Fetus sound monitoring and artificial intelligence auxiliary diagnosis method and system

Abstract

The invention discloses a fetal sound monitoring and artificial intelligence auxiliary diagnosis method and system, relates to the field of medical treatment, and solves the problem of poor diagnosis effect of the existing fetal sound monitoring and artificial intelligence auxiliary diagnosis method.The fetal sound monitoring and artificial intelligence auxiliary diagnosis method comprises the steps that S1, a fetal image recognition model is used for recognizing a maternal fetal medical image, the method comprises the following steps: S1, identifying a target fetus, and dividing the target fetus into a first type of diagnosed fetus and a second type of diagnosed fetus according to an identification result to obtain preliminary diagnosis data of the fetus, S2, respectively obtaining a periodic fetal heart average beating frequency, a fetal heart acceleration monitoring coefficient and a fetal heart deceleration monitoring coefficient to obtain fetal sound monitoring data, and S3, performing auxiliary diagnosis on the target fetus according to the preliminary diagnosis data of the fetus and the fetal sound monitoring data, and performing early warning according to the diagnosis result, so that the pertinence and accuracy of the fetal sound auxiliary diagnosis result can be improved.

Description

Technical Field

[0001] The present invention belongs to the medical field and relates to artificial intelligence technology, specifically a fetal sound monitoring and artificial intelligence-assisted diagnosis method and system. Background Art

[0002] The existing fetal sound monitoring and artificial intelligence-assisted diagnosis methods have the following specific defects when performing medical auxiliary diagnosis on the fetus:

[0003] 1. The existing fetal sound monitoring methods do not classify the monitored fetuses according to medical images, and do not develop targeted monitoring methods for specific fetuses, resulting in a lack of targeting in the monitoring process;

[0004] 2. The existing fetal tone monitoring method can only provide a single fetal tone data, and cannot further analyze the fetal heart monitoring data, thus failing to provide comprehensive auxiliary diagnosis data for medical staff, resulting in poor auxiliary diagnosis effect.

[0005] To this end, we propose a fetal sound monitoring and artificial intelligence-assisted diagnosis method and system. Summary of the invention

[0006] In view of the shortcomings of the prior art, the present invention aims to provide a fetal sound monitoring and artificial intelligence-assisted diagnosis method and system, and the present invention aims to improve the comprehensiveness and pertinence of the fetal sound-assisted diagnosis method;

[0007] In order to achieve the above object, the present invention adopts the following technical solution: a fetal sound monitoring and artificial intelligence assisted diagnosis method, comprising the following specific steps:

[0008] Step S1: obtaining maternal-fetal medical images and multiple fetal medical images, creating a fetal image recognition model through the fetal medical images, using the fetal image recognition model to recognize the maternal-fetal medical images, and classifying the target fetus into a first type of diagnosis fetus and a second type of diagnosis fetus according to the recognition results, to obtain preliminary fetal diagnosis data;

[0009] Step S2: obtaining a fetal heart monitoring cycle, monitoring the number of fetal heart beats of the second type of diagnosed fetus in the fetal heart monitoring cycle, obtaining an average number of fetal heart beats in the cycle, monitoring the fetal heart acceleration rate of the second type of diagnosed fetus in the fetal heart monitoring cycle, obtaining a fetal heart acceleration monitoring coefficient, monitoring the fetal heart deceleration rate of the second type of diagnosed fetus in the fetal heart monitoring cycle, obtaining a fetal heart deceleration monitoring coefficient, and defining the average number of fetal heart beats in the cycle, the fetal heart acceleration monitoring coefficient, and the fetal heart deceleration monitoring coefficient as fetal sound monitoring data;

[0010] Step S3: Perform auxiliary diagnosis on the target fetus based on the preliminary fetal diagnosis data and fetal sound monitoring data, and issue an early warning based on the diagnosis results.

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

[0012] Step S11: obtaining a medical image of the target fetus in the pregnant woman's belly through a medical detection device to obtain a medical image of the mother and the fetus;

[0013] Step S12: Acquire the gestational month of the target fetus to obtain the gestational month of the target fetus;

[0014] Step S13: creating a fetal image recognition model;

[0015] Step S14: Use the fetal image recognition model to recognize maternal and fetal medical images, and classify the target fetus into a first type of diagnosis fetus and a second type of diagnosis fetus according to the recognition result to obtain preliminary fetal diagnosis data.

[0016] Furthermore, the step S13 further includes the following specific steps:

[0017] Step S131: acquiring multiple fetal medical images of the same gestational month as the target fetus, and dividing the multiple fetal medical images into first-type medical images and second-type medical images by manual identification and manual marking to obtain fetal medical image marking data;

[0018] Step S132: dividing the fetal image labeled data into a fetal image training set and a fetal image test set according to the image training test ratio;

[0019] Step S133: creating an image recognition model through an existing artificial intelligence platform, and training the image recognition model using a fetal image training set, until each medical fetal image in the fetal image training set has trained the image recognition model once;

[0020] Step S134: Use the fetal image test set to test the image recognition model and obtain the recognition accuracy. When the recognition accuracy is greater than or equal to the target recognition accuracy, the image recognition model training is completed and the fetal image recognition model is obtained. When the recognition accuracy is less than the target recognition accuracy, continue to use the fetal image training set to train the image recognition model until the recognition accuracy is greater than or equal to the target recognition accuracy.

[0021] Furthermore, the step S2 further includes the following specific steps:

[0022] Step S21: during the process of fetal heart rate monitoring of the target fetus, marking a fetal heart rate monitoring cycle;

[0023] Step S22: dividing the fetal heart monitoring cycle into a number of fetal heart monitoring time periods of equal duration, and naming the fetal heart monitoring time periods as the first fetal heart monitoring time period to the tth fetal heart monitoring time period in chronological order;

[0024] Step S23: acquiring the fetal heart beats of the target fetus corresponding to the first fetal heart monitoring period to the t-th fetal heart monitoring period, and obtaining the fetal heart beats from the first period to the t-th period;

[0025] Step S24: Calculate the average of the fetal heart beats from the first period to the tth period to obtain the average fetal heart beats per period;

[0026] Step S25: monitoring the fetal heart rate and fetal movement frequency of the target fetus, and obtaining a plurality of fetal heart acceleration periods and a plurality of fetal heart deceleration periods according to the monitoring results;

[0027] Step S26: Analyze multiple fetal heart acceleration periods to obtain a fetal heart acceleration monitoring coefficient corresponding to the target fetus;

[0028] Step S27: Analyze multiple fetal heart rate deceleration periods to obtain a fetal heart rate deceleration monitoring coefficient corresponding to the target fetus;

[0029] Step S28: define the average number of fetal heart beats per period, the fetal heart acceleration monitoring coefficient, and the fetal heart deceleration monitoring coefficient as fetal sound monitoring data.

[0030] Furthermore, the step S25 further includes the following specific steps:

[0031] Step S251: marking a number of fetal heart monitoring time points within a fetal heart monitoring cycle, and naming the marked fetal heart monitoring time points as the first fetal heart monitoring time point to the jth fetal heart monitoring time point in chronological order;

[0032] Step S252: acquiring the fetal heart rate corresponding to the first fetal heart rate monitoring time point to the jth fetal heart rate monitoring time point respectively, and obtaining the first fetal heart rate to the jth fetal heart rate;

[0033] Step S253: Using the fetal heart monitoring time point as the horizontal coordinate and the fetal heart rate as the vertical coordinate, a plane rectangular coordinate system is created to obtain a fetal heart monitoring coordinate system;

[0034] Step S254: taking the first fetal heart monitoring time point as the abscissa and the first fetal heart beat rate as the ordinate to obtain the first fetal heart coordinate point, taking the second fetal heart monitoring time point as the abscissa and the second fetal heart beat rate as the ordinate to obtain the second fetal heart coordinate point, and so on, taking the j-th fetal heart monitoring time point as the abscissa and the j-th fetal heart beat rate as the ordinate to obtain the j-th fetal heart coordinate point;

[0035] Step S255: Mark the first fetal heart coordinate point to the j-th fetal heart coordinate point in the fetal heart monitoring coordinate system respectively, connect the first fetal heart coordinate point with the second fetal heart coordinate point to obtain the first fetal heart monitoring line, connect the second fetal heart coordinate point with the third fetal heart coordinate point to obtain the second fetal heart monitoring line, and so on, connect the j-1-th fetal heart coordinate point with the j-th fetal heart coordinate point to obtain the j-1-th fetal heart monitoring line;

