304 results about "Anatomical feature" patented technology

A robotic surgical system includes a robotic manipulator configured to perform surgical procedures under direct surgeon control. A surgical camera system captures real-time intraoperative video. An external imaging interface receives multimodal imaging data, including preoperative and intraoperative data from at least one of magnetic resonance imaging (MRI), computed tomography (CT), ultrasound, and fluoroscopy. An artificial intelligence (AI module has a trained neural network and a deep learning model trained on multi-institutional annotated surgical datasets, The AI module is configured to execute one or more of: fuse acquired video and imaging data into temporally and spatially coherent anatomical visualizations; generate continuously updating overlays aligned with the surgical field, with segmented anatomical features; projected tissue boundaries, proximity indicators for instruments, and predictive deformation trends; provide dynamic predictive trend visualization indicating zones of future anatomical complexity or risk; register and align preoperative imaging data with intraoperative imaging data in real time; adapt overlay presentation in response to tissue deformation without actuating the robotic manipulate or; and passively augment visual feedback without initiating any autonomous actuation of surgical instruments.

The invention relates to the technical field of tumor interventional operations, and discloses a visual navigation method based on a tumor interventional operation robot. The method comprises the following steps: acquiring real-time medical image data of a tumor area containing multi-modal imaging information so as to comprehensively present anatomical details; and performing three-dimensional reconstruction on the image data to generate a tumor area three-dimensional anatomical structure model capable of visually displaying a space structure. Key anatomical feature points are extracted based on the model, space coordinates are calculated, a surgical robot intervention path is planned according to the coordinates, and an initial navigation track is generated; and continuously collecting real-time pose data of the robot in an operation, dynamically matching the real-time pose data with the initial navigation trajectory, adjusting motion parameters according to a matching result, and generating a corrected navigation instruction. The method can reflect the intraoperative anatomy condition in real time, dynamically optimize the path, solve the problems that traditional navigation depends on preoperative static images and lacks real-time adjustment, reduce operative complications and improve the treatment effect of patients.

The invention provides an echocardiogram video abstraction method based on a hierarchical space-time attention mechanism, and the method comprises the steps: obtaining an echocardiogram, inputting the echocardiogram into an anatomy and physiological basic model of a key frame recognition task, and obtaining key frame information; inputting the key frame information into a multi-dimensional spatial-temporal feature extraction model to obtain feature information and frequency domain information; fusing the feature information and the frequency domain information; performing key frame enhancement processing on the fusion information by adopting an SE channel attention mechanism to obtain key frame anatomical feature expression; processing the key frame anatomical feature expression by adopting a key frame backtracking strategy and a classification network to obtain a key frame identification classification result; performing left ventricle segmentation on the classified image by adopting an improved U-Net left ventricle region segmentation network; estimating the area and volume of the left ventricle based on a segmentation result; fusing the estimated left ventricular area and volume by adopting a hierarchical attention mechanism and a cross-layer information fusion strategy to obtain a ventricular volume result; according to the method, through abstract extraction of the key information of the ultrasonic cardiac video, the clinical practicability of intelligent screening and auxiliary diagnosis of cardiovascular diseases based on ultrasonic is improved.

The invention discloses a diabetic foot early-stage detection method combining infrared and visible light imaging, and relates to the technical field of diabetic foot medical imaging diagnos.The diabetic foot early-stage detection method comprises the steps that multi-mode image collection and standardization processing are conducted, infrared images and visible light images are synchronously collected, and standardization processing such as size normalization is conducted; image registration and space alignment are carried out, marking points are set based on foot anatomical features, and feature extraction, mismatching point elimination and transformation matrix calculation are carried out; extracting and screening multi-dimensional features, extracting temperature and structural features, and screening by using a Relief-F algorithm; feature lesion recognition and classification are fused, and lesion probability is output through a double-branch convolutional neural network; carrying out detection result verification and feedback optimization, and comparing a clinical diagnosis optimization model; and generating a detection report and storing data, and generating a report containing the fused image. The early lesion detection precision is improved through multi-modal fusion, individual and environment differences are adjusted and adapted in a personalized mode, and reliable technical support is provided for clinic.

