Music recommendation method and system based on nursing data

Abstract

The invention discloses a music recommendation method and system based on nursing data, relates to the field of nursing, and solves the problem that an existing music recommendation method for a patient in a nursing period is poor in application effect, and the method comprises the steps: S1, obtaining a nursing monitoring period, dividing the nursing monitoring period into a plurality of music recommendation time periods, s2, according to the patient nursing monitoring data, obtaining music recommendation preliminary classification data and a feature expression image appearance ratio, obtaining patient nursing analysis data, and according to the patient nursing analysis data, obtaining a first music recommendation coefficient and a second music recommendation coefficient corresponding to each music recommendation time period; and S3, according to the patient nursing analysis data, music recommendation is carried out on patients in different music recommendation periods, the reasonability and accuracy of the treatment music recommendation method in the nursing period of the patient can be ensured, and thus the body recovery of the patient is promoted.

Description

Technical Field

[0001] The present invention belongs to the field of nursing and relates to data analysis technology, and specifically to a music recommendation method and system based on nursing data. Background Art

[0002] The existing music recommendation methods based on nursing data have the following defects when recommending music to patients:

[0003] 1. Existing nursing data focuses on physiological parameters, but ignores the patient's emotional state. Without monitoring and analysis of the patient's emotions, the recommended music may not have a positive effect on the patient.

[0004] 2. The existing music recommendation methods based on nursing data obtain relatively single data and are unable to fully combine various physiological indicator data and emotional data to make music recommendations, resulting in a lack of accuracy in music recommendations.

[0005] To this end, we propose a music recommendation method and system based on nursing data. Summary of the invention

[0006] In view of the shortcomings of the prior art, the purpose of the present invention is to provide a music recommendation method and system based on nursing data. The present invention is based on obtaining a nursing monitoring cycle, dividing the nursing monitoring cycle into multiple music recommendation time periods, respectively obtaining a first music recommendation coefficient and a second music recommendation coefficient corresponding to each music recommendation time period, obtaining patient nursing monitoring data, obtaining a preliminary music recommendation coefficient based on the patient nursing monitoring data, dividing multiple sample music recommendation time periods into a first type of music recommendation time period and a second type of music recommendation time period, obtaining preliminary music recommendation classification data, and performing facial expression analysis on patients corresponding to the second type of music recommendation time period to obtain a characteristic facial expression image appearance ratio, defining the preliminary music recommendation classification data and the characteristic facial expression image appearance ratio as patient nursing analysis data, and performing music recommendations for patients in different music recommendation time periods based on the patient nursing analysis data.

[0007] In order to achieve the above object, the present invention adopts the following technical solution: A music recommendation method based on nursing data includes the following specific steps:

[0008] Step S1: obtaining a nursing monitoring cycle, dividing the nursing monitoring cycle into multiple music recommendation time periods, respectively obtaining a first music recommendation coefficient and a second music recommendation coefficient corresponding to each music recommendation time period, and obtaining patient nursing monitoring data;

[0009] Step S2: Obtaining preliminary music recommendation coefficients according to patient care monitoring data to divide multiple sample music recommendation periods into a first type of music recommendation period and a second type of music recommendation period, obtaining preliminary music recommendation classification data, and performing facial expression analysis on patients corresponding to the second type of music recommendation period to obtain a characteristic facial expression image appearance ratio, and defining the preliminary music recommendation classification data and the characteristic facial expression image appearance ratio as patient care analysis data;

[0010] Step S3: Recommend music to patients in different music recommendation time periods based on the patient care analysis data.

[0011] Furthermore, the step S1 further includes the following specific steps:

[0012] Step S11: marking the period of time during which the patient receives nursing treatment every day as a nursing monitoring cycle, and dividing the nursing monitoring cycle into a number of music recommendation periods of equal length;

[0013] Step S12: randomly selecting a time period from the obtained multiple music recommendation time periods as a sample music recommendation time period;

[0014] Step S13: monitoring physiological indicators of patients in the sample music recommendation period to obtain a first music recommendation coefficient;

[0015] Step S14: performing skin electricity monitoring on the patient in the sample music recommendation period to obtain a second music recommendation coefficient;

[0016] Step S15: Repeat step S13 and step S14 to obtain the first music recommendation coefficient and the second music recommendation coefficient corresponding to each music recommendation period, respectively, to obtain patient care monitoring data.

[0017] Furthermore, the step S13 further includes the following specific steps:

[0018] Step S131: Mark the start time point of the sample music recommendation period as the first period characteristic time point, mark the time point corresponding to a physiological characteristic duration after the first period characteristic time point as the second period characteristic time point, and mark the time period between the first period characteristic time point and the second period characteristic time point as the physiological monitoring sub-period of the sample music recommendation period;

[0019] Step S132: randomly selecting a number of physiological monitoring time points in the physiological monitoring sub-period, and naming them as the first physiological monitoring time point to the ath physiological monitoring time point;

[0020] Step S133: respectively obtaining the diastolic blood pressure values ​​of the patient corresponding to the first physiological monitoring time point to the ath physiological monitoring time point, obtaining the first diastolic pressure value to the ath diastolic pressure value, and averaging the first diastolic pressure value to the ath diastolic pressure value to obtain the average diastolic pressure of the time period;

[0021] Step S134: respectively obtaining the systolic blood pressure values ​​of the patient corresponding to the first physiological monitoring time point to the ath physiological monitoring time point, obtaining the first systolic blood pressure value to the ath systolic blood pressure value, and averaging the first systolic blood pressure value to the ath systolic blood pressure value to obtain the average systolic blood pressure of the time period;

[0022] Step S135: performing heart rate analysis on the patients in the physiological monitoring sub-periods respectively to obtain heart rate index values ​​for the period;

[0023] Step S136: Calculate the average diastolic pressure, average systolic pressure and heart rate index values ​​of the time period to obtain a first music recommendation coefficient corresponding to the sample music recommendation time period;

[0024] A first music recommendation coefficient is calculated.