[0036] Step S256: in the fetal heart monitoring coordinate system, respectively obtaining the slopes of the lines corresponding to the first fetal heart monitoring line to the j-1th fetal heart monitoring line, to obtain the slope of the first fetal heart line to the j-1th fetal heart line;

[0037] Step S257: Obtaining a reference value of the slope of the fetal heart line;

[0038] Step S258: if the slope of the first fetal heart line is greater than the reference value of the slope of the fetal heart line, the period between the first fetal heart monitoring time point and the second fetal heart monitoring time point is marked as the fetal heart acceleration period, and the slope of the first fetal heart line is named the fetal heart acceleration rate; if the slope of the first fetal heart line is less than or equal to the reference value of the slope of the fetal heart line, the period between the first fetal heart monitoring time point and the second fetal heart monitoring time point is marked as the fetal heart deceleration period, and the absolute value of the slope of the first fetal heart line is named the fetal heart deceleration rate;

[0039] Step S259: numerically compare the slopes of the second fetal heart line to the slopes of the j-1th fetal heart line with the fetal heart line slope reference value to obtain multiple fetal heart acceleration periods and multiple fetal heart deceleration periods.

[0040] Furthermore, the step S26 further includes the following specific steps:

[0041] Step S261: respectively obtaining the fetal heart acceleration rate corresponding to each fetal heart acceleration period, obtaining the fetal heart acceleration rate threshold, numerically comparing the fetal heart acceleration rate with the fetal heart acceleration rate threshold, and dividing the fetal heart acceleration period into a normal fetal heart acceleration period and an abnormal fetal heart acceleration period according to the numerical comparison;

[0042] The details are as follows:

[0043] When the fetal heart acceleration rate is greater than or equal to the fetal heart acceleration rate threshold, the corresponding fetal heart acceleration period is marked as an abnormal fetal heart acceleration period;

[0044] When the fetal heart acceleration rate is less than the fetal heart acceleration rate threshold, the corresponding fetal heart acceleration period is marked as a normal fetal heart acceleration period;

[0045] Step S262: Obtain the number of fetal heart acceleration periods to obtain the number value of the first acceleration period, obtain the number value corresponding to the abnormal fetal heart acceleration period, and obtain the number value of the second acceleration period;

[0046] Step S263: acquiring the fetal heart acceleration rate corresponding to each abnormal fetal heart acceleration period to obtain multiple fetal heart acceleration rates, comparing the values ​​of the multiple fetal heart acceleration rates obtained, and marking the fetal heart acceleration rate with the largest value as the peak fetal heart acceleration rate;

[0047] Step S264: Calculate the peak fetal heart acceleration rate, the first acceleration period value, and the second acceleration period value to obtain a fetal heart acceleration monitoring coefficient corresponding to the target fetus;

[0048] The fetal heart acceleration monitoring coefficient corresponding to the target fetus is calculated, and the specific formula is as follows:

[0049]

[0050] Among them, Tjx is the fetal heart acceleration monitoring coefficient corresponding to the target fetus, Js1 is the number value of the first acceleration period, Js2 is the number value of the second acceleration period, and Jsf is the peak fetal heart acceleration rate.

[0051] Furthermore, the step S27 further includes the following specific steps:

[0052] Step S271: respectively obtaining the fetal heart deceleration rate corresponding to each fetal heart deceleration period, obtaining the fetal heart deceleration rate threshold, numerically comparing the fetal heart deceleration rate with the fetal heart deceleration rate threshold, and dividing the fetal heart deceleration period into a normal fetal heart deceleration period and an abnormal fetal heart deceleration period according to the numerical comparison;

[0053] The details are as follows:

[0054] When the fetal heart rate deceleration rate is greater than or equal to the fetal heart rate deceleration rate threshold, the corresponding fetal heart rate deceleration period is marked as an abnormal fetal heart rate deceleration period;

[0055] When the fetal heart rate deceleration rate is less than the fetal heart rate deceleration rate threshold, the corresponding fetal heart rate deceleration period is marked as a normal fetal heart rate deceleration period;

[0056] Step S272: obtaining the number of fetal heart deceleration periods, obtaining the number value of the first deceleration period, obtaining the number value corresponding to the abnormal fetal heart deceleration period, and obtaining the number value of the second deceleration period;

[0057] Step S273: acquiring the fetal heart deceleration rate corresponding to each abnormal fetal heart deceleration period to obtain multiple fetal heart deceleration rates, comparing the values ​​of the obtained multiple fetal heart deceleration rates, and marking the fetal heart deceleration rate with the largest value as the peak fetal heart deceleration rate;

[0058] Step S274: Calculate the peak fetal heart rate deceleration rate, the number of the first deceleration period, and the number of the second deceleration period to obtain a fetal heart rate deceleration monitoring coefficient corresponding to the target fetus;

[0059] The fetal heart rate deceleration monitoring coefficient corresponding to the target fetus is calculated. The specific formula is as follows:

[0060]

[0061] Among them, Tsx is the fetal heart rate deceleration monitoring coefficient corresponding to the target fetus, Ss1 is the number of the first deceleration period, Ss2 is the number of the second deceleration period, and Ssf is the peak fetal heart rate deceleration rate.

[0062] Furthermore, the step S3 further includes the following specific steps:

[0063] Step S31: obtaining preliminary fetal diagnosis data, and when the target patient is a first-type diagnosed fetus in the preliminary fetal diagnosis data, determining that the target fetus has a diagnosis abnormality, and issuing a diagnosis abnormality warning;

[0064] Step S32: When the target fetus is diagnosed as a second type fetus in the preliminary fetal diagnosis data, further auxiliary diagnosis is performed on the target fetus;

[0065] The step S32 further includes the following specific steps:

[0066] Step S321: Obtain fetal sound monitoring data, and obtain the average number of fetal heart beats per period, the fetal heart acceleration monitoring coefficient, and the fetal heart deceleration monitoring coefficient according to the fetal sound monitoring data;

[0067] Step S322: Calculate the average number of fetal heart beats in a cycle, the fetal heart acceleration monitoring coefficient, and the fetal heart deceleration monitoring coefficient to obtain a fetal sound auxiliary diagnosis coefficient;

[0068] The fetal sound auxiliary diagnosis coefficient is calculated, and the specific formula is as follows:

[0069] Tyf=Tsx 2 +Tjx 2 +Ztc;

[0070] Among them, Tyf is the fetal sound auxiliary diagnosis coefficient, Tsx is the fetal heart deceleration monitoring coefficient, Tjx is the fetal heart acceleration monitoring coefficient, and Ztc is the average number of fetal heart beats per cycle;

[0071] Step S323: Obtain the fetal sound auxiliary diagnosis coefficient threshold, perform numerical comparison between the fetal sound auxiliary diagnosis coefficient and the fetal sound auxiliary diagnosis coefficient threshold, and perform auxiliary diagnosis warning on the fetus based on the numerical comparison result.

[0072] Furthermore, the step S323 further includes the following specific steps:

[0073] Step S3231: respectively obtaining a threshold value of the average number of fetal heart beats in a period, a threshold value of a fetal heart acceleration monitoring coefficient, and a threshold value of a fetal heart deceleration monitoring coefficient;

[0074] Step S3232: Calculate the threshold of the average number of fetal heart beats in a cycle, the threshold of the fetal heart acceleration monitoring coefficient, and the threshold of the fetal heart deceleration monitoring coefficient to obtain a threshold of the fetal sound auxiliary diagnosis coefficient;

[0075] The fetal sound auxiliary diagnosis coefficient threshold is calculated, and the specific formula is as follows:

[0076] Tyfy=Tsxy 2 +Tjxy 2 +Ztcy;

[0077] Among them, Tyfy is the threshold value of the fetal sound auxiliary diagnosis coefficient, Tsxy is the threshold value of the fetal heart deceleration monitoring coefficient, Tjxy is the threshold value of the fetal heart acceleration monitoring coefficient, and Ztcy is the threshold value of the average number of fetal heart beats per cycle;

[0078] Step S3233: If the fetal sound auxiliary diagnosis coefficient is greater than or equal to the fetal sound auxiliary diagnosis coefficient threshold, it is determined that the target fetus has a diagnosis abnormality, and a diagnosis abnormality warning is issued;

[0079] Step S3234: If the fetal sound auxiliary diagnosis coefficient is less than the fetal sound auxiliary diagnosis coefficient threshold, it is determined that there is no diagnostic abnormality in the target fetus, and no diagnostic abnormality warning is issued.