The invention discloses a diffuse large B-cell lymphoma segmentation system based on deep learning, and the system comprises a data collection module which is used for collecting PET and CT images of a patient; the double-branch encoder module is used for processing PET and CT images respectively; the cross-modal feature cross fusion module is used for extracting positioning clues from the CT anatomical features by taking the PET metabolic features as queries; taking CT structure characteristics as query, and screening PET high-metabolism regions; the global-attention suppression module is used for extracting global context features, performing down-sampling, remodeling input features into a low-dimensional matrix, then performing self-attention calculation, generating a sparse weight matrix for the input features based on a full connection layer, and then performing sparse attention calculation; and finally, fusing the two attention calculation results and outputting enhanced features. And the multi-scale jump connection module is used for fusing the encoder features and the decoder features and recovering spatial details.

The invention discloses a high-precision and automatic dental crown generation method. The method comprises seven steps of oral cavity three-dimensional data acquisition, full-mouth tooth semantic segmentation, upper and lower jaw point cloud precise registration, abutment and associated gingiva precise segmentation and trimming, abutment surrounding environment multi-tissue segmentation, neck-edge line secondary precise detection, and dental crown generation and multi-dimensional refinement. The method comprises the steps of oral cavity three-dimensional data acquisition, full-mouth tooth semantic segmentation, upper and lower jaw point cloud precise registration, abutment and associated gingiva precise segmentation and trimming, abutment surrounding environment multi-tissue segmentation, neck-edge line secondary precise detection and dental crown generation. Through combination of deep learning and a point cloud algorithm, high-precision segmentation of teeth and gingiva and precise reduction of an occlusion relationship are realized, an oral anatomical feature library and a clinical repair standard are fused, and through multi-dimensional fine adjustment optimization, a dental crown 3D model which is high in fitting degree, harmonious in occlusion and capable of meeting clinical requirements is generated. The method is full-process automatic, greatly improves efficiency, reduces operation threshold, reduces material waste and diagnosis and treatment cost, remarkably improves repair success rate, and is suitable for various oral repair scenes.

The invention relates to the technical field of medical instruments, and discloses a tracheotomy anatomical structure image recognition method and system, and the method comprises the steps: obtaining real-time visual image data and spatial positioning data of an endoscope, carrying out the time synchronization, carrying out the mode recognition of the spatial positioning data, and locking the remarkable anatomical features in the real-time visual image data, and according to the characteristics and the spatial positioning data, calculating a geometric transformation relationship between an endoscope visual coordinate system and a spatial positioning coordinate system, and finally fusing the data and displaying anatomical structure information and a surgical tool position in real time. The method can effectively solve the problem that in the prior art, an endoscope is irregularly deformed in the narrow trachea with physiological bending of a patient.

Systems and methods for fully automated anatomical analysis of an anatomical structure are provided to facilitate pre-operative planning. The computerized method may include obtaining a plurality of images, e.g., MSCT images, of patient-specific cardiovascular anatomy, and analyzing the MSCT images with a trained artificial intelligence module to identify one or more anatomical landmarks and to construct a virtual three-dimensional model of the anatomical structure. For example, the trained artificial intelligence module may execute segmentation, point detection, curve detection, or plane detection deep learning modules, independently or in combination, to identify the anatomical landmarks. The method further may include deriving anatomical measurements of the one or more identified anatomical landmarks, and displaying the virtual three-dimensional model alongside the anatomical measurements of the one or more identified anatomical landmarks.

The invention relates to the technical field of deep learning, and discloses a foot model generation method and system based on deep learning, and the method comprises the steps: collecting multi-angle foot image data of a user foot; inputting the multi-angle foot image data into a high-order embedded neural network for foot feature extraction to obtain foot feature point set data; inputting the foot feature point set data into a frequency domain diffusion guide neural network to execute image enhancement reconstruction of foot missing details to obtain enhanced foot image data; a point-line fusion neural network is utilized, foot contour points and anatomical feature lines are extracted from the enhanced foot image data at the same time, three-dimensional structure reconstruction is carried out, and foot point cloud model data are obtained.According to the method, differential modeling strategies are applied to different foot areas, the precision and the calculation efficiency are balanced, and the accuracy and the calculation efficiency are improved. The whole modeling process can be efficiently operated on a common smart phone and a cloud system, and the calculation complexity is greatly reduced.

A computer-implemented method for comparing medical images may include obtaining image data produced by multiple imaging devices indicative of anatomical features associated with an arthroplasty procedure, preprocessing the image data, including projecting three dimensional image data produced by a first imaging device of the multiple imaging devices to a plane represented by two dimensional image data produced by a second imaging device of the multiple imaging devices, identifying one or more landmarks associated with the anatomical features for comparison between images in the image data, comparing the identified one or more landmarks across the images to determine whether the images represent different individuals, and producing a notification indicative of a result of the determination of whether the images represent different individuals.

