
In order to solve the technical problems, the invention provides the following technical scheme:
the invention provides an intelligent nursing method for a breast cancer postoperative patient, which comprises the steps of collecting postoperative physiological data of the patient through intelligent wearing equipment, uploading the collected real-time data to a cloud platform, constructing a multi-dimensional data association model by combining individual features of the patient based on the postoperative physiological data of the patient and taking the standardized feature data as input variables, outputting a multi-dimensional feature vector, calculating the association degree of each feature, constructing a lymphedema early prediction model by a machine learning algorithm based on the multi-dimensional data association model, predicting the occurrence probability of lymphedema, generating an individualized intelligent nursing plan according to the early prediction result, continuously monitoring and optimizing the individualized nursing intervention plan, continuously monitoring physiological data of the patient, and comparing the individualized nursing intervention plan with the intelligent nursing plan to obtain a real-time rehabilitation progress report.
The invention relates to an intelligent nursing method for a breast cancer postoperative patient, which is a preferable scheme, wherein, based on postoperative physiological data of the patient, a machine learning algorithm is used for constructing a lymphedema early prediction model, and the method comprises the following steps of extracting key features from cloud platform data, performing standardized processing, constructing a multidimensional data correlation model by taking the standardized feature data as an input variable and combining individual features of the patient, and outputting a multidimensional feature vectorWherein, the method comprises the steps of, wherein,The method comprises the steps of obtaining feature vectors, calculating the association degree of each feature, and constructing a lymphedema early prediction model based on a multidimensional data association model.
As a preferable scheme of the breast cancer postoperative patient intelligent care method, the calculation formula of the association degree of each feature is as follows:
;
Wherein,For the predictive output of the multidimensional data correlation model,Is the weight of the corresponding feature.
As a preferable scheme of the breast cancer postoperative patient intelligent care method, the construction of the lymphedema early-stage prediction model comprises the steps of calculating a prediction value through a logistic regression model to generate the occurrence probability of lymphedema:
;
Dynamically updating the model using a sliding window algorithm, ensuring that the most up-to-date physiological data is used for prediction:
;
Wherein,Is the predicted probability of time t.
The method for intelligently nursing the breast cancer postoperative patient is a preferable scheme, wherein the generation of the personalized intelligent nursing plan according to the early prediction result comprises the following steps of generating the intelligent nursing plan according to the prediction probability of lymphedema, and adjusting nursing intervention intensity and type according to real-time feedback data.
As a preferable scheme of the breast cancer postoperative patient intelligent nursing method, the method for adjusting the nursing intervention intensity and type according to the real-time feedback data comprises the following steps of setting swelling change acquired in real time to beAnd pain index ofAdjust the nursing intervention intensityUpdating is performed by the following formula:
;
Wherein,In order to adjust the intensity of the care intervention after adjustment,For the intensity of the current care intervention,As the amount of change in swelling at the present moment,AndAnd dynamically adjusting the nursing intervention type according to the real-time data fed back by the patient.
The continuous monitoring and optimization of the personalized nursing intervention plan comprises the following steps that after each nursing intervention plan is executed, nursing intervention effect is estimated through comprehensive evaluation feedback data and self-report of the patient, and a nursing intervention effect score is obtained:
;
Wherein,AndIn order to evaluate the coefficients of the coefficients,To care for the amount of swelling change after dry-out,Is the pain index variable; the nursing intervention effect is scored, wherein positive values are good effects, and negative values are poor effects.
The invention provides an intelligent nursing system for a breast cancer postoperative patient, which comprises a data acquisition module, a data analysis module, an intelligent nursing plan generation module and a rehabilitation progress monitoring module, wherein the data acquisition module is used for acquiring postoperative physiological data of the patient through intelligent wearing equipment and uploading the acquired real-time data to a cloud platform, the data analysis module is used for constructing a lymphedema early prediction model through a machine learning algorithm based on the postoperative physiological data of the patient to predict the occurrence probability of lymphedema, the intelligent nursing plan generation module is used for generating an individualized intelligent nursing plan according to an early prediction result, and the rehabilitation progress monitoring module is used for continuously monitoring the physiological data of the patient and comparing the physiological data with the intelligent nursing plan to obtain a real-time rehabilitation progress report.
In a third aspect, the invention provides a computer device comprising a memory and a processor, the memory storing a computer program, wherein the computer program instructions, when executed by the processor, implement the steps of the breast cancer postoperative patient intelligent care method according to the first aspect of the invention.