[0080] Fetal sound monitoring and artificial intelligence-assisted diagnosis system, including:

[0081] Image data module: obtain maternal and fetal medical images and multiple fetal medical images, create a fetal image recognition model through the fetal medical images, use the fetal image recognition model to recognize the maternal and fetal medical images, and classify the target fetus into a first type of diagnosis fetus and a second type of diagnosis fetus according to the recognition results, and obtain preliminary fetal diagnosis data;

[0082] Fetal sound data module: obtain the fetal heart monitoring cycle, monitor the number of fetal heart beats of the second type of diagnosed fetus in the fetal heart monitoring cycle, obtain the average number of fetal heart beats in the cycle, monitor the fetal heart acceleration rate of the second type of diagnosed fetus in the fetal heart monitoring cycle, obtain the fetal heart acceleration monitoring coefficient, monitor the fetal heart deceleration rate of the second type of diagnosed fetus in the fetal heart monitoring cycle, obtain the fetal heart deceleration monitoring coefficient, and define the average number of fetal heart beats in the cycle, the fetal heart acceleration monitoring coefficient and the fetal heart deceleration monitoring coefficient as fetal sound monitoring data;

[0083] Auxiliary diagnosis module: Perform auxiliary diagnosis on the target fetus based on the preliminary diagnosis data of the fetus and fetal sound monitoring data, and issue an early warning based on the diagnosis results.

[0084] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0085] 1. The present invention creates a fetal image recognition model through fetal medical images, uses the fetal image recognition model to recognize maternal and fetal medical images, and divides the target fetus into a first type of diagnosis fetus and a second type of diagnosis fetus according to the recognition result, and performs targeted fetal tone monitoring on the second type of diagnosis fetus, which can improve the pertinence of the fetal tone monitoring process;

[0086] 2. The present invention performs auxiliary diagnosis on the second type of diagnosed fetus by obtaining the average number of fetal heart beats per period, the fetal heart acceleration monitoring coefficient and the fetal heart deceleration monitoring coefficient, thereby improving the accuracy of the auxiliary diagnosis results. BRIEF DESCRIPTION OF THE DRAWINGS

[0087] In order to facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0088] Figure 1 It is a diagram of the implementation steps of the present invention;

[0089] Figure 2 is the overall system block diagram of the present invention;

[0090] Figure 3 It is the fetal heart monitoring coordinate system of the present invention. DETAILED DESCRIPTION

[0091] 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.

[0092] Embodiment 1

[0093] See also Figure 1 The present invention provides a technical solution: a fetal sound monitoring and artificial intelligence-assisted diagnosis method, comprising the following specific steps:

[0094] Step S1: obtaining maternal-fetal medical images and multiple fetal medical images, creating a fetal image recognition model through the fetal medical images, using the fetal image recognition model to recognize the maternal-fetal medical images, and classifying the target fetus into a first type of diagnosis fetus and a second type of diagnosis fetus according to the recognition results, to obtain preliminary fetal diagnosis data;

[0095] The step S1 further includes the following specific steps:

[0096] Step S11: obtaining a medical image of the target fetus in the pregnant woman's belly through a medical detection device to obtain a medical image of the mother and the fetus;

[0097] Step S12: Acquire the gestational month of the target fetus to obtain the gestational month of the target fetus;

[0098] Step S13: creating a fetal image recognition model;

[0099] The step S13 further includes the following specific steps:

[0100] Step S131: acquiring multiple fetal medical images of the same gestational month as the target fetus, and dividing the multiple fetal medical images into first-type medical images and second-type medical images by manual identification and manual marking to obtain fetal medical image marking data;

[0101] Step S132: dividing the fetal image labeled data into a fetal image training set and a fetal image test set according to the image training test ratio;

[0102] Step S133: creating an image recognition model through an existing artificial intelligence platform, and training the image recognition model using a fetal image training set, until each medical fetal image in the fetal image training set has trained the image recognition model once;

[0103] Step S134: testing the image recognition model using the fetal image test set and obtaining the recognition accuracy. When the recognition accuracy is greater than or equal to the target recognition accuracy, the image recognition model training is completed and the fetal image recognition model is obtained. When the recognition accuracy is less than the target recognition accuracy, the image recognition model is continuously trained using the fetal image training set until the recognition accuracy is greater than or equal to the target recognition accuracy.

[0104] Step S14: using the fetal image recognition model to recognize the maternal and fetal medical images, and classifying the target fetus into a first type of diagnosis fetus and a second type of diagnosis fetus according to the recognition result, to obtain preliminary diagnosis data of the fetus;

[0105] The step S14 further includes the following specific steps:

[0106] Step S141: when the maternal-fetal medical image identifies the maternal-fetal medical image as a first type of fetal image, the target fetus is classified as a first type of diagnostic fetus;

[0107] Step S142: when the maternal-fetal medical image identifies the maternal-fetal medical image as a second type of fetal image, the target fetus is classified as a second type of diagnostic fetus;

[0108] Step S2: fetal sound monitoring is performed on the second type of diagnosed fetus, and the average number of fetal heart beats per cycle, the fetal heart acceleration monitoring coefficient, and the fetal heart deceleration monitoring coefficient are respectively obtained to obtain fetal sound monitoring data;

[0109] The step S2 further includes the following specific steps:

[0110] Step S21: during the process of fetal heart rate monitoring of the target fetus, marking a fetal heart rate monitoring cycle;

[0111] Step S22: dividing the fetal heart monitoring cycle into a number of fetal heart monitoring time periods of equal duration, and naming the fetal heart monitoring time periods as the first fetal heart monitoring time period to the tth fetal heart monitoring time period in chronological order;

[0112] Step S23: acquiring the fetal heart beats of the target fetus corresponding to the first fetal heart monitoring period to the t-th fetal heart monitoring period, and obtaining the fetal heart beats from the first period to the t-th period;

[0113] Step S24: Calculate the average of the fetal heart beats from the first period to the tth period to obtain the average fetal heart beats per period;

[0114] Step S25: monitoring the fetal heart rate and fetal movement frequency of the target fetus, and obtaining a plurality of fetal heart acceleration periods and a plurality of fetal heart deceleration periods according to the monitoring results;

[0115] The step S25 further includes the following specific steps:

[0116] Step S251: marking a number of fetal heart monitoring time points within a fetal heart monitoring cycle, and naming the marked fetal heart monitoring time points as the first fetal heart monitoring time point to the jth fetal heart monitoring time point in chronological order;

[0117] Step S252: acquiring the fetal heart rate corresponding to the first fetal heart rate monitoring time point to the jth fetal heart rate monitoring time point respectively, and obtaining the first fetal heart rate to the jth fetal heart rate;

[0118] Step S253: Using the fetal heart monitoring time point as the horizontal coordinate and the fetal heart rate as the vertical coordinate, a plane rectangular coordinate system is created to obtain a fetal heart monitoring coordinate system;

[0119] Step S254: taking the first fetal heart monitoring time point as the abscissa and the first fetal heart beat rate as the ordinate to obtain the first fetal heart coordinate point, taking the second fetal heart monitoring time point as the abscissa and the second fetal heart beat rate as the ordinate to obtain the second fetal heart coordinate point, and so on, taking the j-th fetal heart monitoring time point as the abscissa and the j-th fetal heart beat rate as the ordinate to obtain the j-th fetal heart coordinate point;

[0120] Step S255: Mark the first fetal heart coordinate point to the j-th fetal heart coordinate point in the fetal heart monitoring coordinate system respectively, connect the first fetal heart coordinate point with the second fetal heart coordinate point to obtain the first fetal heart monitoring line, connect the second fetal heart coordinate point with the third fetal heart coordinate point to obtain the second fetal heart monitoring line, and so on, connect the j-1-th fetal heart coordinate point with the j-th fetal heart coordinate point to obtain the j-1-th fetal heart monitoring line;