[0025] Furthermore, the step S135 further includes the following specific steps:

[0026] Step S1351: respectively obtaining the real-time heart rate values ​​of the patient corresponding to the first physiological monitoring time point to the ath physiological monitoring time point, obtaining the first heart rate value to the ath heart rate value, and averaging the first heart rate value to the ath heart rate value to obtain the time period average heart rate value;

[0027] Step S1352: Compare the values ​​of the first heart rate value to the ath heart rate value, and mark the heart rate value with the largest value as the peak heart rate value of the period;

[0028] Step S1353: Calculate the variance of the first heart rate value to the ath heart rate value to obtain the heart rate variance of the time period monitoring;

[0029] Step S1354: Calculate the average heart rate value of the time period, the peak heart rate value of the time period, and the variance of the monitored heart rate of the time period to obtain a heart rate index value of the time period;

[0030] Calculate the heart rate index value for each period.

[0031] Furthermore, the step S14 further includes the following specific steps:

[0032] Step S141: Mark the start time point of the sample music recommendation period as the first period characteristic time point, mark the time point corresponding to the emotion monitoring duration after the first period characteristic time point as the third period characteristic time point, and mark the time period between the first period characteristic time point and the third period characteristic time point as the skin electricity monitoring sub-period of the sample music recommendation period;

[0033] Step S142: In the skin electrical monitoring sub-period, a plurality of consecutive time points with equal time intervals are selected as emotion monitoring time points, and the obtained plurality of emotion monitoring time points are named as the first emotion monitoring time point to the bth emotion monitoring time point respectively;

[0034] Step S143: obtaining the skin conductivities of the patient corresponding to the first emotion monitoring time point to the bth emotion monitoring time point, and obtaining the first skin conductivities to the bth skin conductivities;

[0035] Step S144: obtaining a skin conductivity change value by calculating the first skin conductivity to the bth skin conductivity;

[0036] Calculate the change value of skin conductivity;

[0037] Step S145: Calculate the average of the first skin conductivity to the bth skin conductivity to obtain a monitored average conductivity;

[0038] Step S146: comparing the values ​​of the first skin conductivity to the bth skin conductivity, and marking the skin conductivity with the largest value as the monitored peak skin conductivity;

[0039] Step S147: Calculate the skin conductivity change value, the monitored average conductivity, and the monitored peak skin conductivity to obtain a second music recommendation coefficient;

[0040] The second music recommendation coefficient is calculated.

[0041] Furthermore, the step S2 further includes the following specific steps:

[0042] Step S21: Obtain patient care monitoring data, and obtain a first music recommendation coefficient and a second music recommendation coefficient corresponding to each music recommendation period according to the patient care monitoring data;

[0043] Step S22: Calculating the first music recommendation coefficient and the second music recommendation coefficient corresponding to each music recommendation period to obtain a plurality of preliminary music recommendation coefficients;

[0044] Calculate the preliminary music recommendation coefficient;

[0045] Step S23: obtaining a music preliminary recommendation coefficient threshold, comparing multiple music preliminary recommendation coefficients with the music preliminary recommendation coefficient threshold respectively, and dividing multiple music recommendation periods into a first type of music recommendation period and a second type of music recommendation period according to the comparison result, to obtain music recommendation preliminary classification data;

[0046] Step S24: performing expression analysis on the second type of music recommendation period to obtain a characteristic expression image appearance ratio;

[0047] Step S25: Define the music recommendation preliminary classification data and the characteristic expression image appearance ratio as patient care analysis data.

[0048] Furthermore, the step S23 further includes the following specific steps:

[0049] Step S231: respectively obtaining a first music recommendation coefficient threshold and a second music recommendation coefficient threshold, and calculating the first music recommendation coefficient threshold and the second music recommendation coefficient threshold to obtain a preliminary music recommendation coefficient threshold;

[0050] Step S232: when the music preliminary recommendation coefficient is greater than or equal to the music preliminary recommendation coefficient threshold, the corresponding music recommendation period is divided into a first type of music recommendation period;

[0051] Step S233: When the music preliminary recommendation coefficient is less than the music preliminary recommendation coefficient threshold, the corresponding music recommendation period is divided into a second type of music recommendation period.

[0052] Furthermore, the step S24 further includes the following specific steps:

[0053] Step S241: randomly selecting a time period in the music recommendation time period corresponding to the second type of music recommendation time period as an expression recognition sample time period;

[0054] Step S242: marking the start time point of the expression recognition sample period as the first image feature time point, marking the time point corresponding to an image monitoring duration after the first image feature time point as the second image feature time point, marking the time period between the first image feature time point and the second image feature time point as the expression monitoring sub-period of the expression recognition sample period, and randomly acquiring facial expression images of a number of patients within the expression monitoring sub-period to obtain a plurality of patient expression images;

[0055] Step S243: Counting the number of randomly acquired facial expression images to obtain the total number of facial expression images;

[0056] Step S244: randomly crawling a plurality of human face images from the Internet of Things by using crawler technology, and annotating the obtained plurality of human face images into first expression type images and second expression type images by manual annotation;

[0057] Step S245: dividing the plurality of human face images into a model test set, a model training set, and a model verification set, establishing an image recognition model by using an existing convolutional neural network model technology, training the image recognition model by using the model test set, the model training set, and the model verification set, and testing the trained image recognition model, and naming the tested image recognition model as a feature expression recognition model;

[0058] Step S246: Use the characteristic expression recognition model to divide multiple patient expression images into first expression type images and second expression type images, and obtain the number of patient expression images corresponding to the first expression type images to obtain the characteristic expression image quantity value, calculate the ratio of the characteristic expression image quantity value to the total number of facial expression images, and obtain the characteristic expression image appearance ratio.

[0059] Furthermore, the step S3 further includes the following specific steps:

[0060] Step S31: Obtain patient care analysis data, and obtain music recommendation preliminary classification data and characteristic expression image appearance ratio according to the patient care analysis data;

[0061] Step S32: when the patient is in the first type of music recommendation period, calm music is recommended to the patient;

[0062] Step S33: when the patient is in the second type of music recommendation period, obtaining a characteristic expression image appearance ratio threshold and performing numerical comparison with the characteristic expression image appearance ratio;

[0063] Step S34: when the characteristic expression image appearance ratio is greater than or equal to the characteristic expression image appearance ratio threshold, calming music is recommended to the patient in the second type of music recommendation period;

[0064] Step S35: When the characteristic expression image appearance ratio is less than the characteristic expression image appearance ratio threshold, dynamic music is recommended to the patient in the second type of music recommendation period.