In a fourth aspect, the present invention provides a computer readable storage medium having stored thereon a computer program, wherein the computer program instructions when executed by a processor implement the steps of the breast cancer postoperative patient intelligent care method according to the first aspect of the present invention.
The invention has the beneficial effects that the early prediction of lymphedema is carried out by integrating various sensor data and adopting an advanced machine learning algorithm, and a personalized intelligent care plan is generated on the basis, so that the rehabilitation state of a patient can be monitored in real time, and the care plan can be dynamically adjusted according to the progress. The invention not only can evaluate the rehabilitation progress of the patient in real time according to comprehensive physiological data and movement data, but also can carry out intelligent personalized optimization on the nursing plan through data feedback, thereby greatly improving the accuracy and effect of postoperative nursing.
Detailed Description
In order that the above-recited objects, features and advantages of the present invention will become more readily apparent, a more particular description of the invention will be rendered by reference to specific embodiments thereof which are illustrated in the appended drawings.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention, but the present invention may be practiced in other ways other than those described, and persons skilled in the art will readily appreciate that the present invention is not limited to the specific embodiments disclosed below.
Further, reference herein to "one embodiment" or "an embodiment" means that a particular feature, structure, or characteristic can be included in at least one implementation of the invention. The appearances of the phrase "in one embodiment" in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments.
Embodiment 1 referring to fig. 1 to 2, in a first embodiment of the present invention, an intelligent care method for a patient after breast cancer surgery is provided, as shown in fig. 1, including:
S1, acquiring postoperative physiological data of a patient through intelligent wearing equipment (such as an intelligent armband, a bracelet, a sensor and the like), wherein the postoperative physiological data comprise multidimensional data of swelling degree (measured through pressure and volume change), activity, body temperature, heart rate and the like, and the data can provide a basis for subsequent analysis.
Wherein, the data acquisition should have high frequency and high accuracy, especially the swelling, heart rate and body temperature data, and the real-time monitoring is very critical, and it is recommended to acquire the data every 30 seconds, so as to ensure that sudden changes can be tracked.
The acquisition frequency of the activity data can be properly adjusted, and the data is usually acquired for 3-5 times within 24 hours or recorded after the patient moves according to a preset program.
The data collected by various sensors need to be synchronized, so that the consistency and consistency of the multidimensional data are ensured. For example, swelling, heart rate and body temperature data need to be collected and processed simultaneously in order to analyze the postoperative condition of the patient as a whole, and data can be collected in different dimensions through a multi-sensor data fusion technology to form a comprehensive physiological data set.
Further, the collected real-time data is uploaded to a cloud platform, cloud storage and processing are carried out through a data transmission interface (such as Bluetooth and Wi-Fi), and the cloud platform integrates and preprocesses the data to provide a high-quality data set for subsequent analysis.
S2, constructing a lymphedema early prediction model through a machine learning algorithm based on postoperative physiological data of the patient, and predicting the occurrence probability of lymphedema by combining the postoperative physiological data of the patient, so as to provide basis for nursing intervention.
S2.1 key features such as degree of swelling (rate of change of pressure and volume change), amount of activity (frequency of movement, intensity and duration), body temperature (local temperature change) and heart rate (heart rate fluctuation amplitude) are extracted from cloud platform data. In order to ensure data consistency and effectiveness, multidimensional data acquired by different sensors are subjected to standardized processing, so that all features are in the same dimension for subsequent modeling and analysis.
Specifically, the pressure change rate (degree of swelling) was calculated as follows:
;
Wherein,In order to provide a rate of change of pressure,The pressure value (unit: pa) at the current time,The pressure value (unit: pa) at the previous time.
The volume change rate (degree of swelling) was calculated as follows:
;
Wherein,In order to provide a rate of change of volume,The volume value (unit: cm 3) at the current time,The volume value (unit: cm 3) at the previous time is given.
The activity was calculated as follows:
;
Wherein,Is the activity (unit: m/s 2),Acceleration values (unit: m/s 2) for the i-th sampling point,Is the total number of sampling points.
Body temperature changeThe (local temperature change) is the difference between the body temperature at the current time and the body temperature at the previous time.
The formula of the heart rate fluctuation amplitude is:
;
Wherein,Is the heart rate fluctuation amplitude (unit: bpm 2),For the heart rate value (unit: bpm) of the ith sample,Is the average of heart rate (unit: bpm).
