40 results about "Late pregnancy" patented technology

The invention discloses a pregnant sow feed for increasing birth weight of piglets and improving placenta inflammation of sows, and relates to the technical field of biology. The pregnant sow feed comprises a lysozyme oligomer and a basic ration. The pregnant sow feed for increasing birth weight of piglets and improving placenta inflammation of sows provided by the invention is used for dynamically feeding sows in the pregnancy stage, meets the nutrition and energy requirements of the sows from the late pregnancy stage to delivery, and is fed for several times two days before delivery, so that the dystocia condition of the sows can be relieved, the stillbirth rate can be effectively reduced, and the survival rate of the sows is increased. The lysozyme content in colostrum and serum is increased, the anti-inflammatory factor expression of placenta is obviously promoted, the birth weight of piglets is obviously increased, the stillbirth rate is reduced, and the reproductive performance of sows is improved.

The invention relates to a middle-late pregnancy fetus ultrasonic cerebellum standard section image parallel recognition and measurement method and device, computer equipment, a storage medium and a computer program product, and the method comprises the steps: obtaining a middle-late pregnancy fetus ultrasonic image and a deep learning model of cerebellum standard section recognition, the deep learning model is obtained based on multi-node parallel technology accelerated training; inputting the fetus ultrasonic image in the middle-late pregnancy period into a deep learning model to identify a detection frame of each structure in a cerebellar section and a corresponding confidence coefficient; screening each structure in the recognized cerebellar section based on the detection frame and the corresponding confidence to obtain a structure recognition result in the cerebellar section; determining a cerebellum center line according to a structure identification result in the cerebellum section; and measuring the transverse diameter of the cerebellum according to the midline of the cerebellum. According to the whole scheme, efficient and accurate parallel recognition and measurement of the ultrasonic cerebellum standard section image of the fetus in the middle-late pregnancy period can be achieved.

The invention provides an optimal administration method for promoting cervical maturation by misoprostol based on a PopPK model, and belongs to the technical field of drug administration scheme optimization, the method comprises the following steps: constructing the PopPK model for oral administration of misoprostol by a pregnant woman in late pregnancy, the PopPK model is a model structure containing absorption lag time zero-order-first-order absorption parallel, one-atrioventricular distribution and first-order elimination, and the PopPK model is a model structure containing absorption lag time zero-order-first-order absorption parallel, one-atrioventricular distribution and first-order elimination; the PopPK model comprises two absorption chambers and one central chamber; and based on the PopPK model, determining a staged administration scheme through simulation optimization. The limitation of a traditional fixing scheme is effectively overcome, and the safety, effectiveness and clinical practicability of drug administration are considered.

The invention relates to the technical field of intelligent medical treatment and accurate risk management of obstetrics, in particular to a dynamic risk assessment system and method for venous thromboembolism of pregnant and lying-in women based on multi-modal data fusion. The system is composed of a multi-modal data acquisition module, a cloud data processing module and an early warning feedback module. The multi-modal data acquisition module is integrated with the micro-fluidic chip to rapidly detect blood coagulation markers, the genetic information analysis unit is used for screening gene mutation, and the intelligent pressure socks are used for monitoring hemodynamic parameters in real time; the cloud data processing module adopts a long-short-term memory network model, inputs dynamic feature sequences of continuous 30 days, and outputs VTE risk level probabilities in the future 7 days; the early warning feedback module pushes a grading intervention scheme; dynamic monitoring is implemented in early, middle and late pregnancy stages and postpartum stages, and dynamic risk updating is realized in combination with clinical rules and deep learning. Clinical tests show that the sensitivity, the specificity and the early warning timeliness of the system are remarkably improved, and the ICU transfer-in rate is reduced.