The invention relates to the field of remote quality control of kinesio, and provides a remote kinesio treatment guidance and quality control system. The image acquisition module is used for acquiring high-definition image data of a to-be-treated area through intelligent terminal equipment and marking anatomical feature points; wherein the high-definition image data comprises a static image and a dynamic muscle movement video stream; the AI analysis engine is used for carrying out feature extraction on the high-definition image data based on a muscle dynamics model of a to-be-treated region constructed by a convolutional neural network, identifying a muscle tension abnormal region and grading a nerve injury degree, and generating an initial sticking path recommendation scheme; the remote auditing interface transmits the recommended scheme generated by the AI to the doctor end interaction interface, obtains a target treatment scheme after the doctor end performs scheme correction based on the medical history data of the patient, and confirms a final treatment instruction through a digital signature; and the block chain quality control module is used for encrypting key operation data in the treatment process and then writing the key operation data into a distributed account book to generate a treatment process traceability chain which cannot be tampered.

Systems and methods are provided for multi-schema analysis of patient specific anatomical features from medical images. The system may receive medical images of a patient and metadata associated with the medical images indicative of a selected pathology, and automatically classify the medical images using a segmentation algorithm. The system may use an anatomical landmark detection algorithm leveraging Deep Reinforcement Learning (DRL) techniques to automatically locate one or more anatomical landmarks associated with the patient specific anatomical feature within the medical images. A 3D surface mesh model may be generated representing the patient specific anatomical features including the located one or more anatomical landmarks. The located one or more anatomical landmarks may be used to guide placement of a 3D model of a medical device that may be fused with the 3D surface mesh model to generate a patient specific 3D model of the medical device.

The invention discloses a visual interface display method and system for medical communication, and belongs to the technical field of medical communication visualization. The method comprises the following steps: acquiring an operation type, calling a corresponding scene template, synchronizing an image, completing three-dimensional reconstruction and generating an interactive anatomical model; main view real-time rendering is carried out, multi-dimensional data linkage display is realized, and a remote connection signal is obtained; when the remote connection is successful, allowing an expert to label the anatomical model; when abnormity is detected, risk early warning is triggered; calculating a network quality index, performing weighted calculation, and generating a network quality score; when the network quality score does not meet the communication condition, key anatomical feature points are extracted, and feature vectors are generated; traversing a feature database of the anatomical model, and matching with the feature vectors; positioning a focus area in the main view field, and executing area replacement logic; calling a medical network topology library, and screening alternative networks; and when no single network in the topological structure meets the requirement, starting a link aggregation mechanism.

The invention discloses a construction method of an obstructive sleep apnea feature recognition model. The obstructive sleep apnea feature recognition model is used for recognizing obstructive sleep apnea features, and comprises the following steps: collecting an apnea hypopnea index and a front face video of a subject; performing frame sampling and image enhancement on the front face video; performing automatic positioning and geometric normalization on face key points; constructing a weighted adjacency graph, and dividing mutually exclusive functional subnets; extracting multi-scale topological features based on the weighted adjacency graph and the functional subnet; splicing the multi-scale topological features, the facial anatomical features and the clinical variables to form comprehensive feature vectors, and screening the comprehensive feature vectors; and adopting an OSA risk prediction model to predict the probability of the individual suffering from obstructive sleep apnea. According to the method, screening can be completed in a non-invasive and low-cost manner under common illumination and natural postures.

The invention provides an ultrasonic probe sterile sleeve intelligent matching and image optimization method, which comprises the following steps: acquiring probe surface geometric data and examination part anatomical feature data through a three-dimensional scanning technology, generating a probe shape model and a part adaptation model, and obtaining probe shape adaptation parameters and examination part adaptation parameters; according to the shape adaptation parameters of the probe, adopting a finite element analysis method to simulate the fitting state of the sterile sleeve on the surface of the probe, calculating the fitting degree score of the sterile sleeve, and obtaining a fitting degree quantification result; a support vector machine algorithm is adopted, classification training is carried out on the multi-dimensional matching feature vectors, an intelligent recommendation model is generated, and sterile sleeve recommendation lists for different probes and examination parts are obtained; and if the fitness score of the matching feature vector is lower than a preset threshold value, optimizing geometric parameters of the sterile sleeve through a genetic algorithm, generating improved design parameters of the sterile sleeve, and obtaining an optimized matching precision result.