S2.2, constructing a multidimensional data association model.
Taking the standardized characteristic data as an input variable, combining individual characteristics (such as age, sex, operation type and the like) of a patient, constructing a multi-dimensional data association model, and outputting a multi-dimensional characteristic vector by the modelWherein, the method comprises the steps of, wherein,The number of the features in the feature vector is used for obtaining an association matrix through a training process, and the association degree of each feature can be expressed as follows:
;
Wherein,For the predictive output of the multidimensional data correlation model,Is the weight of the corresponding feature.
It should be noted that the multidimensional data correlation model finds potential correlations between these physiological parameters by analyzing the relationships in the historical data and evaluates the extent to which they affect the occurrence of lymphedema.
S2.3, establishing a lymphedema early prediction model and calculating the occurrence probability of lymphedema.
And constructing a lymphedema early prediction model based on the multidimensional data correlation model, wherein the lymphedema early prediction model can calculate the occurrence probability of lymphedema according to the input real-time physiological data.
Preferably, the prediction value is calculated through a logistic regression model to generate the occurrence probability of lymphedema:
;
Further, the model is dynamically updated by using a sliding window algorithm, so that the latest physiological data is used for prediction:
;
Wherein,Is the predicted probability of time t.
It should be noted that the lymphedema early prediction model is outputted as a continuous prediction probability value (between 0 and 1) reflecting the risk level of lymphedema occurrence.
And S3, generating a personalized intelligent care plan (such as local pressure therapy, massage, exercise and the like) according to the early prediction result, and automatically adjusting the care intervention intensity according to feedback of a patient and real-time monitoring data (such as swelling change, pain index and the like) to ensure the effect.
And S3.1, generating an intelligent care plan according to the prediction probability of lymphedema.
According to step S2.3Determining whether a care intervention needs to be initiated, wherein a threshold (adjustable according to clinical experience, typically between 0.5 and 0.8) for initiation of the care intervention needs to be set, willIn contrast to the threshold for initiation of care interventions, ifGreater than or equal to a threshold for initiation of a care intervention. Personalized care interventions are generated based on the patient's predicted condition and medical guidelines including, but not limited to, localized pressure therapy (e.g., using pressure socks or straps), exercise (e.g., moderate physical therapy exercises, gait training), massage (e.g., lymphatic drainage; massage), and drug care interventions (e.g., anti-inflammatory drugs).
And S3.2, adjusting the nursing intervention intensity and type according to the real-time feedback data.
By monitoring real-time physiological data (e.g., swelling changes, pain index, activity, etc.) of the patient, the care intervention strategy is continually adjusted.
Suppose that the swelling change acquired in real time isAnd pain index ofAdjust the nursing intervention intensityUpdating is performed by the following formula:
;
Wherein,For the adjusted care intervention intensity (which may be local pressure, massage intensity, etc.),For the intensity of the current care intervention,As the amount of change in swelling at the present moment,AndFor adjustment factors, it is learned through clinical data.
Further, the nursing intervention type is dynamically adjusted according to real-time data fed back by the patient. For example, if the patient feedback swelling is reduced but the pain index is rising, the care intervention regimen may be adjusted to more focus on the strategy of pain relief, such as increasing the intensity of the massage and decreasing the intensity of the local pressure therapy.
And S4, continuously monitoring and optimizing the personalized nursing intervention plan, continuously monitoring physiological data of the patient, comparing the physiological data with the intelligent nursing plan, evaluating the rehabilitation progress in real time, providing scientific feedback according to the data, and helping the patient and doctors to know the recovery condition.
After each nursing intervention scheme is executed, the nursing intervention effect is estimated by comprehensively estimating feedback data and self-report of the patient, and a grading of the nursing intervention effect is obtainedSuppose that the change in swelling after the dry nursing isThe pain index change amount isThe care intervention effect can be evaluated by the following formula:
;
Wherein,AndTo evaluate the coefficients.
The nursing intervention effect is scored, wherein positive values are good effects, and negative values are poor effects.
If nursing intervention effect scoresBelow the set effect threshold, a further adjustment of the care intervention plan is required, updating the care intervention strategy by the following formula:
;
Wherein,In order to optimize the intensity of the care intervention after it has been performed,To optimize the factor, the magnitude of the care intervention intensity or type adjustment is expressed.