The application provides a misoprostol cervical ripening optimization dosing method based on a PopPK model, belongs to the technical field of drug dosing scheme optimization, and comprises the following steps: constructing a PopPK model of oral misoprostol for pregnant women in the third trimester of pregnancy, the PopPK model is a model structure containing zero-order-first-order absorption in parallel, one-compartment distribution and first-order elimination, the PopPK model comprises two absorption compartments and one central compartment; and determining a phased dosing scheme based on the PopPK model through simulation optimization. The method effectively overcomes the limitations of traditional fixed schemes, and takes into account the safety, effectiveness and clinical practicability of dosing.

The present application relates to the field of traditional Chinese medicinal materials, in particular to a composition for resisting PRRSv and improving the reproductive performance of sows and application thereof. The present application formulates a composition composed of 10 plant extracts (Angelica sinensis extract, Atractylodes macrocephala Koidz extract, Glycyrrhiza uralensis Fisch extract, Eucommia ulmoides Oliver extract, Eucommia ulmoides leaf extract, Populus davidiana Dode extract, Perilla frutescens leaf extract, Gynostemma pentaphyllum Makino extract, Leonurus japonicus extract and Astragalus membranaceus Bge extract) for sows in late pregnancy, and collects sow blood and placenta samples at the time of delivery, studies the influence of the composition on sow reproduction, and explores the mechanism of action. The experimental results show that the composition can improve the reproductive performance of sows by increasing the total number of piglets and other indicators; meanwhile, the composition can effectively inhibit the replication of PRRSv.

The utility model discloses a telescopic antenatal examination chair, and relates to the technical field of medical instruments. The device comprises a supporting plate, supporting legs are fixedly connected to the bottom of the supporting plate, a supporting shaft is rotationally connected to the rear end of the supporting plate, a back plate is fixedly connected to the surface of the supporting shaft, and an angle adjusting assembly used for controlling rotation of the supporting shaft is arranged on the bottom face of the supporting plate. During use, the effect of facilitating prenatal examination of a pregnant woman is achieved, meanwhile, the sitting function and the lying function are provided for the pregnant woman, the waist can be effectively supported in the sitting posture state, and therefore the discomfort of the waist of the pregnant woman is relieved, the difficulty of waist lifting cooperation of the pregnant woman is reduced during prostration examination, shoe taking-off operation is not needed, and the labor intensity of the pregnant woman is lowered. And the pregnant woman can conveniently lift the leg part, promote blood backflow, increase fetal blood supply, relieve the problem of edema of the pregnant woman in the late pregnancy, promote the health of the mother and the child, effectively reduce the horizontal position duration of the pregnant woman and reduce the risk of supine position hypotension syndrome.

The utility model relates to the technical field of nipple stickers, in particular to a nipple sticker capable of removing nipple scabs, which comprises a nipple sticker and a secondary throwing box, a groove is arranged in the middle of the inner wall of the nipple sticker, a first cavity and a second cavity are formed in the secondary throwing box through a partition plate, two placing cylinders are arranged in the first cavity, and the two placing cylinders are arranged in the second cavity. Ten paraffin oil nipple stickers are placed in the two containing barrels respectively, tweezers are placed in the second cavity, the nipple stickers can be repeatedly pasted through adhesive sticker parts, and the effect of repeated use is achieved. During use, the secondary throwing box is opened, the tweezers are used for clamping one paraffin oil nipple sticker and placing the paraffin oil nipple stickers into the grooves in the inner wall of the nipple stickers, and the nipple stickers are placed in the grooves in the inner wall of the nipple stickers. When the nipple patch is used, the nipple patch is adhered to the breast, the paraffin oil nipple patch is positioned at the nipple, a mother in late pregnancy can continuously or discontinuously wear the nipple patch, paraffin oil can soften and scabs, and then the nipple patch is lightly cleaned with warm water, so that the nipple patch is convenient to use, and meanwhile, the nipple can be protected and is prevented from being damaged.