[0065] A music recommendation system based on nursing data. The specific working process of each module is as follows:

[0066] Data acquisition module: used to obtain a nursing monitoring cycle, divide the nursing monitoring cycle into multiple music recommendation time periods, respectively obtain a first music recommendation coefficient and a second music recommendation coefficient corresponding to each music recommendation time period, and obtain patient nursing monitoring data;

[0067] Data analysis module: used to obtain preliminary music recommendation coefficients according to patient care monitoring data, divide multiple sample music recommendation periods into first-type music recommendation periods and second-type music recommendation periods, obtain preliminary music recommendation classification data, and perform facial expression analysis on patients corresponding to the second-type music recommendation periods to obtain characteristic facial expression image appearance ratios, and define the preliminary music recommendation classification data and characteristic facial expression image appearance ratios as patient care analysis data;

[0068] Music recommendation module: used to make music recommendations to patients in different music recommendation periods based on patient care analysis data.

[0069] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0070] 1. The present invention divides a nursing monitoring cycle into multiple music recommendation time periods, and obtains a first music recommendation coefficient and a second music recommendation coefficient corresponding to each music recommendation time period, which can effectively improve the comprehensiveness of nursing data analysis and fully ensure that the music played during the treatment process can have a positive effect on the patient's physical recovery;

[0071] 2. The present invention performs expression analysis on patients corresponding to the second type of music recommendation period, can combine nursing data with expression data, and use comprehensive thinking to recommend music to patients, thereby ensuring the accuracy of music recommendations for patients. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] In order to facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0073] Figure 1 It is a diagram of the implementation steps of the present invention;

[0074] Figure 2 is the overall system block diagram of the present invention;

[0075] Figure 3 Schematic diagram of the convolutional neural network technology in the present invention. DETAILED DESCRIPTION

[0076] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0077] Embodiment 1

[0078] See also Figure 1The present invention provides a technical solution: a music recommendation method based on nursing data, comprising the following specific steps:

[0079] Step S1: obtaining a nursing monitoring cycle, dividing the nursing monitoring cycle into multiple music recommendation time periods, respectively obtaining a first music recommendation coefficient and a second music recommendation coefficient corresponding to each music recommendation time period, and obtaining patient nursing monitoring data;

[0080] Step S11: marking the period of time during which the patient receives nursing treatment every day as a nursing monitoring cycle, and dividing the nursing monitoring cycle into a number of music recommendation periods of equal length;

[0081] Step S12: randomly selecting a time period from the obtained multiple music recommendation time periods as a sample music recommendation time period;

[0082] Step S13: monitoring physiological indicators of patients in the sample music recommendation period to obtain a first music recommendation coefficient;

[0083] The step S13 further includes the following specific steps:

[0084] Step S131: Mark the start time point of the sample music recommendation period as the first period characteristic time point, mark the time point corresponding to a physiological characteristic duration after the first period characteristic time point as the second period characteristic time point, and mark the time period between the first period characteristic time point and the second period characteristic time point as the physiological monitoring sub-period of the sample music recommendation period;

[0085] Step S132: randomly selecting a number of physiological monitoring time points in the physiological monitoring sub-period, and naming them as the first physiological monitoring time point to the ath physiological monitoring time point;

[0086] Step S133: respectively obtaining the diastolic blood pressure values ​​of the patient corresponding to the first physiological monitoring time point to the ath physiological monitoring time point, obtaining the first diastolic pressure value to the ath diastolic pressure value, and averaging the first diastolic pressure value to the ath diastolic pressure value to obtain the average diastolic pressure of the time period;

[0087] Step S134: respectively obtaining the systolic blood pressure values ​​of the patient corresponding to the first physiological monitoring time point to the ath physiological monitoring time point, obtaining the first systolic blood pressure value to the ath systolic blood pressure value, and averaging the first systolic blood pressure value to the ath systolic blood pressure value to obtain the average systolic blood pressure of the time period;

[0088] Step S135: performing heart rate analysis on the patients in the physiological monitoring sub-periods respectively to obtain heart rate index values ​​for the period;

[0089] The step S135 further includes the following specific steps:

[0090] Step S1351: respectively obtaining the real-time heart rate values ​​of the patient corresponding to the first physiological monitoring time point to the ath physiological monitoring time point, obtaining the first heart rate value to the ath heart rate value, and averaging the first heart rate value to the ath heart rate value to obtain the time period average heart rate value;

[0091] Step S1352: Compare the values ​​of the first heart rate value to the ath heart rate value, and mark the heart rate value with the largest value as the peak heart rate value of the period;

[0092] Step S1353: Calculate the variance of the first heart rate value to the ath heart rate value to obtain the heart rate variance of the time period monitoring;

[0093] Step S1354: Calculate the average heart rate value of the time period, the peak heart rate value of the time period, and the variance of the monitored heart rate of the time period to obtain a heart rate index value of the time period;

[0094] The specific formula for calculating the heart rate index value of the time period is as follows:

[0095]

[0096] Among them, Zxl is the heart rate index value of the period, Xpl is the average heart rate value of the period, Xfc is the variance of the monitored heart rate of the period, and Fzc is the peak heart rate value of the period;

[0097] Step S136: Calculate the average diastolic pressure, average systolic pressure and heart rate index values ​​of the time period to obtain a first music recommendation coefficient corresponding to the sample music recommendation time period;

[0098] The first music recommendation coefficient is calculated, and the specific formula is as follows:

[0099]

[0100] Among them, Ytl1 is the first music recommendation coefficient corresponding to the sample music recommendation period, Zxl is the heart rate index value of the period, Ssy is the average diastolic blood pressure of the period, and Szy is the average systolic blood pressure of the period;

[0101] Step S14: performing skin electricity monitoring on the patient in the sample music recommendation period to obtain a second music recommendation coefficient;

[0102] The step S14 further includes the following specific steps:

[0103] Step S141: Mark the start time point of the sample music recommendation period as the first period characteristic time point, mark the time point corresponding to the emotion monitoring duration after the first period characteristic time point as the third period characteristic time point, and mark the time period between the first period characteristic time point and the third period characteristic time point as the skin electricity monitoring sub-period of the sample music recommendation period;

[0104] Step S142: In the skin electrical monitoring sub-period, a plurality of consecutive time points with equal time intervals are selected as emotion monitoring time points, and the obtained plurality of emotion monitoring time points are named as the first emotion monitoring time point to the bth emotion monitoring time point respectively;

[0105] Step S143: obtaining the skin conductivities of the patient corresponding to the first emotion monitoring time point to the bth emotion monitoring time point, and obtaining the first skin conductivities to the bth skin conductivities;

[0106] Step S144: obtaining a skin conductivity change value by calculating the first skin conductivity to the bth skin conductivity;

[0107] The skin conductivity change value is calculated, and the specific formula is as follows:

[0108] Pdd=(Dpl2-Dpl1)+(Dpl3-Dpl2)+...+(Dplb-Dpl(b-1));

[0109] Wherein, Pdd is the skin conductivity change value, Dpl1 to Dplb are the first skin conductivity to the bth skin conductivity respectively;

[0110] Step S145: Calculate the average of the first skin conductivity to the bth skin conductivity to obtain a monitored average conductivity;

[0111] Step S146: comparing the values ​​of the first skin conductivity to the bth skin conductivity, and marking the skin conductivity with the largest value as the monitored peak skin conductivity;

[0112] Step S147: Calculate the skin conductivity change value, the monitored average conductivity, and the monitored peak skin conductivity to obtain a second music recommendation coefficient;

[0113] The second music recommendation coefficient is calculated, and the specific formula is as follows:

[0114] Ytl2=Pdd+Jpd+Fzd;

[0115] Wherein, Ytl2 is the second music recommendation coefficient, Pdd is the skin conductivity change value, Jpd is the monitored average conductivity, and Fzd is the monitored peak skin conductivity;

[0116] Step S15: Repeat step S13 and step S14 to obtain the first music recommendation coefficient and the second music recommendation coefficient corresponding to each music recommendation period, respectively, to obtain patient care monitoring data;

[0117] Step S2: Obtaining preliminary music recommendation coefficients according to patient care monitoring data to divide multiple sample music recommendation periods into a first type of music recommendation period and a second type of music recommendation period, obtaining preliminary music recommendation classification data, and performing facial expression analysis on patients corresponding to the second type of music recommendation period to obtain a characteristic facial expression image appearance ratio, and defining the preliminary music recommendation classification data and the characteristic facial expression image appearance ratio as patient care analysis data;

[0118] Step S21: Obtain patient care monitoring data, and obtain a first music recommendation coefficient and a second music recommendation coefficient corresponding to each music recommendation period according to the patient care monitoring data;

[0119] Step S22: Calculating the first music recommendation coefficient and the second music recommendation coefficient corresponding to each music recommendation period to obtain a plurality of preliminary music recommendation coefficients;

[0120] The preliminary music recommendation coefficient is calculated, and the specific formula is as follows:

[0121] Yyx=Ytl1+Ytl2;

[0122] Among them, Yyx is the preliminary music recommendation coefficient, Ytl1 is the first music recommendation coefficient, and Ytl2 is the second music recommendation coefficient;

[0123] Step S23: obtaining a music preliminary recommendation coefficient threshold, comparing multiple music preliminary recommendation coefficients with the music preliminary recommendation coefficient threshold respectively, and dividing multiple music recommendation periods into a first type of music recommendation period and a second type of music recommendation period according to the comparison result, to obtain music recommendation preliminary classification data;

[0124] The step S23 further includes the following specific steps:

[0125] Step S231: respectively obtaining a first music recommendation coefficient threshold and a second music recommendation coefficient threshold, and calculating the first music recommendation coefficient threshold and the second music recommendation coefficient threshold to obtain a preliminary music recommendation coefficient threshold;

[0126] Step S232: when the music preliminary recommendation coefficient is greater than or equal to the music preliminary recommendation coefficient threshold, the corresponding music recommendation period is divided into a first type of music recommendation period;

[0127] Step S233: when the music preliminary recommendation coefficient is less than the music preliminary recommendation coefficient threshold, the corresponding music recommendation period is divided into a second type of music recommendation period;

[0128] Step S24: performing expression analysis on the second type of music recommendation period to obtain a characteristic expression image appearance ratio;

[0129] The step S24 further includes the following specific steps:

[0130] Step S241: randomly selecting a time period in the music recommendation time period corresponding to the second type of music recommendation time period as an expression recognition sample time period;

[0131] Step S242: marking the start time point of the expression recognition sample period as the first image feature time point, marking the time point corresponding to an image monitoring duration after the first image feature time point as the second image feature time point, marking the time period between the first image feature time point and the second image feature time point as the expression monitoring sub-period of the expression recognition sample period, and randomly acquiring facial expression images of a number of patients within the expression monitoring sub-period to obtain a plurality of patient expression images;

[0132] Step S243: Counting the number of randomly acquired facial expression images to obtain the total number of facial expression images;

[0133] Step S244: randomly crawling a plurality of human face images from the Internet of Things by using crawler technology, and annotating the obtained plurality of human face images into first expression type images and second expression type images by manual annotation;

[0134] Step S245: dividing the plurality of human face images into a model test set, a model training set, and a model verification set, establishing an image recognition model by using an existing convolutional neural network model technology, training the image recognition model by using the model test set, the model training set, and the model verification set, and testing the trained image recognition model, and naming the tested image recognition model as a feature expression recognition model;

[0135] Step S246: using a characteristic expression recognition model to divide a plurality of patient expression images into a first expression type image and a second expression type image, and obtaining the number of patient expression images corresponding to the first expression type image to obtain a characteristic expression image quantity value, and calculating a ratio of the characteristic expression image quantity value to the total number of facial expression images to obtain a characteristic expression image appearance ratio;

[0136] Step S25: defining the music recommendation preliminary classification data and the characteristic expression image appearance ratio as patient care analysis data;