The invention discloses a brain high-resolution vascular wall image reconstruction system based on deep learning. The system comprises an image acquisition module, a frequency band decomposition module, a core structure recombination module, a frequency band detail enhancement module, a frequency band synthesis optimization module and a cerebrovascular wall image reconstruction module. The invention belongs to the field of image processing, and particularly relates to a brain high-resolution blood vessel wall image reconstruction system based on deep learning. According to the scheme, the reliability of subsequent cerebral vessel wall image reconstruction is improved through filtering processing of anatomical feature constraint, design of a grayscale adaptive basis function and dynamic adjustment of an effective interval based on an image grayscale histogram; gradient direction consistency constraint is introduced for core point discrimination, and non-core background interference of the blood vessel wall image is eliminated; on the basis of gray-scale adaptive second derivative weighting, detail enhancement and noise directional suppression of the blood vessel wall image are realized; and carrying out interpolation completion on the missing fine structure elements, and introducing a guide factor to adjust the weight, thereby improving the reconstruction stability of the cerebrovascular wall image.

The invention discloses an AI auxiliary radiotherapy dose intelligent optimization device, which comprises a multi-modal image fusion module, a dose distribution prediction module, a dynamic adaptive optimization module and a prognosis model integration module, and is characterized in that the multi-modal image fusion module is used for aligning anatomical features of different images, eliminating respiratory motion artifacts and predicting the dose distribution of the different images; the dose distribution prediction module is used for rapidly predicting radiotherapy dose distribution based on anatomical features and historical data, the dynamic adaptive optimization module is used for monitoring anatomical changes in real time and dynamically adjusting a dose plan, and the prognosis model integration module is used for quantifying correlation between dose distribution and radioactive injury risks. High-precision alignment and respiratory motion artifact elimination of an anatomical structure are achieved through the multi-modal image fusion module, three-dimensional dose calculation is rapidly and accurately conducted through the dose distribution prediction module, organ displacement and deformation are effectively coped with through the dynamic self-adaptive optimization module through a real-time image monitoring and reinforcement learning algorithm, and the accuracy of the three-dimensional dose calculation is improved. And the prognosis model integration module constructs an individualized risk prediction model.

The invention relates to the field of medical image processing, in particular to a cross-modal medical image alignment method based on anatomical feature one-dimensional distribution, and the method comprises the steps: obtaining a computed tomography (CT) image sequence and a magnetic resonance imaging (MRI) image sequence of a hip joint region of a patient; extracting a CT one-dimensional anatomical feature sequence representing the morphological change of the first side femoral head from the CT image sequence; extracting an MRI one-dimensional anatomical feature sequence representing the morphological change of the first side femoral head from the MRI image sequence; based on the CT one-dimensional anatomical feature sequence and the MRI one-dimensional anatomical feature sequence, a similarity alignment process is executed; wherein the similarity alignment process comprises the following steps: determining an optimal similarity alignment offset between a CT image sequence and an MRI image sequence, and aligning the MRI image sequence and the CT image sequence according to the optimal similarity alignment offset. According to the method, the cross-modal image alignment problem is subjected to dimensionality reduction into one-dimensional anatomical feature sequence matching, so that the precision, efficiency and robustness of registration are remarkably improved.

The invention relates to the technical field of augmented reality image processing, and discloses a traditional Chinese medicine acupuncture auxiliary positioning system based on an AR technology, which comprises a body surface anatomical feature intelligent extraction module, a personalized meridian model adaptive deformation module, a dynamic virtual-real registration tracking module and an AR visual rendering and quality closed loop module, according to the method, a thin-plate spline transformation algorithm is adopted to elastically deform a standard meridian model to the body surface of an individual, the individualized cun unit length is dynamically calculated in combination with a bone degree cun method, virtual-real registration is optimized through an iterative nearest point algorithm, movement and breathing deviation of a patient are compensated, the positioning quality is evaluated through a confidence coefficient thermodynamic diagram, closed-loop reacquisition optimization is driven, and the positioning accuracy is improved. Accurate AR-assisted positioning of acupuncture points of the whole body is realized.

