4 results about "Abdominal surface" patented technology

The invention discloses a uterine contraction detection method based on multi-modal signal fusion reconstruction, which comprises the following steps of: 1, synchronously acquiring an abdominal surface uterine electromyographic signal in real time through a high-density electrode array, acquiring a uterine contraction pressure signal in cooperation with a fundus pressure sensor, synchronizing a triaxial accelerometer signal to acquire a maternal motion signal, and preprocessing; 2, performing time sequence alignment and resampling on the data, and constructing a quality mask to form a multi-mode uterine contraction signal data set; 3, constructing a reconstruction network based on multi-modal feature fusion; 4, constructing a total loss function of the multi-modal feature fusion reconstruction network; and 5, training a reconstruction network of multi-modal feature fusion, wherein the reconstruction network is used for fusion reconstruction of the multi-modal uterine contraction data. According to the invention, through multi-modal signal fusion and deep feature correlation analysis, accurate monitoring and prediction of uterine contraction and premature delivery risks in the pregnant and birth period can be realized.

The present application belongs to the technical field of holographic breath analysis, and discloses a non-contact respiratory mechanics reverse reconstruction system based on 4D holographic field perception, which comprises the following steps: receiving the echo sequence of a millimeter wave sparse aperture array to the chest and abdominal area, generating a 4D point cloud sequence of the time-varying shape of the chest and abdominal surface according to frames, and constructing a respiratory observation sequence of the chest and abdominal surface; by analyzing the correlation between the length change attribute of the geodesic line and the energy change attribute of the curvature tensor evolution sequence, a respiratory driving feature is generated; according to the time change of the chest and abdominal area in the respiratory driving feature, an observation confidence is obtained; the respiratory driving feature is input into a reverse reconstruction network containing a nonlinear viscoelastic respiratory dynamics constraint, and a mechanical parameter set of airway resistance and pleural cavity pressure change is output. The present application converts 4D morphological observation into structured boundary excitation, and uses observation confidence to weight the constraint inversion process, enhances the complete expression of inhalation and exhalation driving, and improves the stability and reliability of the mechanical parameter output.

The invention belongs to the technical field of holographic respiration analysis, and discloses a non-contact respiratory mechanics reverse reconstruction system based on 4D holographic field perception. Comprising the following steps: receiving an echo sequence of a millimeter wave sparse aperture array to a thoracico-abdominal region, generating a 4D point cloud sequence of a time-varying form of the thoracico-abdominal surface according to frames, and constructing a respiration observation sequence of the thoracico-abdominal surface; by analyzing the relevance between the geodesic length change attribute and the curvature tensor evolution sequence energy change attribute, respiration driving characteristics are generated; obtaining an observation confidence coefficient according to the time change of the thoracic and abdominal region in the respiration driving characteristics; and inputting the respiration driving characteristics into a reverse reconstruction network containing nonlinear viscoelastic respiration dynamic constraints, and outputting a mechanical parameter set of airway resistance and pleural cavity pressure change. According to the method, 4D morphological observation is converted into structured boundary excitation, and the inversion process is weighted and constrained by observation confidence, so that the inspiration driving expression integrity is enhanced, and the stability and credibility of mechanical parameter output are improved.

The application relates to the field of acupuncture point positioning, in particular to an acupuncture point positioning method for acupuncture and physiotherapy. The method first acquires a plurality of acupuncture points in an abdominal surface image of a patient to be measured and initial coordinates of each acupuncture point, taking the Shenjue point of the patient to be measured as a coordinate origin, and uses a contour scanner to collect a plurality of contour lines of the abdominal surface of the patient to be measured along a body axis direction, analyzes a proportion between distances in the image and actual distances, obtains a proportion coefficient, screens a target contour line passing through a target acupuncture point according to an initial longitudinal coordinate of the target acupuncture point, a proportion coefficient, a moving speed of the contour scanner and a scanning frequency, analyzes a bending degree of a local part of the target acupuncture point on the target contour line, corrects the initial coordinate of the target acupuncture point according to the obtained bending degree, and obtains an adjusted coordinate of the target acupuncture point. The application can reduce deviation of acupuncture point positioning on the abdominal surface of the patient to be measured and improve the accuracy of the acupuncture point positioning.