[0137] Step S3: Recommending music to patients in different music recommendation time periods according to the patient care analysis data;

[0138] Step S31: Obtain patient care analysis data, and obtain music recommendation preliminary classification data and characteristic expression image appearance ratio according to the patient care analysis data;

[0139] Step S32: when the patient is in the first type of music recommendation period, calm music is recommended to the patient;

[0140] Step S33: when the patient is in the second type of music recommendation period, obtaining a characteristic expression image appearance ratio threshold and performing numerical comparison with the characteristic expression image appearance ratio;

[0141] Step S34: when the characteristic expression image appearance ratio is greater than or equal to the characteristic expression image appearance ratio threshold, calming music is recommended to the patient in the second type of music recommendation period;

[0142] Step S35: when the characteristic expression image appearance ratio is less than the characteristic expression image appearance ratio threshold, dynamic music is recommended to the patient in the second type of music recommendation period;

[0143] In this application, if corresponding calculation formulas appear, the above calculation formulas are all dimensionless and take their numerical calculations. The weight coefficients, proportional coefficients and other coefficients in the formulas are set to a result value obtained by quantifying each parameter. The size of the weight coefficient and the proportional coefficient can be determined as long as it does not affect the proportional relationship between the parameter and the result value.

[0144] Embodiment 2

[0145] See also Figure 2 , based on another concept of the same invention, a music recommendation system based on nursing data is now proposed, including a data acquisition module, an expression analysis module, a music recommendation module and a server, wherein the data acquisition module, the expression analysis module and the music recommendation module are respectively connected to the server, and the server controls the data acquisition module, the expression analysis module and the music recommendation module respectively;

[0146] The data acquisition module acquires a nursing monitoring cycle, divides the nursing monitoring cycle into multiple music recommendation time periods, respectively acquires a first music recommendation coefficient and a second music recommendation coefficient corresponding to each music recommendation time period, and obtains patient nursing monitoring data;

[0147] The data acquisition module includes a physiological monitoring unit and a skin electricity unit;

[0148] Mark the period of time when the patient receives nursing treatment every day as a nursing monitoring cycle, and divide the nursing monitoring cycle into several music recommendation periods of equal length;

[0149] Randomly select a time period from the obtained multiple music recommendation time periods as a sample music recommendation time period;

[0150] The physiological monitoring unit monitors the physiological indexes of the patient in the sample music recommendation period to obtain a first music recommendation coefficient;

[0151] The details are as follows:

[0152] Mark the start time point of the sample music recommendation period as the first period characteristic time point, mark the time point corresponding to a physiological characteristic duration after the first period characteristic time point as the second period characteristic time point, and mark the time period between the first period characteristic time point and the second period characteristic time point as the physiological monitoring sub-period of the sample music recommendation period;

[0153] In the physiological monitoring sub-period, a number of physiological monitoring time points are randomly selected and named as the first physiological monitoring time point to the ath physiological monitoring time point;

[0154] It should be noted here that:

[0155] In this application, a is a quantitative value corresponding to a physiological monitoring time point, and a is an integer greater than 0;

[0156] Respectively obtain the diastolic blood pressure values ​​of the patient corresponding to the first physiological monitoring time point to the ath physiological monitoring time point, obtain the first diastolic pressure value to the ath diastolic pressure value, and average the first diastolic pressure value to the ath diastolic pressure value to obtain the average diastolic pressure of the time period;

[0157] Respectively obtain the systolic blood pressure values ​​of the patient corresponding to the first physiological monitoring time point to the ath physiological monitoring time point, obtain the first systolic blood pressure value to the ath systolic blood pressure value, and average the first systolic blood pressure value to the ath systolic blood pressure value to obtain the average systolic blood pressure of the time period;

[0158] Perform heart rate analysis on patients in physiological monitoring sub-periods respectively to obtain heart rate index values ​​for the period;

[0159] The details are as follows:

[0160] Respectively obtain the real-time heart rate values ​​of the patient corresponding to the first physiological monitoring time point to the ath physiological monitoring time point, obtain the first heart rate value to the ath heart rate value, and average the first heart rate value to the ath heart rate value to obtain the average heart rate value of the time period;

[0161] Compare the values ​​of the first heart rate value to the ath heart rate value, and mark the heart rate value with the largest value as the peak heart rate value of the period;

[0162] Calculate the variance of the first heart rate value to the ath heart rate value to obtain the heart rate variance of the time period monitoring;

[0163] The average heart rate value of the period, the peak heart rate value of the period and the variance of the monitored heart rate of the period are calculated to obtain the heart rate index value of the period;

[0164] The specific formula for calculating the heart rate index value of the time period is as follows:

[0165]

[0166] Among them, Zxl is the heart rate index value of the period, Xpl is the average heart rate value of the period, Xfc is the variance of the monitored heart rate of the period, and Fzc is the peak heart rate value of the period;

[0167] The average diastolic blood pressure, the average systolic blood pressure and the heart rate index values ​​of the time period are calculated to obtain the first music recommendation coefficient corresponding to the sample music recommendation time period;

[0168] The first music recommendation coefficient is calculated, and the specific formula is as follows:

[0169]

[0170] Among them, Ytl1 is the first music recommendation coefficient corresponding to the sample music recommendation period, Zxl is the heart rate index value of the period, Ssy is the average diastolic blood pressure of the period, and Szy is the average systolic blood pressure of the period;

[0171] The skin electricity unit performs skin electricity monitoring on the patient in the sample music recommendation period to obtain a second music recommendation coefficient;

[0172] The details are as follows:

[0173] The start time point of the sample music recommendation period is marked as the first period characteristic time point, the time point corresponding to the emotion monitoring duration after the first period characteristic time point is marked as the third period characteristic time point, and the time period between the first period characteristic time point and the third period characteristic time point is marked as the skin electricity monitoring sub-period of the sample music recommendation period;

[0174] In the skin electrical monitoring sub-period, a number of consecutive time points with equal time intervals are selected as emotion monitoring time points, and the obtained multiple emotion monitoring time points are named as the first emotion monitoring time point to the bth emotion monitoring time point respectively;

[0175] It should be noted here that:

[0176] In this application, b is the quantity value corresponding to the emotion monitoring time point, and b is an integer greater than 0;

[0177] Obtain the skin conductivities of the patient corresponding to the first emotion monitoring time point to the bth emotion monitoring time point, and obtain the first skin conductivities to the bth skin conductivities;

[0178] The skin conductivity change value is obtained by calculating the first skin conductivity to the bth skin conductivity;

[0179] The skin conductivity change value is calculated, and the specific formula is as follows:

[0180] Pdd=(Dpl2-Dpl1)+(Dpl3-Dpl2)+...+(Dplb-Dpl(b-1));

[0181] Wherein, Pdd is the skin conductivity change value, Dpl1 to Dplb are the first skin conductivity to the bth skin conductivity respectively;

[0182] Calculate the average of the first skin conductivity to the bth skin conductivity to obtain the monitored average conductivity;

[0183] Compare the values ​​of the first skin conductivity to the bth skin conductivity, and mark the skin conductivity with the largest value as the monitored peak skin conductivity;

[0184] The skin conductivity change value, the monitored average conductivity, and the monitored peak skin conductivity are calculated to obtain a second music recommendation coefficient;

[0185] The second music recommendation coefficient is calculated, and the specific formula is as follows:

[0186] Ytl2=Pdd+Jpd+Fzd;

[0187] Wherein, Ytl2 is the second music recommendation coefficient, Pdd is the skin conductivity change value, Jpd is the monitored average conductivity, and Fzd is the monitored peak skin conductivity;

[0188] Repeat the process of obtaining the first music recommendation coefficient and the second music recommendation coefficient corresponding to the sample music recommendation period, respectively obtain the first music recommendation coefficient and the second music recommendation coefficient corresponding to each music recommendation period, and obtain patient care monitoring data;

[0189] The data acquisition module acquires the patient care monitoring data and transmits it to the data analysis module;

[0190] The data analysis module obtains the preliminary music recommendation coefficient according to the patient care monitoring data, divides the multiple sample music recommendation periods into the first type of music recommendation period and the second type of music recommendation period, obtains the preliminary music recommendation classification data, and performs expression analysis on the patients corresponding to the second type of music recommendation period to obtain the characteristic expression image appearance ratio, and defines the preliminary music recommendation classification data and the characteristic expression image appearance ratio as the patient care analysis data;

[0191] Acquire patient care monitoring data, and acquire a first music recommendation coefficient and a second music recommendation coefficient corresponding to each music recommendation period according to the patient care monitoring data;

[0192] The first music recommendation coefficient and the second music recommendation coefficient corresponding to each music recommendation period are calculated to obtain a plurality of preliminary music recommendation coefficients;

[0193] The preliminary music recommendation coefficient is calculated, and the specific formula is as follows:

[0194] Yyx=Ytl1+Ytl2;

[0195] Among them, Yyx is the preliminary music recommendation coefficient, Ytl1 is the first music recommendation coefficient, and Ytl2 is the second music recommendation coefficient;

[0196] Obtaining a music preliminary recommendation coefficient threshold, numerically comparing multiple music preliminary recommendation coefficients with the music preliminary recommendation coefficient threshold, and dividing multiple music recommendation periods into a first type of music recommendation period and a second type of music recommendation period according to the comparison results, to obtain music recommendation preliminary classification data;

[0197] The threshold of the preliminary music recommendation coefficient is obtained as follows:

[0198] Obtain a first music recommendation coefficient threshold and a second music recommendation coefficient threshold respectively, and calculate the first music recommendation coefficient threshold and the second music recommendation coefficient threshold to obtain a music preliminary recommendation coefficient threshold;

[0199] The threshold of the preliminary music recommendation coefficient is calculated, and the specific formula is as follows:

[0200] Yyxy=Ytly1+Ytly2;

[0201] Among them, Yyxy is the preliminary music recommendation coefficient threshold, Ytly1 is the first music recommendation coefficient threshold, and Ytly2 is the second music recommendation coefficient threshold;

[0202] It should be noted here that:

[0203] The first music recommendation coefficient threshold and the second music recommendation coefficient threshold involved here are respectively the minimum first music recommendation coefficient and the minimum second music recommendation coefficient corresponding to the first type of music recommendation period;

[0204] When the music preliminary recommendation coefficient is greater than or equal to the music preliminary recommendation coefficient threshold, the corresponding music recommendation period is divided into a first type of music recommendation period;

[0205] When the music preliminary recommendation coefficient is less than the music preliminary recommendation coefficient threshold, the corresponding music recommendation period is divided into a second type of music recommendation period;

[0206] Performing expression analysis on the second type of music recommendation period to obtain the appearance ratio of characteristic expression images;

[0207] The details are as follows:

[0208] Randomly selecting a time period in the music recommendation time period corresponding to the second type of music recommendation time period as an expression recognition sample time period;

[0209] Marking the start time point of the expression recognition sample period as a first image feature time point, marking a time point corresponding to an image monitoring duration after the first image feature time point as a second image feature time point, marking the time period between the first image feature time point and the second image feature time point as an expression monitoring sub-period of the expression recognition sample period, and randomly acquiring facial expression images of a number of patients within the expression monitoring sub-period to obtain a plurality of patient expression images;

[0210] Counting the randomly acquired facial expression images to obtain the total number of facial expression images;

[0211] Randomly crawling a plurality of human face images from the Internet of Things by using crawler technology, and annotating the obtained plurality of human face images into first expression type images and second expression type images by manually annotating;

[0212] It should be noted here that:

[0213] In the present application, the expression corresponding to the first expression type image is specifically a high-intensity type expression, including but not limited to anger, excitement and anxiety, and the expression corresponding to the second expression type image is specifically a low-intensity type expression, including but not limited to pleasure, calmness and boredom;

[0214] Divide multiple human face images into model test sets, model training sets, and model validation sets. Figure 3 , establish an image recognition model through the existing convolutional neural network technology, use the model test set, model training set and model verification set to train the image recognition model, and test the trained image recognition model, and name the tested image recognition model as a feature expression recognition model;