[0121] Step S256: in the fetal heart monitoring coordinate system, respectively obtaining the slopes of the lines corresponding to the first fetal heart monitoring line to the j-1th fetal heart monitoring line, to obtain the slope of the first fetal heart line to the j-1th fetal heart line;

[0122] Step S257: Obtaining a reference value of the slope of the fetal heart line;

[0123] Step S258: if the slope of the first fetal heart line is greater than the reference value of the slope of the fetal heart line, the period between the first fetal heart monitoring time point and the second fetal heart monitoring time point is marked as the fetal heart acceleration period, and the slope of the first fetal heart line is named the fetal heart acceleration rate; if the slope of the first fetal heart line is less than or equal to the reference value of the slope of the fetal heart line, the period between the first fetal heart monitoring time point and the second fetal heart monitoring time point is marked as the fetal heart deceleration period, and the absolute value of the slope of the first fetal heart line is named the fetal heart deceleration rate;

[0124] Step S259: numerically comparing the slopes of the second fetal heart line to the slopes of the j-1th fetal heart line with the reference value of the fetal heart line slope to obtain a plurality of fetal heart acceleration periods and a plurality of fetal heart deceleration periods;

[0125] Step S26: Analyze multiple fetal heart acceleration periods to obtain a fetal heart acceleration monitoring coefficient corresponding to the target fetus;

[0126] The step S26 further includes the following specific steps:

[0127] Step S261: respectively obtaining the fetal heart acceleration rate corresponding to each fetal heart acceleration period, obtaining the fetal heart acceleration rate threshold, numerically comparing the fetal heart acceleration rate with the fetal heart acceleration rate threshold, and dividing the fetal heart acceleration period into a normal fetal heart acceleration period and an abnormal fetal heart acceleration period according to the numerical comparison;

[0128] The details are as follows:

[0129] When the fetal heart acceleration rate is greater than or equal to the fetal heart acceleration rate threshold, the corresponding fetal heart acceleration period is marked as an abnormal fetal heart acceleration period;

[0130] When the fetal heart acceleration rate is less than the fetal heart acceleration rate threshold, the corresponding fetal heart acceleration period is marked as a normal fetal heart acceleration period;

[0131] Step S262: Obtain the number of fetal heart acceleration periods to obtain the number value of the first acceleration period, obtain the number value corresponding to the abnormal fetal heart acceleration period, and obtain the number value of the second acceleration period;

[0132] Step S263: acquiring the fetal heart acceleration rate corresponding to each abnormal fetal heart acceleration period to obtain multiple fetal heart acceleration rates, comparing the values ​​of the multiple fetal heart acceleration rates obtained, and marking the fetal heart acceleration rate with the largest value as the peak fetal heart acceleration rate;

[0133] Step S264: Calculate the peak fetal heart acceleration rate, the first acceleration period value, and the second acceleration period value to obtain a fetal heart acceleration monitoring coefficient corresponding to the target fetus;

[0134] The fetal heart acceleration monitoring coefficient corresponding to the target fetus is calculated, and the specific formula is as follows:

[0135]

[0136] Among them, Tjx is the fetal heart acceleration monitoring coefficient corresponding to the target fetus, Js1 is the number value of the first acceleration period, Js2 is the number value of the second acceleration period, and Jsf is the peak fetal heart acceleration rate;

[0137] Step S27: Analyze multiple fetal heart rate deceleration periods to obtain a fetal heart rate deceleration monitoring coefficient corresponding to the target fetus;

[0138] The step S27 further includes the following specific steps:

[0139] Step S271: respectively obtaining the fetal heart deceleration rate corresponding to each fetal heart deceleration period, obtaining the fetal heart deceleration rate threshold, numerically comparing the fetal heart deceleration rate with the fetal heart deceleration rate threshold, and dividing the fetal heart deceleration period into a normal fetal heart deceleration period and an abnormal fetal heart deceleration period according to the numerical comparison;

[0140] The details are as follows:

[0141] When the fetal heart rate deceleration rate is greater than or equal to the fetal heart rate deceleration rate threshold, the corresponding fetal heart rate deceleration period is marked as an abnormal fetal heart rate deceleration period;

[0142] When the fetal heart rate deceleration rate is less than the fetal heart rate deceleration rate threshold, the corresponding fetal heart rate deceleration period is marked as a normal fetal heart rate deceleration period;

[0143] Step S272: obtaining the number of fetal heart deceleration periods, obtaining the number value of the first deceleration period, obtaining the number value corresponding to the abnormal fetal heart deceleration period, and obtaining the number value of the second deceleration period;

[0144] Step S273: acquiring the fetal heart deceleration rate corresponding to each abnormal fetal heart deceleration period to obtain multiple fetal heart deceleration rates, comparing the values ​​of the obtained multiple fetal heart deceleration rates, and marking the fetal heart deceleration rate with the largest value as the peak fetal heart deceleration rate;

[0145] Step S274: Calculate the peak fetal heart rate deceleration rate, the number of the first deceleration period, and the number of the second deceleration period to obtain a fetal heart rate deceleration monitoring coefficient corresponding to the target fetus;

[0146] The fetal heart rate deceleration monitoring coefficient corresponding to the target fetus is calculated. The specific formula is as follows:

[0147]

[0148] Wherein, Tsx is the fetal heart rate deceleration monitoring coefficient corresponding to the target fetus, Ss1 is the number of the first deceleration period, Ss2 is the number of the second deceleration period, and Ssf is the peak fetal heart rate deceleration rate;

[0149] Step S28: defining the average number of fetal heart beats per period, the fetal heart acceleration monitoring coefficient, and the fetal heart deceleration monitoring coefficient as fetal sound monitoring data;

[0150] Step S3: Perform auxiliary diagnosis on the target fetus based on the preliminary diagnosis data of the fetus and the fetal sound monitoring data, and issue an early warning based on the diagnosis results;

[0151] The step S3 further includes the following specific steps:

[0152] Step S31: obtaining preliminary fetal diagnosis data, and when the target patient is a first-type diagnosed fetus in the preliminary fetal diagnosis data, determining that the target fetus has a diagnosis abnormality, and issuing a diagnosis abnormality warning;

[0153] Step S32: When the target fetus is diagnosed as a second type fetus in the preliminary fetal diagnosis data, further auxiliary diagnosis is performed on the target fetus;

[0154] The step S32 further includes the following specific steps:

[0155] Step S321: Obtain fetal sound monitoring data, and obtain the average number of fetal heart beats per period, the fetal heart acceleration monitoring coefficient, and the fetal heart deceleration monitoring coefficient according to the fetal sound monitoring data;

[0156] Step S322: Calculate the average number of fetal heart beats in a cycle, the fetal heart acceleration monitoring coefficient, and the fetal heart deceleration monitoring coefficient to obtain a fetal sound auxiliary diagnosis coefficient;

[0157] The fetal sound auxiliary diagnosis coefficient is calculated, and the specific formula is as follows:

[0158] Tyf=Tsx 2 +Tjx 2 +Ztc;

[0159] Among them, Tyf is the fetal sound auxiliary diagnosis coefficient, Tsx is the fetal heart deceleration monitoring coefficient, Tjx is the fetal heart acceleration monitoring coefficient, and Ztc is the average number of fetal heart beats per cycle;

[0160] Step S323: obtaining a fetal sound auxiliary diagnosis coefficient threshold, performing numerical comparison between the fetal sound auxiliary diagnosis coefficient and the fetal sound auxiliary diagnosis coefficient threshold, and performing auxiliary diagnosis warning for the fetus according to the numerical comparison result;

[0161] The step S323 further includes the following specific steps:

[0162] Step S3231: respectively obtaining a threshold value of the average number of fetal heart beats in a period, a threshold value of a fetal heart acceleration monitoring coefficient, and a threshold value of a fetal heart deceleration monitoring coefficient;

[0163] Step S3232: Calculate the threshold of the average number of fetal heart beats in a cycle, the threshold of the fetal heart acceleration monitoring coefficient, and the threshold of the fetal heart deceleration monitoring coefficient to obtain a threshold of the fetal sound auxiliary diagnosis coefficient;

[0164] The fetal sound auxiliary diagnosis coefficient threshold is calculated, and the specific formula is as follows:

[0165] Tyfy=Tsxy 2 +Tjxy 2 +Ztcy;

[0166] Among them, Tyfy is the threshold value of the fetal sound auxiliary diagnosis coefficient, Tsxy is the threshold value of the fetal heart deceleration monitoring coefficient, Tfxy is the threshold value of the fetal heart acceleration monitoring coefficient, and Ztcy is the threshold value of the average number of fetal heart beats in a period;

[0167] Step S3233: If the fetal sound auxiliary diagnosis coefficient is greater than or equal to the fetal sound auxiliary diagnosis coefficient threshold, it is determined that the target fetus has a diagnosis abnormality, and a diagnosis abnormality warning is issued;

[0168] Step S3234: If the fetal sound auxiliary diagnosis coefficient is less than the fetal sound auxiliary diagnosis coefficient threshold, it is determined that there is no diagnostic abnormality in the target fetus, and no diagnostic abnormality warning is issued.