The invention discloses a delivery time prediction method and system in combination with an ultrasonic image and clinical information, and relates to the field of medical image analysis and artificial intelligence, and the method comprises the steps: automatically extracting image features related to natural delivery time from an antenatal care ultrasonic image of a pregnant woman in the late pregnancy period; performing characterization processing on the clinical information of the pregnant woman to obtain clinical features related to natural delivery time; through a multi-modal fusion strategy, carrying out fusion processing on the image features and the clinical features to obtain fusion features with better delivery time predictability; and predicting the natural childbirth time of the pregnant woman based on the fusion features to obtain a natural childbirth time prediction result of the pregnant woman. Compared with a traditional expected date-of-confinement prediction method, the method combines the automatically extracted ultrasonic features with the clinical features, can more accurately predict the occurrence time of natural delivery, is beneficial to resource allocation of medical institutions, and is beneficial to alleviation of anxiety mood of pregnant women.

The invention discloses a system, method and device for premature delivery risk prediction and a storage medium, and belongs to the technical field of biomedical information. The system comprises a data acquisition module and a data processing device. The data acquisition module is used for acquiring hemoglobin concentration, serum ferritin concentration, age, BMI, pregnancy times, delivery times, marriage conditions, education degree and other clinical characteristic data of the pregnant woman in the early pregnancy period, the middle pregnancy period and the late pregnancy period; the input characteristics of the model comprise all data of the three stages of pregnancy; the risk score is then compared with a specific threshold value set based on a Youden index, and a risk level of preterm delivery, spontaneous preterm delivery, or iatrogenic preterm delivery is output. The Hb-SF-Clinical multi-feature prediction model is constructed by fusing the three-stage pregnancy dynamic biomarkers and clinical features, the AUC (early birth prediction) of the Hb-SF-Clinical multi-feature prediction model in a test set reaches up to 96.0%, the Hb-SF-Clinical multi-feature prediction model is remarkably superior to the prior art, high-precision and subtype early birth risk assessment is realized, and the clinical applicability is high.

The invention discloses an antenatal care bed with a massage relieving function, and belongs to the field of antenatal care cots.The antenatal care bed comprises an upper supporting plate and a fixed base, a plurality of circular grooves are formed in the upper surface of the upper supporting plate, a plurality of ventilation holes are formed in the upper surface of the upper supporting plate, and a rotating plate is connected to the inner bottom wall of each circular groove; according to the massage device, a heating module above a soft protection pad is electrified to generate heat, temperature adjustment is achieved through a controller, the massage head is connected to the upper surface of each rotating plate, the massage head is connected to the upper surface of each rotating plate, and a plurality of first motors are installed on the bottom face of the upper supporting plate. A stable and warm lying environment can be provided for a pregnant woman in autumn and winter or in a cold physique, the situation that the waist, the back and the abdomen are cold due to exposure during examination is avoided, local blood circulation of the waist and the back of the pregnant woman in the middle and late pregnancy period can be promoted through mild heat, waist and back muscle stiffness and aching pain caused by abdominal upheaval are relieved, and meanwhile the flexible heating film is attached to the body curve, so that the health of the pregnant woman is improved. Therefore, the safety of the fetus is guaranteed.

The present invention relates to the field of medical devices, and in particular to an amniotic fluid detection pad for premature rupture of membranes, comprising a pad body, wherein a water absorption layer, a dilution layer and an internal standard layer are sequentially arranged in the pad body from top to bottom; a first hollow cavity is arranged inside the dilution layer, a diluent is arranged in the first hollow cavity, a first isolation layer is detachably attached to the upper end of the first hollow cavity, and a second isolation layer is detachably attached to the lower end; a second hollow cavity is arranged inside the internal standard layer, and an internal standard liquid is arranged in the second hollow cavity; the first isolation layer is used to isolate the water absorption layer and the first hollow cavity, and the second isolation layer is used to isolate the first hollow cavity and the second hollow cavity. The present application can filter and collect liquid leaked by pregnant women in the late pregnancy, pretreat the subject's amniotic fluid with the diluent, and extract the diluted amniotic fluid with the internal standard liquid, thereby ensuring the purity of the extracted amniotic fluid and improving the accuracy of detection.