[0215] Using a characteristic expression recognition model, a plurality of patient expression images are divided into a first expression type image and a second expression type image, and the number of patient expression images corresponding to the first expression type image is obtained to obtain a characteristic expression image quantity value, and a ratio of the characteristic expression image quantity value to the total number of facial expression images is calculated to obtain a characteristic expression image appearance ratio;

[0216] The music recommendation preliminary classification data and the characteristic expression image appearance ratio were defined as patient care analysis data;

[0217] The data analysis module acquires the patient care analysis data and transmits it to the music recommendation module;

[0218] The music recommendation module recommends music to patients in different music recommendation periods based on patient care analysis data;

[0219] Acquire patient care analysis data, and acquire music recommendation preliminary classification data and characteristic expression image appearance ratio according to the patient care analysis data;

[0220] When the patient is in the first type of music recommendation period, calm music is recommended to the patient;

[0221] When the patient is in the second type of music recommendation period, obtaining a characteristic expression image appearance ratio threshold and comparing the characteristic expression image appearance ratio with each other;

[0222] When the characteristic expression image appearance ratio is greater than or equal to the characteristic expression image appearance ratio threshold, calm type music is recommended to the patient in the second type music recommendation period;

[0223] The characteristic expression image appearance ratio threshold involved here is the minimum characteristic expression image appearance ratio corresponding to the first type of music recommendation period;

[0224] When the characteristic expression image appearance ratio is less than the characteristic expression image appearance ratio threshold, dynamic music is recommended to the patient in the second type of music recommendation period;

[0225] It should be noted here that:

[0226] In the present application, calm music includes but is not limited to simulated nature music, classical music and soothing piano music, and dynamic music includes but is not limited to rock, hip-hop rap and pop music.

[0227] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A music recommendation method based on nursing data, characterized in that: The specific steps include: Step S1: obtaining a nursing monitoring cycle, dividing the nursing monitoring cycle into multiple music recommendation time periods, respectively obtaining a first music recommendation coefficient and a second music recommendation coefficient corresponding to each music recommendation time period, and obtaining patient nursing monitoring data; Step S2: obtaining a preliminary music recommendation coefficient according to the patient care monitoring data, dividing a plurality of sample music recommendation time periods into a first type of music recommendation time period and a second type of music recommendation time period, obtaining preliminary music recommendation classification data, and performing facial expression analysis on patients corresponding to the second type of music recommendation time period to obtain a characteristic facial expression image appearance ratio, and defining the preliminary music recommendation classification data and the characteristic facial expression image appearance ratio as patient care analysis data; Step S3: Recommend music to patients in different music recommendation time periods based on the patient care analysis data.

2. A music recommendation method based on nursing data according to claim 1, characterized in that: The step S1 further includes the following specific steps: Step S11: marking the period of time during which the patient receives nursing treatment every day as a nursing monitoring cycle, and dividing the nursing monitoring cycle into a number of music recommendation periods of equal length; Step S12: randomly selecting a time period from multiple music recommendation time periods as a sample music recommendation time period; Step S13: monitoring physiological indicators of patients in the sample music recommendation period to obtain a first music recommendation coefficient; Step S14: performing skin electricity monitoring on the patient in the sample music recommendation period to obtain a second music recommendation coefficient; Step S15: Repeat step S13 and step S14 to obtain the first music recommendation coefficient and the second music recommendation coefficient corresponding to each music recommendation period, respectively, to obtain patient care monitoring data.

3. A music recommendation method based on nursing data according to claim 2, characterized in that: The step S13 further includes the following specific steps: Step S131: Mark the start time point of the sample music recommendation period as the first period characteristic time point, mark the time point corresponding to a physiological characteristic duration after the first period characteristic time point as the second period characteristic time point, and mark the time period between the first period characteristic time point and the second period characteristic time point as the physiological monitoring sub-period; Step S132: randomly selecting a number of physiological monitoring time points in the physiological monitoring sub-period, and naming them as the first physiological monitoring time point to the ath physiological monitoring time point; Step S133: respectively obtaining the diastolic blood pressure values ​​of the patient corresponding to the first physiological monitoring time point to the ath physiological monitoring time point, obtaining the first diastolic pressure value to the ath diastolic pressure value, and averaging the first diastolic pressure value to the ath diastolic pressure value to obtain the average diastolic pressure of the time period; Step S134: respectively obtaining the systolic blood pressure values ​​of the patient corresponding to the first physiological monitoring time point to the ath physiological monitoring time point, obtaining the first systolic blood pressure value to the ath systolic blood pressure value, and averaging the first systolic blood pressure value to the ath systolic blood pressure value to obtain the average systolic blood pressure of the time period; Step S135: performing heart rate analysis on the patients in the physiological monitoring sub-periods respectively to obtain heart rate index values ​​for the period; Step S136: Calculate the average diastolic pressure, average systolic pressure and heart rate index values ​​of the time period to obtain a first music recommendation coefficient corresponding to the sample music recommendation time period; A first music recommendation coefficient is calculated.

4. The method for music recommendation based on nursing data according to claim 3, characterized in that: The step S135 further includes the following specific steps: Step S1351: respectively obtaining the real-time heart rate values ​​of the patient corresponding to the first physiological monitoring time point to the ath physiological monitoring time point, obtaining the first heart rate value to the ath heart rate value, and averaging the first heart rate value to the ath heart rate value to obtain the time period average heart rate value; Step S1352: Compare the values ​​of the first heart rate value to the ath heart rate value, and mark the heart rate value with the largest value as the peak heart rate value of the period; Step S1353: Calculate the variance of the first heart rate value to the ath heart rate value to obtain the heart rate variance of the time period monitoring; Step S1354: Calculate the average heart rate value of the time period, the peak heart rate value of the time period, and the variance of the monitored heart rate of the time period to obtain a heart rate index value of the time period; Calculate the heart rate index value for each period.