The invention provides a non-paired graph unsupervised conversion method and system based on an anatomical fidelity Schrodinger bridge. The method comprises the following steps: acquiring a non-paired CBCT image set and a CT image set; based on a preset interpolation mechanism IPM, selecting an image sample from the CBCT image set for iterative processing, and generating an intermediate state sample; predicting the intermediate state sample through a generator network to obtain a synthesized CT image; based on the synthesized CT image and a real CT image randomly extracted from the CT image set, constructing a loss function to drive a training generator network; and converting the input CBCT image into a synthetic CT image by using the trained generator network. According to the method, the distribution difference between the CBCT image and the CT image is minimized by using the optimal transmission framework of entropy regularization, anatomical details are captured and fused under multiple scales, it is ensured that key anatomical features of the generated CT image are accurately reserved, and the accuracy of the CBCT image is improved. The method can effectively solve the problem that the modal conversion precision and the anatomical structure are lost due to lack of pairing data in an existing method.

The invention relates to the technical field of medical image processing and surgical navigation, in particular to an arthroscopic surgery real-time navigation method and system based on multi-modal image fusion, and the method comprises the following steps: S1, collecting preoperative CT (Computed Tomography) or MRI (Magnetic Resonance Imaging) data of a patient to generate a three-dimensional model, and collecting an intra-operative arthroscopic real-time video to provide input data for subsequent registration; and S2, performing image preprocessing, including denoising and enhancement, on the three-dimensional model and the arthroscope video, and extracting key anatomical features and lesion boundaries to obtain a processing result which can be used for registration. According to the method, the preoperative three-dimensional model and the intra-operative arthroscope real-time video are fused, and multi-strategy registration, closed-loop correction and dynamic updating are combined, so that continuous registration and real-time navigation of multi-modal data are realized, and the problems that the traditional arthroscope surgical navigation mostly depends on a single-modal image, and the navigation accuracy is low are solved. And the problem of insufficient intraoperative registration precision caused by lack of multi-source information fusion is solved.

The invention provides a physical manipulation apparatus that is engageable with a skin surface of a human or animal subject. The apparatus may include a skin-engageable pad configured to conform to the morphology of a physical manipulation member such as a digit of a human therapist or a stylus tip of a chiropractic adjustment instrument. The skin-engageable pad may include a friction-inducing component in the form of a biomimetic dry adhesive. The apparatus may include a digit cover (110, 120, 130) configured at least partially to cover a human digit. Instead, the apparatus may be configured as a skin patch having two release liners superimposed, respectively, on opposite sides of the skin-engageable pad. The skin patch may include indicia arranged to indicate a required orientation of the skin patch relative to a targeted anatomical feature such as a spinal segment, or a required direction of movement of the manipulation member.

The invention relates to the technical field of knee joint brace design, and discloses a knee joint brace personalized design method based on 3D printing. The method comprises information of a skeleton structure, a soft tissue contour and a joint motion range. Determining a personalized anatomical feature set of the knee joint by combining a preset knee joint biomechanical knowledge base on the basis of the three-dimensional scanning data; according to the personalized anatomical feature set and the corresponding at least one design rule file, dividing the stereo scanning data into a macroscopic structure region and a microscopic detail region; the macrostructure area outputs skeleton frame data, and the data is transmitted to the microcosmic detail area for guiding soft tissue matching analysis; constructing a three-dimensional model of the knee joint brace based on a division result; and converting the three-dimensional model into a 3D printing instruction file which is used for driving 3D printing equipment to produce the personalized knee joint brace. According to the method, personalized manufacturing of the knee joint brace is achieved through multi-dimensional data fusion and structural collaborative design.

The invention discloses a multi-scale anatomical prior and attention-enhanced liver lesion segmentation method, which comprises the following steps of: training a three-dimensional convolution denoising auto-encoder by using an unsupervised training data set of an original image to obtain a trained denoising auto-encoder model; inputting a supervised training data set of an original image into the trained denoising auto-encoder model to obtain prior anatomical features of each image at four different resolution levels; inputting the prior anatomical features and the original image into a 3D attention enhancement segmentation network for network training output to obtain a trained segmentation model; and obtaining a liver focus segmentation result of the three-dimensional image of the to-be-segmented liver region through the trained denoising auto-encoder model segmentation model. Dependence on labeled data is reduced through a semi-supervised learning strategy, global context construction of a focus area is enhanced by introducing a hierarchical three-dimensional attention mechanism, and segmentation precision, robustness and clinical generalization ability of a tiny, low-contrast and boundary fuzzy focus are improved.