The invention discloses a safety auxiliary device for ultrasonic examination of fetal development, and relates to the field of ultrasonic examines.The safety auxiliary device comprises a base, a hydraulic telescopic rod is installed on one side of the top end of the base, a lifting rod is welded to the telescopic end of the hydraulic telescopic rod, and a motor base is welded to the end face of one side of the lifting rod; the device is scientific and reasonable in structure and safe and convenient to use, a pregnant woman is slowly rotated to a horizontal state from a vertical state, then ultrasonic detection is conducted, the auxiliary detection bed is rotated to a vertical state again after detection is finished, and therefore detection is completed, the whole ultrasonic detection process is accurate, the detection efficiency is high, and the detection efficiency is high. A pregnant woman, especially a pregnant woman in late pregnancy, does not need to stoop to sit on the detection bed and does not need to stand up after detection is finished, movement of the pregnant woman is facilitated, the detection efficiency is improved, a fetus is not extruded in the detection process, the fetus is prevented from being injured, the detection efficiency is further improved, and the safety of the pregnant woman is guaranteed.

The invention discloses an efficient oxytocic for hippocampus in the late pregnancy period and an oxytocic method of the efficient oxytocic. Research verifies that the odinagogue taking prostaglandin E2 as a main active component can rapidly induce the hippocampus to give birth in the late pregnancy (take effect within 60 minutes after odinagogue), and the odinagogue time is remarkably shortened. The juvenile fish produced by the method has independent viability, has no significant adverse effect on the sea horse with incomplete embryonic development, can effectively improve the mature fry acquisition rate when the embryonic development condition is unknown, reduces the fry loss in the culture process, and is beneficial to sea horse resource protection and industrial layout optimization.

The invention discloses a fetal giant child prediction method and system based on machine learning in the field of medical health of artificial intelligence, and the method comprises the following steps: S1, data collection: collecting ultrasonic examination data and clinical feature data of a late pregnancy (greater than or equal to 37 weeks) single-fetal pregnant woman, and constructing an original data set; the ultrasonic examination data comprises fetal biparietal diameter (BPD), femur length (FL), abdominal circumference (AC), head circumference (HC), amniotic fluid index (AFI) and ultrasonic weight estimation, and the clinical characteristic data comprises pregnant woman pre-pregnancy body weight index (BMI), pregnancy body weight increment (GWG), examination week of pregnancy (GA) and fetal gender. The giant child prediction method based on machine learning provided by the invention has significant clinical application value and technical advantages, can effectively improve the prediction ability of fetal giant child, promotes the intelligent and precise development of the obstetrical field, and has important practical significance for improving the health level of mothers and infants.

The application provides a high water-retention insoluble fiber and a preparation method and application thereof. The high water-retention insoluble fiber used in the application is derived from wood and wood chip processing by-products, and is obtained through a technical process of removing bark and cambium, cutting, crushing, high-temperature vaporization, pickling and drying, has a water-retention of 634-789%, an insoluble fiber component content of 84.32-88.24%, and a soluble fiber component content of 1.23-1.62%, and can be added into the feed of pregnant sows as a functional fiber raw material, so as to improve the constipation and empty chewing ratio of the pregnant sows, increase the number of live piglets and weaning litter weight of the sows and the mating rate in the next reproduction cycle, reduce the empty chewing ratio in the late pregnancy (100-102d) of the sows and the elimination rate in the next reproduction cycle, and has a wide application prospect in the daily feed of the pregnant sows.