[0169] 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.

[0170] Embodiment 2

[0171] See also Figure 2 Based on another concept of the same invention, a fetal sound monitoring and artificial intelligence-assisted diagnosis system is proposed, including an image data module, a fetal sound data module, an auxiliary diagnosis module and a server, wherein the image data module, the fetal sound data module and the auxiliary diagnosis module are respectively connected to the server, and the server controls the image data module, the fetal sound data module and the auxiliary diagnosis module respectively:

[0172] The image data module obtains maternal and fetal medical images and multiple fetal medical images, creates a fetal image recognition model through the fetal medical images, uses the fetal image recognition model to recognize the maternal and fetal medical images, and divides the target fetus into a first type of diagnosis fetus and a second type of diagnosis fetus according to the recognition result, and obtains preliminary fetal diagnosis data;

[0173] The details are as follows:

[0174] Obtain medical images of the target fetus in the pregnant woman's belly through medical detection equipment to obtain medical images of the mother and fetus;

[0175] It should be noted here that:

[0176] In this application, the medical image designed here is specifically a B-ultrasound image;

[0177] In the present application, the target fetus referred to herein is specifically the fetus that needs auxiliary diagnosis in the present application;

[0178] The gestational month of the target fetus is obtained to obtain the gestational month of the target fetus;

[0179] Create a fetal image recognition model;

[0180] The details are as follows:

[0181] Acquire multiple fetal medical images of the same gestational month as the target fetus, and classify the multiple fetal medical images into first-type medical images and second-type medical images by manual identification and manual labeling to obtain fetal medical image labeling data;

[0182] It should be noted here that:

[0183] In the present application, the first type of fetal image referred to herein is a normal fetal image, and the second type of fetal image referred to herein is an abnormal fetal image;

[0184] In the present application, the specific abnormal types corresponding to the abnormal fetal images designed herein include but are not limited to nervous system abnormalities, cardiac structural abnormalities, and digestive system abnormalities.

[0185] The fetal image labeled data is divided into a fetal image training set and a fetal image test set according to the image training and testing ratio;

[0186] It should be noted here that:

[0187] In the present application, the image training-test ratio involved herein is specifically 7:3, that is, the ratio of the number of images in the fetal image training set and the fetal image test set is 7:3;

[0188] Create an image recognition model through an existing artificial intelligence platform, and train the image recognition model using a fetal image training set until the image recognition model is trained once for each medical fetal image in the fetal image training set;

[0189] It should be noted here that:

[0190] In this application, the artificial intelligence platform designed here is specifically TensorFlow;

[0191] The image recognition model is tested using the fetal image test set, and the recognition accuracy is obtained. When the recognition accuracy is greater than or equal to the target recognition accuracy, the image recognition model training is completed and the fetal image recognition model is obtained. When the recognition accuracy is less than the target recognition accuracy, the image recognition model is continuously trained using the fetal image training set until the recognition accuracy is greater than or equal to the target recognition accuracy.

[0192] Using a fetal image recognition model to recognize maternal and fetal medical images, and classifying the target fetus into a first-type diagnosis fetus and a second-type diagnosis fetus according to the recognition results, to obtain preliminary fetal diagnosis data;

[0193] The details are as follows:

[0194] When the maternal-fetal medical image identifies the maternal-fetal medical image as a first type of fetal image, the target fetus is classified as a first type of diagnostic fetus;

[0195] When the maternal-fetal medical image identifies the maternal-fetal medical image as a second type of fetal image, the target fetus is classified as a second type of diagnostic fetus;

[0196] The image data module acquires preliminary fetal diagnostic data and transmits it to the fetal sound data module;

[0197] The fetal sound data module performs fetal sound monitoring on the second type of diagnosed fetus, and obtains the average number of fetal heart beats in a cycle, the fetal heart acceleration monitoring coefficient, and the fetal heart deceleration monitoring coefficient, to obtain fetal sound monitoring data;

[0198] The details are as follows:

[0199] During the process of fetal heart rate monitoring of the target fetus, marking a fetal heart rate monitoring cycle;

[0200] Dividing the fetal heart monitoring cycle into a number of fetal heart monitoring periods of equal time, and naming the fetal heart monitoring periods in chronological order as the first fetal heart monitoring period to the tth fetal heart monitoring period;

[0201] It should be noted here that:

[0202] In the present application, the fetal sound involved is specifically the sound of the fetal heart beating;

[0203] In this application, t referred to herein is a numerical value corresponding to a fetal heart rate monitoring period, and t is an integer greater than 0;

[0204] The fetal heart beats of the target fetus corresponding to the first fetal heart monitoring period to the tth fetal heart monitoring period are respectively obtained to obtain the fetal heart beats from the first period to the tth period;

[0205] Calculate the average of the fetal heart beats from the first period to the tth period to obtain the average fetal heart beats per period;

[0206] Perform fetal heart acceleration monitoring on the target fetus to obtain a fetal heart acceleration monitoring coefficient;

[0207] The details are as follows:

[0208] In a fetal heart monitoring cycle, a number of fetal heart monitoring time points are marked, and the marked fetal heart monitoring time points are named as the first fetal heart monitoring time point to the jth fetal heart monitoring time point in chronological order;

[0209] It should be noted here that:

[0210] In this application, j referred to herein is a quantity value corresponding to a fetal heart monitoring time point, and j is an integer greater than 0;

[0211] The fetal heart rate corresponding to the first fetal heart monitoring time point to the jth fetal heart monitoring time point is respectively acquired to obtain the first fetal heart rate to the jth fetal heart rate;

[0212] See also Figure 3, taking the fetal heart monitoring time point as the horizontal coordinate and the fetal heart rate as the vertical coordinate, a plane rectangular coordinate system is created to obtain a fetal heart monitoring coordinate system;

[0213] The first fetal heart monitoring time point is used as the abscissa and the first fetal heart beat rate is used as the ordinate to obtain the first fetal heart coordinate point. The second fetal heart monitoring time point is used as the abscissa and the second fetal heart beat rate is used as the ordinate to obtain the second fetal heart coordinate point. Similarly, the j-th fetal heart monitoring time point is used as the abscissa and the j-th fetal heart beat rate is used as the ordinate to obtain the j-th fetal heart coordinate point.