Further, based on the personalized intelligent care plan formulated in the step S3, the real-time monitoring data is compared with a preset rehabilitation target in the intelligent care plan to obtain the difference between the real-time data and the preset planAnd evaluating the difference of the current progress according to the comparison result. If it isExceeding a certain threshold (e.g., a set recovery tolerance range), it is indicated that the patient's recovery schedule may not be as expected, requiring care intervention or adjustment of the treatment regimen.
Based on the real-time data comparison and the rehabilitation progress assessment, the patient will obtain clear and concise progress reports including, but not limited to, whether the swelling changes, pain indexes, whether the pain is relieved, whether the medication or nursing intervention is required to be adjusted, and whether the activity meets the rehabilitation requirements.
The physician will receive detailed reports based on patient real-time data and feedback, including patient physiological data changes, assessment of rehabilitation progress, performance of care intervention programs, whether adjustments to treatment are needed, etc.
For patients with poor progress, the system will automatically generate corresponding care intervention adjustment recommendations and notify the physician. For example, a certain care intervention (e.g., increasing local pressure or increasing movement) may be added, or the medication regimen may be adjusted.
The adjusted proposal can be pushed in real time through intelligent equipment or APP, and reminds patients and doctors of paying attention to important changes.
Further, this embodiment also provides a breast cancer postoperative patient intelligent care system, including:
The data acquisition module is used for acquiring postoperative physiological data of a patient through the intelligent wearable device and uploading the acquired real-time data to the cloud platform;
the data analysis module is used for constructing an early lymphedema prediction model through a machine learning algorithm based on postoperative physiological data of the patient and predicting the occurrence probability of lymphedema;
the intelligent care plan generation module is used for generating a personalized intelligent care plan according to the early prediction result;
And the rehabilitation progress monitoring module is used for continuously monitoring physiological data of a patient and comparing the physiological data with the intelligent care plan to obtain a real-time rehabilitation progress report.
The embodiment also provides computer equipment, which is suitable for the condition of the breast cancer postoperative patient intelligent care method and comprises a memory and a processor, wherein the memory is used for storing computer executable instructions, and the processor is used for realizing the breast cancer postoperative patient intelligent care method according to the embodiment when executing the computer executable instructions.
The computer device may be a terminal comprising a processor, a memory, a communication interface, a display screen and input means connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The communication interface of the computer device is used for carrying out wired or wireless communication with an external terminal, and the wireless mode can be realized through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer equipment can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, can also be keys, a track ball or a touch pad arranged on the shell of the computer equipment, and can also be an external keyboard, a touch pad or a mouse and the like.
The present embodiment also provides a storage medium having stored thereon a computer program which, when executed by a processor, implements a method for implementing intelligent care of a patient after breast cancer surgery as set forth in the above embodiments.
In summary, the invention integrates various sensor data, adopts advanced machine learning algorithm to predict lymphedema early, generates personalized intelligent nursing plan on the basis, can monitor the rehabilitation state of the patient in real time, and dynamically adjusts the nursing plan according to the progress. The invention not only can evaluate the rehabilitation progress of the patient in real time according to comprehensive physiological data and movement data, but also can carry out intelligent personalized optimization on the nursing plan through data feedback, thereby greatly improving the accuracy and effect of postoperative nursing.
Example 2 referring to tables 1 and 2, a second example of the present invention provides a method for intelligent care of a patient after breast cancer surgery, which is scientifically demonstrated through simulation experiments in order to verify the beneficial effects of the present invention.
The embodiment focuses on evaluating the role of a multidimensional physiological data monitoring and machine learning prediction model based on intelligent wearable equipment in lymphedema early prediction and personalized care intervention.
In the experiment, 10 breast cancer postoperative patients are selected, all patients wear intelligent wearing equipment such as intelligent wrist bands and armbands, and multidimensional physiological data after the operation are monitored in real time. The frequency of data acquisition was that swelling, body temperature and heart rate were acquired every 30 seconds, and activity was acquired at different time points per day. The data includes the extent of swelling (measured by pressure and volume changes), the amount of activity (frequency, intensity and duration of movement), body temperature, heart rate, etc.
In the experiment, the real-time physiological data of the patient is firstly transmitted to the cloud platform through Bluetooth for storage and preprocessing, and the cloud platform performs standardized processing on the data to ensure the data consistency. These data are then analyzed by machine learning algorithms, an early prediction model of lymphedema is constructed, and a personalized intelligent care regimen is generated based on the prediction results. Nursing interventions include localized pressure therapy (e.g., using pressure socks), moderate exercise, lymphatic drainage massage, and the like.