The invention provides a dynamic progressive pregnancy and delivery prediction method and system, a storage medium and equipment, and relates to the technical field of smart medical treatment, and the method comprises the steps: obtaining baseline data and dynamic monitoring data of a pregnant woman, and dividing the dynamic monitoring data into collection indexes according to pregnancy stages; based on the baseline data and the dynamic monitoring data, calling a pre-constructed staged prediction model for prediction so as to generate a progressive prediction report; the staged prediction model comprises early, middle and late pregnancy sub-models, and each sub-model outputs a prediction result of a corresponding stage; in the prediction generation process, when a new pregnancy stage is entered, the corresponding sub-model is activated, and the prediction result of the previous stage is fused to generate a progressive prediction report. According to the technical scheme of the invention, personalized adjustment is carried out on the prediction basis for different pregnancy stages, algorithm adaptation is carried out for data characteristics of different pregnancy stages, and a composite prediction system covering the whole pregnancy is formed, so that the dynamic adaptability and accuracy of the prediction process are improved.

The application provides application of maternal blood exosome miR-1909-3p as a biomarker in preparation of a product for diagnosing or assisting in diagnosing fetal congenital heart disease, and belongs to the technical field of in vitro diagnosis. The application uses miRNA microarray analysis to analyze the expression characteristics of serum exosome microRNAs affected by CHD compared with matched healthy controls. In the early and late pregnancy of TOF fetus, miR-1909-3p is significantly overexpressed in the maternal circulation, indicating that miR-1909-3p is a potential biomarker for fetal congenital heart disease. Through ROC curve analysis of the verification set, it is found that the area under the curve is 0.953, P<0.001, the sensitivity is 95%, and the specificity is 95%, indicating that miR-1909-3p can be used as a potential non-invasive biomarker for prenatal CHD screening.

The utility model discloses a flexible and multi-purpose pregnant woman pillow which is characterized in that a pillowcase is provided with a first pillow cavity, a second pillow cavity and an adjusting cavity, the first pillow cavity is used for placing a first pillow inner, the second pillow cavity is used for placing a second pillow inner and a third pillow inner, and the distance between the two pillow inner is adjusted by the adjusting cavity; the third pillow inner is in a long and narrow long strip shape, the outer sides of the two pillow cavities are both subjected to opening binding through the opening binding assemblies, and when the thicknesses needed by the pregnant women in different pregnancy periods and early pregnancy periods are low, the thickness of the first pillow inner or the second pillow inner can be reduced by storing a part of contents through the first pillow inner or the second pillow inner extending storage strip; the first pillow inner and the second pillow inner are arranged in the first pillow inner, the third pillow inner is removed, pregnant mothers in the middle and later pregnancy can squeeze contents in the extended storage strip into the first pillow inner or the second pillow inner, meanwhile, the third pillow inner is used for supporting the abdomen and legs, the first pillow inner and the second pillow inner can be removed in the later lactation period, only the third pillow inner is reserved, and the pillow has multiple purposes.

The utility model relates to the technical field of fetal heart measuring equipment, and discloses a split type ultrasonic fetus-voice meter which comprises a host unit and a display screen unit which are arranged in a split mode, and the host unit is connected with the display screen unit in a wireless communication mode. And the host unit and the display screen unit are fixedly positioned and connected with each other in a magnetic attraction manner. A display screen unit carrying a display screen and a host unit carrying other parts are arranged in a split mode, the display screen unit and the host unit are combined together through magnetic force, and fetal heart signals of the host unit are transmitted to the display screen unit to be displayed in a wireless transmission mode. A user can hold the display screen unit or place the display screen unit at any position to check real-time data, fatigue caused by long-time head-lowering data checking of a pregnant woman user and inconvenience in operation in the middle and late pregnancy are avoided, user experience and monitoring efficiency are improved, and the problems that an ultrasonic fetus-voice meter in the prior art is prone to causing cervical vertebra fatigue, cable constraint and monitoring interruption are solved.