[0214] Mark the first fetal heart coordinate point to the jth fetal heart coordinate point in the fetal heart monitoring coordinate system respectively, connect the first fetal heart coordinate point with the second fetal heart coordinate point to obtain the first fetal heart monitoring line, connect the second fetal heart coordinate point with the third fetal heart coordinate point to obtain the second fetal heart monitoring line, and so on, connect the j-1th fetal heart coordinate point with the jth fetal heart coordinate point to obtain the j-1th fetal heart monitoring line;

[0215] In the fetal heart monitoring coordinate system, respectively obtain the slopes of the lines corresponding to the first fetal heart monitoring line to the j-1th fetal heart monitoring line, and obtain the slope of the first fetal heart line to the j-1th fetal heart line;

[0216] Obtaining a fetal heart line slope reference value, performing numerical comparison on the first fetal heart line slope to the j-1th fetal heart line slope, and obtaining multiple fetal heart acceleration periods and multiple fetal heart deceleration periods according to the numerical comparison results;

[0217] It should be noted here that:

[0218] In this application, the reference value of the slope of the fetal heart line involved here is a value of 0;

[0219] The details are as follows:

[0220] If the slope of the first fetal heart line is greater than the reference value of the slope of the fetal heart line, the period between the first fetal heart monitoring time point and the second fetal heart monitoring time point is marked as the fetal heart acceleration period, and the slope of the first fetal heart line is named the fetal heart acceleration rate; if the slope of the first fetal heart line is less than or equal to the reference value of the slope of the fetal heart line, the period between the first fetal heart monitoring time point and the second fetal heart monitoring time point is marked as the fetal heart deceleration period, and the absolute value of the slope of the first fetal heart line is named the fetal heart deceleration rate;

[0221] Repeat the process of comparing the first fetal heart line slope with the fetal heart line slope reference value, respectively compare the second fetal heart line slope to the j-1th fetal heart line slope with the fetal heart line slope reference value, and obtain multiple fetal heart acceleration periods and multiple fetal heart deceleration periods;

[0222] Analyze multiple fetal heart acceleration periods to obtain the fetal heart acceleration monitoring coefficient corresponding to the target fetus;

[0223] The details are as follows:

[0224] respectively obtaining the fetal heart acceleration rate corresponding to each fetal heart acceleration period, obtaining the fetal heart acceleration rate threshold, performing numerical comparison between the fetal heart acceleration rate and the fetal heart acceleration rate threshold, and dividing the fetal heart acceleration period into a normal fetal heart acceleration period and an abnormal fetal heart acceleration period according to the numerical comparison;

[0225] The details are as follows:

[0226] When the fetal heart acceleration rate is greater than or equal to the fetal heart acceleration rate threshold, the corresponding fetal heart acceleration period is marked as an abnormal fetal heart acceleration period;

[0227] When the fetal heart acceleration rate is less than the fetal heart acceleration rate threshold, the corresponding fetal heart acceleration period is marked as a normal fetal heart acceleration period;

[0228] Obtain the number of fetal heart acceleration periods to obtain the number value of the first acceleration period, obtain the number value corresponding to the abnormal fetal heart acceleration period, and obtain the number value of the second acceleration period;

[0229] Acquiring the fetal heart acceleration rate corresponding to each abnormal fetal heart acceleration period to obtain multiple fetal heart acceleration rates, comparing the values ​​of the multiple fetal heart acceleration rates obtained, and marking the fetal heart acceleration rate with the largest value as the peak fetal heart acceleration rate;

[0230] The peak fetal heart acceleration rate, the first acceleration period value and the second acceleration period value are calculated to obtain a fetal heart acceleration monitoring coefficient corresponding to the target fetus;

[0231] The fetal heart acceleration monitoring coefficient corresponding to the target fetus is calculated, and the specific formula is as follows:

[0232]

[0233] Among them, Tjx is the fetal heart acceleration monitoring coefficient corresponding to the target fetus, Js1 is the number value of the first acceleration period, Js2 is the number value of the second acceleration period, and Jsf is the peak fetal heart acceleration rate;

[0234] Analyze multiple fetal heart rate deceleration periods to obtain the fetal heart rate deceleration monitoring coefficient corresponding to the target fetus;

[0235] The details are as follows:

[0236] respectively obtaining the fetal heart deceleration rate corresponding to each fetal heart deceleration period, obtaining the fetal heart deceleration rate threshold, numerically comparing the fetal heart deceleration rate with the fetal heart deceleration rate threshold, and dividing the fetal heart deceleration period into a normal fetal heart deceleration period and an abnormal fetal heart deceleration period according to the numerical comparison;

[0237] The details are as follows:

[0238] When the fetal heart rate deceleration rate is greater than or equal to the fetal heart rate deceleration rate threshold, the corresponding fetal heart rate deceleration period is marked as an abnormal fetal heart rate deceleration period;

[0239] When the fetal heart rate deceleration rate is less than the fetal heart rate deceleration rate threshold, the corresponding fetal heart rate deceleration period is marked as a normal fetal heart rate deceleration period;

[0240] Obtain the number of fetal heart deceleration periods to obtain the number value of the first deceleration period, obtain the number value corresponding to the abnormal fetal heart deceleration period, and obtain the number value of the second deceleration period;

[0241] The fetal heart deceleration rate corresponding to each abnormal fetal heart deceleration period is obtained to obtain multiple fetal heart deceleration rates, and the multiple fetal heart deceleration rates obtained are compared in numerical value, and the fetal heart deceleration rate with the largest value is marked as the peak fetal heart deceleration rate;

[0242] The peak fetal heart rate deceleration rate, the first deceleration period value and the second deceleration period value are calculated to obtain a fetal heart rate deceleration monitoring coefficient corresponding to the target fetus;

[0243] The fetal heart rate deceleration monitoring coefficient corresponding to the target fetus is calculated. The specific formula is as follows:

[0244]

[0245] Wherein, Tsx is the fetal heart rate deceleration monitoring coefficient corresponding to the target fetus, Ss1 is the number of the first deceleration period, Ss2 is the number of the second deceleration period, and Ssf is the peak fetal heart rate deceleration rate;

[0246] The average number of fetal heart beats per cycle, the fetal heart acceleration monitoring coefficient and the fetal heart deceleration monitoring coefficient are defined as fetal sound monitoring data;

[0247] The fetal sound data module acquires the fetal sound monitoring data and transmits it to the auxiliary diagnosis module;

[0248] The auxiliary diagnosis module performs auxiliary diagnosis on the target fetus based on the preliminary diagnosis data of the fetus and the fetal sound monitoring data, and issues an early warning based on the diagnosis results;

[0249] Obtaining preliminary fetal diagnosis data, when the target patient is diagnosed as a first-type fetus in the preliminary fetal diagnosis data, it is determined that the target fetus has a diagnosis abnormality, and a diagnosis abnormality warning is issued;

[0250] When the target fetus is diagnosed as a second-type fetus in the preliminary fetal diagnosis data, further auxiliary diagnosis is performed on the target fetus;

[0251] The details are as follows:

[0252] Obtaining fetal sound monitoring data, and obtaining the average number of fetal heart beats per cycle, the fetal heart acceleration monitoring coefficient, and the fetal heart deceleration monitoring coefficient according to the fetal sound monitoring data;

[0253] The average number of fetal heart beats in a cycle, the fetal heart acceleration monitoring coefficient and the fetal heart deceleration monitoring coefficient are calculated to obtain the fetal sound auxiliary diagnosis coefficient;

[0254] The fetal sound auxiliary diagnosis coefficient is calculated, and the specific formula is as follows:

[0255] Tyf=Tsx 2 +Tjx 2 +Ztc;

[0256] Among them, Tyf is the fetal sound auxiliary diagnosis coefficient, Tsx is the fetal heart deceleration monitoring coefficient, Tjx is the fetal heart acceleration monitoring coefficient, and Ztc is the average number of fetal heart beats per cycle;

[0257] Obtaining a fetal sound auxiliary diagnosis coefficient threshold, performing numerical comparison between the fetal sound auxiliary diagnosis coefficient and the fetal sound auxiliary diagnosis coefficient threshold, and performing auxiliary diagnosis warning for the fetus according to the numerical comparison result;

[0258] The details are as follows:

[0259] The threshold of the average number of fetal heart beats in a cycle, the threshold of the fetal heart acceleration monitoring coefficient, and the threshold of the fetal heart deceleration monitoring coefficient are respectively obtained;

[0260] It should be noted here that:

[0261] The threshold of average fetal heart rate, fetal heart acceleration monitoring coefficient and fetal heart deceleration monitoring coefficient involved here are respectively the minimum average fetal heart rate, fetal heart acceleration monitoring coefficient and fetal heart deceleration monitoring coefficient corresponding to a fetus without diagnostic abnormality;

[0262] The threshold of the average number of fetal heart beats in a cycle, the threshold of the fetal heart acceleration monitoring coefficient, and the threshold of the fetal heart deceleration monitoring coefficient are calculated to obtain the threshold of the fetal sound auxiliary diagnosis coefficient;

[0263] The fetal sound auxiliary diagnosis coefficient threshold is calculated, and the specific formula is as follows:

[0264] Tyfy=Tsxy 2 +Tjxy 2 +Ztcy;

[0265] Among them, Tyfy is the threshold value of the fetal sound auxiliary diagnosis coefficient, Tsxy is the threshold value of the fetal heart deceleration monitoring coefficient, Tjxy is the threshold value of the fetal heart acceleration monitoring coefficient, and Ztcy is the threshold value of the average number of fetal heart beats per cycle;

[0266] If the fetal sound auxiliary diagnosis coefficient is greater than or equal to the fetal sound auxiliary diagnosis coefficient threshold, it is judged that the target fetus has a diagnosis abnormality, and a diagnosis abnormality warning is issued;

[0267] If the fetal sound auxiliary diagnosis coefficient is less than the fetal sound auxiliary diagnosis coefficient threshold, it is judged that there is no diagnostic abnormality in the target fetus and no diagnostic abnormality warning is issued.