A method for tracking patient features in a surgical field includes capturing an image stream in a field of view demonstrating patient features of a patient anatomy. A first feature set is identified in a tracking region of the field of view of the image stream. The relative motion among a first plurality of the image features in the first feature set is monitored in the image stream over time. Moving features from the first features set are filtered to generate a second feature set. The second feature set defines tracking features. The location and orientation of the patient anatomy is then tracked based on the feature locations of the plurality of tracking features.

A surgical navigation system is provided, which includes a handheld surgical instrument configured to receive an end effector, a localizer configured to determine a pose of the handheld surgical instrument, an alert module for providing notifications to a user, and a control system in communication with the localizer and the alert module. The control system configured to receive a selection of the end effector, associate an end effector model with the handheld surgical instrument based on the selection of the end effector, receive a selection of a security margin, modify a spatial dimension of the end effector model based on the selected security margin, determine a pose of the modified end effector model, determine a pose of a patient boundary associated with a target anatomical feature, and control the alert module based on the pose of the patient boundary and the pose of the modified model of the end effector.

The invention belongs to the technical field of medical image processing, and particularly relates to a medical image sequence positioning method and system. The method comprises the steps that a medical image sequence is acquired and preprocessed, wherein the medical image sequence comprises a to-be-positioned target anatomical structure; inputting the preprocessed sequence into a pre-trained positioning model frame by frame to obtain a space attention map of the target anatomical structure in each frame and a corresponding time phase; and aggregating the space attention map of each frame in the medical image sequence and the corresponding time phase, and generating a space-time positioning report of the target anatomical structure. According to the method, the anatomical features and the phase features are explicitly decoupled, the robustness to complex clinical scenes is remarkably enhanced, and the positioning accuracy is effectively improved.

A computer-implemented method of processing repeat B-scans of an imaged region of a body part, the imaged region including an anatomical feature in a first subregion and vasculature of interest in a second subregion of the imaged region, to generate cropped B-scans for use in generating OCT angiography data which provides a representation of the vasculature of interest, the method comprising processing each B-scan of the repeat B-scans by: selecting a subset of the data elements of the B-scan such that the selected data elements are distributed along a representation of the anatomical feature in the B-scan; using the selected data elements to define a respective subregion of interest in the B-scan, which includes OCT data acquired from the second subregion containing the vasculature of interest; and cropping the B-scan to leave the subregion of interest in the B-scan.

The invention relates to a medical beauty implant material intelligent optimization method and system based on artificial intelligence, and the method comprises the following steps: S1, obtaining the performance parameters of different medical beauty implant materials, capturing the anatomical features of a patient through 3D scanning and MRI images, and carrying out the statistics of the personalized biological data of the patient; s2, constructing a multi-task learning model and training based on a multi-target supervised learning data set, wherein the target is to predict the performance of different implant material formulas in a medical beauty implant environment; s3, performing multi-objective optimization by adopting an NSGA-III algorithm to obtain an optimal implant material formula; s4, reconstructing a three-dimensional model of the patient by using a reverse modeling technology based on anatomical features of the patient; and S5, based on the three-dimensional model, according to the optimal implant material formula and the personalized biological data, simulating the biomechanical characteristics of the patient by utilizing finite element analysis, and outputting biocompatibility scores and suitability suggestions of the material in different implant environments. According to the method, the clinical requirements on customized design and application of the medical beauty implant material are effectively met.

The invention discloses an ultrasonic robot inspection quality adaptive optimization method, which relates to the field of ultrasonic robots, and comprises the following steps: acquiring anatomical structure feature information of a to-be-inspected part, and setting an initial working frequency and an initial focusing depth of an ultrasonic probe of an ultrasonic robot based on the anatomical structure feature information; an ultrasonic probe of the ultrasonic robot is controlled to execute scanning operation according to the initial working frequency and the initial focusing depth, ultrasonic images are collected in real time, quality quantitative indexes of the images are extracted, and current probe posture data of the ultrasonic robot in the scanning process are synchronously obtained; the initial working parameters of the ultrasonic probe are set by obtaining the anatomical structure characteristics of the to-be-inspected part, the initial scanning pertinence is improved to reduce invalid operation, the ultrasonic image quality quantitative index and the probe posture data are collected in real time, the scanning state can be grasped in time, the dynamic quality threshold value is set in combination with the anatomical characteristics, and the detection accuracy is improved. And the image quality judgment is more fit with the actual anatomy condition to avoid the judgment deviation of a unified threshold value.