The utility model provides a girdle for a pregnant woman, which relates to the technical field of girdles and comprises spandex cloth, a bearing component is fixedly connected in the spandex cloth, a supporting component is fixedly connected on the outer side of the spandex cloth, a connecting component is rotatably connected at the bottom end of the outer side of the spandex cloth, and a bandage component is fixedly connected at the top end of the outer side of the spandex cloth. The bearing assembly comprises an elastic polyester fiber block, the outer side of the elastic polyester fiber block is fixedly connected to the two sides of the interior of spandex cloth, the outer side of the elastic polyester fiber block is fixedly connected with a pneumatic control valve, and the interior of the elastic polyester fiber block is fixedly connected with a protective air bag, so that the supporting strength can be adjusted according to the abdomen change of the pregnant woman; and the arc-shaped supporting elastic pieces and the spine supporting blocks help to adapt to body curves and maintain correct postures, so that discomfort and fatigue of the waist are effectively relieved, and the problem that static supporting of a traditional waistband is insufficient or excessive under the condition that the abdomen is enlarged at the late pregnancy period can be solved.

Embodiments include a computing device that executes a software routine and / or one or more machine learning architectures that provide a clinical predictive model to predict health complications caused by pregnancy. The prediction model may identify and enhance variables available in the electronic health recording system during the first three months of pregnancy and utilize a machine learning method to predict problematic outcomes during late pregnancy or postpartum. The predictive model follows several steps to identify possible pregnancy conditions, such as identifying data sources and experts for classifiers, checking data with clinical experts, applying statistical and machine learning methods, and evaluating model performance and interpretability. Predictive models and related methods provide unprecedented early detection of pregnancy complications for early intervention and treatment to improve health outcomes of mothers and children.

The invention discloses a feed additive for improving reproduction and lactation performance of sows in the late pregnancy-lactation period. The feed additive is composed of the following components in parts by weight: 1-4 parts of a cinnamon leaf extract, 1-4 parts of a rosemary extract, 10-20 parts of serine, 10-20 parts of N-carbamylglutamic acid, 40-60 parts of a probiotic composition and 10-20 parts of wheat bran. The probiotic composition is prepared from the following bacteria: 1.5 * 10 < 13 > to 4.5 * 10 < 13 > CFU / kg of bacillus subtilis, 1.5 * 10 < 13 > to 4.5 * 10 < 13 > CFU / kg of bacillus coagulans, 4 * 10 < 12 > to 6 * 10 < 12 > CFU / kg of clostridium butyricum and 5 * 10 < 12 > to 7 * 10 < 12 > CFU / kg of bacillus licheniformis.

According to the feed for improving the lactation performance and the immunity of the sows, soluble fibers and hyodeoxycholic acid are combined to feed the sows at the later stage of pregnancy, body fat metabolism of organisms is improved by improving serum biochemistry and colostrum milk fat level, and the lactation performance and the immunity of the sows are improved through mother-child transfer, so that healthy breeding of the local pigs is achieved, and the breeding period of the sows is shortened. A new thought is provided for improving filial generation health and development problems caused by poor lactation performance of the local pigs.

The invention relates to the technical field of medical monitoring equipment, in particular to an intelligent bracelet system which integrates multiple physiological function monitoring functions and can be networked with a medical institution information library in real time, the intelligent bracelet system comprises a bracelet, the bracelet comprises a wrist strap, a watch body, a display screen, a recorder and a patch, the watch body is internally provided with a Bluetooth module, a communication module and a sensor module, and the display screen is connected with the patch. A pressure sensor and a fetal heart sensor are arranged in the patch, the pressure sensor and the fetal heart sensor are electrically connected with the sensor module, pressure data are sent to a signed hospital through the communication module, and an operator on duty in the hospital sends the pressure data to a pregnant woman according to the pressure data of the pregnant woman and according to the condition that the pregnant woman is in the early pregnancy period or the late pregnancy period. And a predetermined scheme is quickly selected to provide help for the pregnant woman, so that unforeseen circumstances such as no threatened abortion or threatened premature birth due to the fact that the pregnant woman cannot see a doctor in time are avoided.