[0268] 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.

Claims

1. Fetal sound monitoring and artificial intelligence-assisted diagnosis method, characterized in that: include: Step S1: obtaining a maternal-fetal medical image and multiple fetal medical images, creating a fetal image recognition model through the multiple fetal medical images, using the fetal image recognition model to recognize the maternal-fetal medical images, and classifying the target fetus into a first type of diagnosis fetus and a second type of diagnosis fetus according to the recognition result, to obtain preliminary fetal diagnosis data; Step S2: obtaining a fetal heart monitoring cycle, monitoring the number of fetal heart beats of the second type of diagnosed fetus in the fetal heart monitoring cycle to obtain an average number of fetal heart beats in the cycle, monitoring the fetal heart acceleration rate of the second type of diagnosed fetus in the fetal heart monitoring cycle to obtain a fetal heart acceleration monitoring coefficient, monitoring the fetal heart deceleration rate of the second type of diagnosed fetus in the fetal heart monitoring cycle to obtain a fetal heart deceleration monitoring coefficient, and defining the average number of fetal heart beats in the cycle, the fetal heart acceleration monitoring coefficient, and the fetal heart deceleration monitoring coefficient as fetal sound monitoring data; Step S3: Perform auxiliary diagnosis on the target fetus based on the preliminary fetal diagnosis data and fetal sound monitoring data, and issue an early warning based on the diagnosis results.

2. The fetal sound monitoring and artificial intelligence-assisted diagnosis method according to claim 1, characterized in that: The step S1 further includes the following specific steps: Step S11: obtaining a medical image of the target fetus in the pregnant woman's belly through a medical detection device to obtain a medical image of the mother and the fetus; Step S12: Acquire the gestational month of the target fetus to obtain the gestational month of the target fetus; Step S13: creating a fetal image recognition model; Step S14: Use the fetal image recognition model to recognize maternal and fetal medical images, and classify the target fetus into a first type of diagnosis fetus and a second type of diagnosis fetus according to the recognition result to obtain preliminary fetal diagnosis data.

3. The fetal sound monitoring and artificial intelligence-assisted diagnosis method according to claim 2, characterized in that: The step S13 further includes the following specific steps: Step S131: acquiring multiple fetal medical images of the same gestational month as the target fetus, and dividing the multiple fetal medical images into first-type medical images and second-type medical images by manual identification and manual marking to obtain fetal medical image marking data; Step S132: dividing the fetal image labeled data into a fetal image training set and a fetal image test set according to the image training test ratio; Step S133: creating an image recognition model through an existing artificial intelligence platform, and training the image recognition model using a fetal image training set; Step S134: Use the fetal image test set to test the image recognition model and obtain the recognition accuracy. When the recognition accuracy is greater than or equal to the target recognition accuracy, the image recognition model training is completed and the fetal image recognition model is obtained. When the recognition accuracy is less than the target recognition accuracy, continue to use the fetal image training set to train the image recognition model until the recognition accuracy is greater than or equal to the target recognition accuracy.

4. The fetal sound monitoring and artificial intelligence-assisted diagnosis method according to claim 1, characterized in that: The step S2 further includes the following specific steps: Step S21: during the process of fetal heart rate monitoring of the target fetus, marking a fetal heart rate monitoring cycle; Step S22: dividing the fetal heart monitoring cycle into a number of fetal heart monitoring time periods of equal duration, and naming the fetal heart monitoring time periods as the first fetal heart monitoring time period to the tth fetal heart monitoring time period in chronological order; Step S23: acquiring the fetal heart beats of the target fetus corresponding to the first fetal heart monitoring period to the t-th fetal heart monitoring period, and obtaining the fetal heart beats from the first period to the t-th period; Step S24: Calculate the average of the fetal heart beats from the first period to the tth period to obtain the average fetal heart beats per period; Step S25: monitoring the fetal heart rate and fetal movement frequency of the target fetus, and obtaining a plurality of fetal heart acceleration periods and a plurality of fetal heart deceleration periods according to the monitoring results; Step S26: Analyze multiple fetal heart acceleration periods to obtain a fetal heart acceleration monitoring coefficient corresponding to the target fetus; Step S27: Analyze multiple fetal heart rate deceleration periods to obtain a fetal heart rate deceleration monitoring coefficient corresponding to the target fetus; Step S28: define the average number of fetal heart beats per period, the fetal heart acceleration monitoring coefficient, and the fetal heart deceleration monitoring coefficient as fetal sound monitoring data.

5. The fetal sound monitoring and artificial intelligence-assisted diagnosis method according to claim 4, characterized in that: The step S25 further includes the following specific steps: Step S251: marking a number of fetal heart monitoring time points within a fetal heart monitoring cycle, and naming the marked fetal heart monitoring time points as the first fetal heart monitoring time point to the jth fetal heart monitoring time point in chronological order; Step S252: acquiring the fetal heart rate corresponding to the first fetal heart rate monitoring time point to the jth fetal heart rate monitoring time point respectively, and obtaining the first fetal heart rate to the jth fetal heart rate; Step S253: Using the fetal heart monitoring time point as the horizontal coordinate and the fetal heart rate as the vertical coordinate, a plane rectangular coordinate system is created to obtain a fetal heart monitoring coordinate system; Step S254: taking the first fetal heart monitoring time point as the abscissa and the first fetal heart beat rate as the ordinate to obtain the first fetal heart coordinate point, taking the second fetal heart monitoring time point as the abscissa and the second fetal heart beat rate as the ordinate to obtain the second fetal heart coordinate point, and so on, taking the j-th fetal heart monitoring time point as the abscissa and the j-th fetal heart beat rate as the ordinate to obtain the j-th fetal heart coordinate point; Step S255: Mark the first fetal heart coordinate point to the j-th fetal heart coordinate point in the fetal heart monitoring coordinate system respectively, connect the first fetal heart coordinate point with the second fetal heart coordinate point to obtain the first fetal heart monitoring line, connect the second fetal heart coordinate point with the third fetal heart coordinate point to obtain the second fetal heart monitoring line, and so on, connect the j-1-th fetal heart coordinate point with the j-th fetal heart coordinate point to obtain the j-1-th fetal heart monitoring line; Step S256: in the fetal heart monitoring coordinate system, obtaining the slope of the line corresponding to the first fetal heart monitoring line to the j-1th fetal heart monitoring line, and obtaining the slope of the first fetal heart line to the j-1th fetal heart line; Step S257: Obtaining a reference value of the slope of the fetal heart line; Step S258: if the slope of the first fetal heart line is greater than the reference value of the slope of the fetal heart line, the period between the first fetal heart monitoring time point and the second fetal heart monitoring time point is marked as the fetal heart acceleration period, and the slope of the first fetal heart line is named the fetal heart acceleration rate; if the slope of the first fetal heart line is less than or equal to the reference value of the slope of the fetal heart line, the period between the first fetal heart monitoring time point and the second fetal heart monitoring time point is marked as the fetal heart deceleration period, and the absolute value of the slope of the first fetal heart line is named the fetal heart deceleration rate; Step S259: numerically compare the slopes of the second fetal heart line to the slopes of the j-1th fetal heart line with the fetal heart line slope reference value to obtain multiple fetal heart acceleration periods and multiple fetal heart deceleration periods.