5. The method for music recommendation based on nursing data according to claim 1, characterized in that: The step S14 further includes the following specific steps: Step S141: Mark the start time point of the sample music recommendation period as the first period characteristic time point, mark the time point corresponding to the emotion monitoring duration after the first period characteristic time point as the third period characteristic time point, and mark the time period between the first period characteristic time point and the third period characteristic time point as the skin electricity monitoring sub-period; Step S142: In the skin electrical monitoring sub-period, a plurality of consecutive time points with equal time intervals are selected as emotion monitoring time points, and the obtained plurality of emotion monitoring time points are named as the first emotion monitoring time point to the bth emotion monitoring time point respectively; Step S143: obtaining the skin conductivities of the patient corresponding to the first emotion monitoring time point to the bth emotion monitoring time point, and obtaining the first skin conductivities to the bth skin conductivities; Step S144: obtaining a skin conductivity change value by calculating the first skin conductivity to the bth skin conductivity; Calculate the change value of skin conductivity; Step S145: Calculate the average of the first skin conductivity to the bth skin conductivity to obtain a monitored average conductivity; Step S146: comparing the values ​​of the first skin conductivity to the bth skin conductivity, and marking the skin conductivity with the largest value as the monitored peak skin conductivity; Step S147: Calculate the skin conductivity change value, the monitored average conductivity, and the monitored peak skin conductivity to obtain a second music recommendation coefficient; The second music recommendation coefficient is calculated.

6. The method for music recommendation based on nursing data according to claim 1, characterized in that: The step S2 further includes the following specific steps: Step S21: Obtain patient care monitoring data, and obtain a first music recommendation coefficient and a second music recommendation coefficient corresponding to each music recommendation period according to the patient care monitoring data; Step S22: Calculating the first music recommendation coefficient and the second music recommendation coefficient corresponding to each music recommendation period to obtain a plurality of preliminary music recommendation coefficients; Calculate the preliminary music recommendation coefficient; Step S23: obtaining a music preliminary recommendation coefficient threshold, comparing multiple music preliminary recommendation coefficients with the music preliminary recommendation coefficient threshold respectively, and dividing multiple music recommendation periods into a first type of music recommendation period and a second type of music recommendation period according to the comparison result, to obtain music recommendation preliminary classification data; Step S24: performing expression analysis on the second type of music recommendation period to obtain a characteristic expression image appearance ratio; Step S25: Define the music recommendation preliminary classification data and the characteristic expression image appearance ratio as patient care analysis data.

7. A music recommendation method based on nursing data according to claim 6, characterized in that: The step S23 further includes the following specific steps: Step S231: respectively obtaining a first music recommendation coefficient threshold and a second music recommendation coefficient threshold, and calculating the first music recommendation coefficient threshold and the second music recommendation coefficient threshold to obtain a preliminary music recommendation coefficient threshold; Step S232: when the music preliminary recommendation coefficient is greater than or equal to the music preliminary recommendation coefficient threshold, the corresponding music recommendation period is divided into a first type of music recommendation period; Step S233: When the music preliminary recommendation coefficient is less than the music preliminary recommendation coefficient threshold, the corresponding music recommendation period is divided into a second type of music recommendation period.

8. The method for music recommendation based on nursing data according to claim 6, characterized in that: The step S24 further includes the following specific steps: Step S241: randomly selecting a time period in the music recommendation time period corresponding to the second type of music recommendation time period as an expression recognition sample time period; Step S242: marking the start time point of the expression recognition sample period as the first image feature time point, marking the time point corresponding to the image monitoring duration after the first image feature time point as the second image feature time point, marking the time period between the first image feature time point and the second image feature time point as the expression monitoring sub-period, and randomly acquiring facial expression images of several patients within the expression monitoring sub-period to obtain multiple patient expression images; Step S243: Counting the number of randomly acquired facial expression images to obtain the total number of facial expression images; Step S244: randomly crawling a plurality of human face images from the Internet of Things by using crawler technology, and annotating the obtained plurality of human face images into first expression type images and second expression type images by manual annotation; Step S245: dividing the plurality of human face images into a model test set, a model training set, and a model verification set, establishing an image recognition model by using an existing convolutional neural network model technology, training the image recognition model by using the model test set, the model training set, and the model verification set, and testing the trained image recognition model, and naming the tested image recognition model as a feature expression recognition model; Step S246: using the characteristic expression recognition model to divide the plurality of patient expression images into a first expression type image and a second expression type image, and obtaining the number of patient expression images corresponding to the first expression type image to obtain a characteristic expression image quantity value, and calculating the ratio of the characteristic expression image quantity value to the total number of facial expression images, Get the appearance ratio of characteristic expression images.

9. The method for music recommendation based on nursing data according to claim 1, characterized in that: The step S3 further includes the following specific steps: Step S31: Obtain patient care analysis data, and obtain music recommendation preliminary classification data and characteristic expression image appearance ratio according to the patient care analysis data; Step S32: when the patient is in the first type of music recommendation period, calm music is recommended to the patient; Step S33: when the patient is in the second type of music recommendation period, obtaining a characteristic expression image appearance ratio threshold and performing numerical comparison with the characteristic expression image appearance ratio; Step S34: when the characteristic expression image appearance ratio is greater than or equal to the characteristic expression image appearance ratio threshold, calming music is recommended to the patient in the second type of music recommendation period; Step S35: When the characteristic expression image appearance ratio is less than the characteristic expression image appearance ratio threshold, dynamic music is recommended to the patient in the second type of music recommendation period.

10. A music recommendation system based on nursing data, applicable to a music recommendation method based on nursing data as claimed in any one of claims 1 to 9, characterized in that: The specific working process of each module of the music recommendation system is as follows: Data acquisition module: used to obtain a nursing monitoring cycle, divide the nursing monitoring cycle into multiple music recommendation time periods, respectively obtain a first music recommendation coefficient and a second music recommendation coefficient corresponding to each music recommendation time period, and obtain patient nursing monitoring data; Data analysis module: used to obtain preliminary music recommendation coefficients according to patient care monitoring data, divide multiple sample music recommendation periods into first-type music recommendation periods and second-type music recommendation periods, obtain preliminary music recommendation classification data, and perform facial expression analysis on patients corresponding to the second-type music recommendation periods to obtain characteristic facial expression image appearance ratios, and define the preliminary music recommendation classification data and characteristic facial expression image appearance ratios as patient care analysis data; Music recommendation module: used to make music recommendations to patients in different music recommendation periods based on patient care analysis data.

Images