The invention relates to a fetal growth limitation intelligent prediction method based on a hidden Markov model, and belongs to the field of fetal growth limitation prediction. The method comprises the following steps: carrying out early, middle and late pregnancy physiological index data acquisition, cleaning and preprocessing; dividing the marked and classified early, middle and late pregnancy physiological index data set into a training set and a test set; taking the pregnancy period as a time sequence, and taking the physiological indexes under early, middle and late pregnancy as observation features to construct a feature sequence; constructing and training hidden Markov models for the normal fetal sample group and the FGR fetal sample group by using the training set to obtain a normal state HMM (HMM-Normal) and an FGR state HMM (HMM-FGR); and respectively calculating log-likelihood values generated with the HMM-Normal and the HMM-FGR by using the samples in the test set, and determining whether the samples have the FGR risk by comparing the likelihood values. According to the method, the hidden factors appearing in the FGR can be continuously analyzed subsequently through a deep learning method.

The application provides a supplementing concentrate for Mongolian horses in the late pregnancy, a supplementing method and application, and relates to the technical field of animal nutrition and feed science.The supplementing concentrate is composed of the following raw materials in parts by weight: corn 70-72%, soybean meal 6-9%, bran 8.7-16.2%, molasses 1-5%, vegetable oil 1-3%, stone powder 0.5-1%, calcium hydrogen phosphate 0.1-1%, lysine 0.1-1%, salt 0.1-1%, and premix 0.1-1%.The application systematically explores the influence of supplementing concentrates with different energy and protein levels on the growth performance, nutrient digestion rate, blood biochemical indexes, reproductive hormone level, antioxidant performance, immune function, fecal microbial flora and metabolomics of Mongolian horses in the late pregnancy, so as to provide a theoretical basis and data support for determining a scientific supplementing scheme of energy and protein in the daily ration of Mongolian horses in the late pregnancy.

The invention relates to a method and a device for judging and measuring a standard section of an epigastric ultrasound image of a fetus in the middle-late pregnancy period. The method comprises the following steps: acquiring a to-be-measured epigastric ultrasound image of the fetus in the middle-late pregnancy period and a deep learning model for identifying the standard section of the epigastric ultrasound image; inputting the upper abdomen ultrasound image of the fetus in the middle-late pregnancy period into the deep learning model, and identifying to obtain a detection frame of each target structure in the standard section of the upper abdomen ultrasound image and corresponding confidence; deleting the redundant detection frames according to the confidence coefficient and the interaction ratio between the detection frames, and obtaining the reserved detection frame corresponding to each target structure; scoring is carried out according to the target structure in the reserved detection frame and the corresponding weight, and a scored upper abdomen horizontal cross section is generated; according to the standard information of the epigastric ultrasound image of the fetus in the middle-late pregnancy period, obtaining an ROI region of the image, carrying out image segmentation, and generating a segmentation mask of the image; and carrying out abdominal girth profile curve fitting based on the segmentation mask, and measuring the abdominal girth length according to the fitted curve.

The invention discloses a preparation method and application of a late gestation feed for improving the survival rate of Hu sheep lambs, and belongs to the technical field of sheep breeding feeds, and the preparation steps are as follows: S1, preparation of silage mulberry: rubbing the whole-plant feed mulberry into shreds and crushing the shreds into 1-2cm in length, adding a complex microbial inoculant containing corn flour, cellulase and Lignin plum, and uniformly mixing to obtain silage mulberry powder; carrying out anaerobic fermentation; and S2, preparing a feed in the later stage of pregnancy, mixing the silage with the rest feed mixture, and adding the premix to prepare a feed finished product. Compared with a mulberry leaf type feed, the whole-plant mulberry leaf feed is used for silage fermentation, branches are not thrown away, the raw material utilization rate is increased, the labor cost is saved, meanwhile, after the prepared feed is used for feeding ewes in the breeding stage and the lamb lactation period, the abortion rate of the ewes is remarkably reduced, and the survival rate of the lambs is increased.