6. The fetal sound monitoring and artificial intelligence-assisted diagnosis method according to claim 4, characterized in that: The step S26 further includes the following specific steps: Step S261: respectively obtaining the fetal heart acceleration rate corresponding to each fetal heart acceleration period, obtaining the fetal heart acceleration rate threshold, numerically comparing the fetal heart acceleration rate with the fetal heart acceleration rate threshold, and dividing the fetal heart acceleration period into a normal fetal heart acceleration period and an abnormal fetal heart acceleration period according to the numerical comparison; The details are as follows: When the fetal heart acceleration rate is greater than or equal to the fetal heart acceleration rate threshold, the corresponding fetal heart acceleration period is marked as an abnormal fetal heart acceleration period; When the fetal heart acceleration rate is less than the fetal heart acceleration rate threshold, the corresponding fetal heart acceleration period is marked as a normal fetal heart acceleration period; Step S262: Obtain the number of fetal heart acceleration periods to obtain the number value of the first acceleration period, obtain the number value corresponding to the abnormal fetal heart acceleration period, and obtain the number value of the second acceleration period; Step S263: acquiring the fetal heart acceleration rate corresponding to each abnormal fetal heart acceleration period to obtain multiple fetal heart acceleration rates, comparing the values ​​of the multiple fetal heart acceleration rates obtained, and marking the fetal heart acceleration rate with the largest value as the peak fetal heart acceleration rate; Step S264: Calculate the peak fetal heart acceleration rate, the first acceleration period value, and the second acceleration period value to obtain a fetal heart acceleration monitoring coefficient corresponding to the target fetus; The fetal heart acceleration monitoring coefficient corresponding to the target fetus is calculated, and the specific formula is as follows: Among them, Tjx is the fetal heart acceleration monitoring coefficient corresponding to the target fetus, Js1 is the number value of the first acceleration period, Js2 is the number value of the second acceleration period, and Jsf is the peak fetal heart acceleration rate.

7. The fetal sound monitoring and artificial intelligence-assisted diagnosis method according to claim 4, characterized in that: The step S27 further includes the following specific steps: Step S271: respectively obtaining the fetal heart deceleration rate corresponding to each fetal heart deceleration period, obtaining the fetal heart deceleration rate threshold, numerically comparing the fetal heart deceleration rate with the fetal heart deceleration rate threshold, and dividing the fetal heart deceleration period into a normal fetal heart deceleration period and an abnormal fetal heart deceleration period according to the numerical comparison; The details are as follows: When the fetal heart rate deceleration rate is greater than or equal to the fetal heart rate deceleration rate threshold, the corresponding fetal heart rate deceleration period is marked as an abnormal fetal heart rate deceleration period; When the fetal heart rate deceleration rate is less than the fetal heart rate deceleration rate threshold, the corresponding fetal heart rate deceleration period is marked as a normal fetal heart rate deceleration period; Step S272: obtaining the number of fetal heart deceleration periods, obtaining the number value of the first deceleration period, obtaining the number value corresponding to the abnormal fetal heart deceleration period, and obtaining the number value of the second deceleration period; Step S273: acquiring the fetal heart deceleration rate corresponding to each abnormal fetal heart deceleration period to obtain multiple fetal heart deceleration rates, comparing the values ​​of the obtained multiple fetal heart deceleration rates, and marking the fetal heart deceleration rate with the largest value as the peak fetal heart deceleration rate; Step S274: Calculate the peak fetal heart rate deceleration rate, the number of the first deceleration period, and the number of the second deceleration period to obtain a fetal heart rate deceleration monitoring coefficient corresponding to the target fetus; The fetal heart rate deceleration monitoring coefficient corresponding to the target fetus is calculated. The specific formula is as follows: Among them, Tsx is the fetal heart rate deceleration monitoring coefficient corresponding to the target fetus, Ss1 is the number of the first deceleration period, Ss2 is the number of the second deceleration period, and Ssf is the peak fetal heart rate deceleration rate.

8. The fetal sound monitoring and artificial intelligence-assisted diagnosis method according to claim 1, characterized in that: The step S3 further includes the following specific steps: Step S31: obtaining preliminary fetal diagnosis data, and when the target patient is a first-type diagnosed fetus in the preliminary fetal diagnosis data, determining that the target fetus has a diagnosis abnormality, and issuing a diagnosis abnormality warning; Step S32: When the target fetus is diagnosed as a second type fetus in the preliminary fetal diagnosis data, further auxiliary diagnosis is performed on the target fetus; The step S32 further includes the following specific steps: Step S321: Obtain fetal sound monitoring data, and obtain the average number of fetal heart beats per period, the fetal heart acceleration monitoring coefficient, and the fetal heart deceleration monitoring coefficient according to the fetal sound monitoring data; Step S322: Calculate the average number of fetal heart beats in a cycle, the fetal heart acceleration monitoring coefficient, and the fetal heart deceleration monitoring coefficient to obtain a fetal sound auxiliary diagnosis coefficient; The fetal sound auxiliary diagnosis coefficient is calculated, and the specific formula is as follows: Type=Tsx 2 +Tjx 2 +Ztc; Among them, Tyf is the fetal sound auxiliary diagnosis coefficient, Tsx is the fetal heart deceleration monitoring coefficient, Tjx is the fetal heart acceleration monitoring coefficient, and Ztc is the average number of fetal heart beats per cycle; Step S323: Obtain the fetal sound auxiliary diagnosis coefficient threshold, perform numerical comparison between the fetal sound auxiliary diagnosis coefficient and the fetal sound auxiliary diagnosis coefficient threshold, and perform auxiliary diagnosis warning on the fetus based on the numerical comparison result.

9. The fetal sound monitoring and artificial intelligence-assisted diagnosis method according to claim 8, characterized in that: The step S323 further includes the following specific steps: Step S3231: respectively obtaining a threshold value of the average number of fetal heart beats in a period, a threshold value of a fetal heart acceleration monitoring coefficient, and a threshold value of a fetal heart deceleration monitoring coefficient; Step S3232: Calculate the threshold of the average number of fetal heart beats in a cycle, the threshold of the fetal heart acceleration monitoring coefficient, and the threshold of the fetal heart deceleration monitoring coefficient to obtain a threshold of the fetal sound auxiliary diagnosis coefficient; The fetal sound auxiliary diagnosis coefficient threshold is calculated, and the specific formula is as follows: Tyfy=Tsxy 2 +Tjxy 2 +Ztcy; Among them, Tyfy is the threshold value of the fetal sound auxiliary diagnosis coefficient, Tsxy is the threshold value of the fetal heart deceleration monitoring coefficient, Tjxy is the threshold value of the fetal heart acceleration monitoring coefficient, and Ztcy is the threshold value of the average number of fetal heart beats in a period; Step S3233: If the fetal sound auxiliary diagnosis coefficient is greater than or equal to the fetal sound auxiliary diagnosis coefficient threshold, it is determined that the target fetus has a diagnosis abnormality, and a diagnosis abnormality warning is issued; Step S3234: If the fetal sound auxiliary diagnosis coefficient is less than the fetal sound auxiliary diagnosis coefficient threshold, it is determined that there is no diagnostic abnormality in the target fetus, and no diagnostic abnormality warning is issued.

10. A fetal sound monitoring and artificial intelligence-assisted diagnosis system, applicable to the fetal sound monitoring and artificial intelligence-assisted diagnosis method according to any one of claims 1 to 9, characterized in that: The auxiliary diagnosis system comprises: Image data module: obtain maternal and fetal medical images and multiple fetal medical images, create a fetal image recognition model through the fetal medical images, use the fetal image recognition model to recognize the maternal and fetal medical images, and classify the target fetus into a first type of diagnosis fetus and a second type of diagnosis fetus according to the recognition results, and obtain preliminary fetal diagnosis data; Fetal sound data module: obtain the fetal heart monitoring cycle, monitor the number of fetal heart beats of the second type of diagnosed fetus in the fetal heart monitoring cycle, obtain the average number of fetal heart beats in the cycle, monitor the fetal heart acceleration rate of the second type of diagnosed fetus in the fetal heart monitoring cycle, obtain the fetal heart acceleration monitoring coefficient, monitor the fetal heart deceleration rate of the second type of diagnosed fetus in the fetal heart monitoring cycle, obtain the fetal heart deceleration monitoring coefficient, and define the average number of fetal heart beats in the cycle, the fetal heart acceleration monitoring coefficient and the fetal heart deceleration monitoring coefficient as fetal sound monitoring data; Auxiliary diagnosis module: Perform auxiliary diagnosis on the target fetus based on the preliminary diagnosis data of the fetus and fetal sound monitoring data, and issue an early warning based on the diagnosis results.

Images