For each patient, the care intervention intensity and type are adjusted in real time according to the physiological data and feedback after the operation. In the intervention process, the data are synchronously updated, and the nursing effect is periodically evaluated. If the swelling level and pain index are not significantly improved, the care plan is dynamically adjusted to optimize the patient's rehabilitation regimen.
In this experiment, physiological data, predictive results, and care intervention data for all patients are recorded in two tables:
table 1 patient physiological data recording table
| Parameters (parameters) | Patient 1 | Patient 2 | Patient 3 | Patient 4 | Patient 5 | Patient 6 | Patient 7 |
| Swelling variable (cm 3) | 12.5 | 15.2 | 10.3 | 14.8 | 13.7 | 16.1 | 14.3 |
| Body temperature change (°c) | 0.4 | 0.5 | 0.3 | 0.6 | 0.5 | 0.4 | 0.4 |
| Heart rate fluctuation amplitude (bpm 2) | 30 | 28 | 25 | 35 | 29 | 27 | 32 |
| Activity amount (m/s 2) | 0.25 | 0.30 | 0.28 | 0.26 | 0.24 | 0.29 | 0.31 |
| Pain index (0-10) | 3 | 4 | 2 | 3 | 3 | 4 | 2 |
| Lymphedema prediction probability (%) | 45 | 65 | 30 | 72 | 58 | 61 | 55 |
Table 2 nursing intervention data record table
| Parameters (parameters) | Patient 1 | Patient 2 | Patient 3 | Patient 4 | Patient 5 | Patient 6 | Patient 7 |
| Initial swelling change (cm 3) | 12.5 | 15.2 | 10.3 | 14.8 | 13.7 | 16.1 | 14.3 |
| Intensity of care intervention expected (%) | 70 | 75 | 65 | 80 | 72 | 78 | 76 |
| Local pressure therapy (hours) | 4 | 5 | 3 | 5 | 4 | 4 | 4 |
| Exercise intervention (hours) | 2 | 3 | 1.5 | 2.5 | 2 | 2.5 | 3 |
| Massage intervention (hours) | 1 | 1 | 1 | 1 | 1 | 1 | 1 |
| Swelling variable after operation (cm 3) | 9.8 | 12.3 | 8.9 | 12.0 | 11.2 | 13.0 | 11.0 |
First, as can be seen from the data in table 1, all patients have a predicted lymphedema probability between 40% and 70%, demonstrating that the risk of early post-operative lymphedema can be effectively captured by the predictive model of the present invention. For example, patient 4 has a predictive probability as high as 72%, suggesting a greater risk of lymphedema, at which point the system recommends intensive care intervention.
In the care intervention data record form of table 2, the amount of swelling was relieved to varying degrees after the patient underwent a personalized intervention regimen. Especially, the initial swelling amount of the patient 1 is 12.5 cm3, and the postoperative swelling change amount is reduced to 9.8 cm3 through local pressure treatment (4 hours), exercise intervention (2 hours) and massage (1 hour), so that the intelligent nursing method can effectively reduce postoperative swelling and relieve discomfort of the patient.
Further analysis shows that nursing intervention schemes of all patients are remarkably improved in postoperative effect, and early warning and personalized intervention provided by the invention through real-time data monitoring and a machine learning model are remarkably improved in rehabilitation effect. This also demonstrates the limitations of traditional care approaches in facing individual differences and complex physiological data, while smart care approaches can provide accurate treatment regimens for each patient, maximally reducing the risk of post-operative complications (such as lymphedema) occurring.
Compared with the prior art, the invention not only breaks through the accuracy of swelling prediction and nursing intervention, but also ensures individuation and effect optimization of a nursing scheme through real-time data feedback and dynamic adjustment of intervention intensity. In addition, the model based on machine learning can predict and evaluate risks according to big data, errors and hysteresis of manual judgment are reduced, and nursing effect and recovery speed of patients are improved.
It should be noted that the above embodiments are only for illustrating the technical solution of the present invention and not for limiting the same, and although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solution of the present invention may be modified or substituted without departing from the spirit and scope of the technical solution of the present invention, which is intended to be covered in the scope of the claims of the present invention.