Search

CN-122025097-A - Bone microstructure change real-time tracking decision management system

CN122025097ACN 122025097 ACN122025097 ACN 122025097ACN-122025097-A

Abstract

The invention belongs to the field of artificial intelligence, and provides a bone microstructure change real-time tracking decision management system which comprises an image data access module, a dynamic structure analysis unit and an interactive decision output module, wherein the image data access module is used for receiving bone three-dimensional time sequence data streams from high-resolution microscopic imaging equipment and clinical imaging equipment in real time, and the dynamic structure analysis unit comprises a bone microstructure evolution identification module, a mechanical property prediction module and an intervention strategy generation module. The invention realizes the full-chain closed-loop management from microstructure perception, mechanical property deduction to intelligent intervention decision-making through the cooperative work of the image data access module, the dynamic structure analysis unit and the interactive decision-making output module, and remarkably improves the early warning and accurate intervention capability of bone diseases.

Inventors

  • LU HAITAO
  • LU JINZHI
  • LIU YULIN

Assignees

  • 东莞市呈夕可健康管理咨询有限公司

Dates

Publication Date
20260512
Application Date
20260129

Claims (10)

  1. 1. The real-time tracking decision-making management system for the bone microstructure change is characterized by comprising an image data access module, a dynamic structure analysis unit and an interactive decision-making output module; The image data access module is used for receiving bone three-dimensional time sequence data streams from the high-resolution microscopic imaging equipment and the clinical imaging equipment in real time; The dynamic structure analysis unit comprises a bone microstructure evolution identification module, a mechanical property prediction module and an intervention strategy generation module; The bone microstructure evolution recognition module is used for extracting bone trabecula thickness, bone trabecula spacing, porosity and connection density parameters based on bone image data of continuous time points, and recognizing abnormal evolution trend of a microstructure through a space-time diagram neural network model; The mechanical property prediction module is used for predicting the thickness of the bone trabecula Spacing of bone trabeculae Connection density By the formula: ; calculating the equivalent elastic modulus at time t Wherein The learning weight coefficient is obtained through clinical data training; the intervention strategy generation module is used for calling a bone metabolism intervention rule base according to the mechanical property result output by the mechanical property prediction module and the structure evolution trend output by the bone microstructure evolution identification module to generate nutrition supplement, medicine intervention and mechanical stimulation schemes; The interactive decision output module is used for visually displaying skeleton microstructure evolution tracks, current mechanical property evaluation results and recommended intervention strategies, and the integrated virtual health assistant interprets risk levels and countermeasures to a user in a natural language interactive mode.
  2. 2. The bone microstructure variation real-time tracking decision management system according to claim 1, wherein the bone three-dimensional time sequence data stream received by the image data access module is derived from at least one of micro-CT, confocal microscope and dual-energy X-ray absorber DXA, and the image data access module further comprises an image preprocessing unit for performing at least one of the following operations of format unification, motion artifact correction, gray scale normalization, automatic bone tissue segmentation and time sequence registration on the received original image data to generate standardized three-dimensional time sequence bone structure data.
  3. 3. The bone microstructure variation real-time tracking decision management system as claimed in claim 1, wherein the bone microstructure evolution recognition module extracts bone microstructure parameters including bone trabecular number The bone microstructure evolution recognition module further comprises a bone seam analysis submodule, wherein the bone seam analysis submodule is used for carrying out fine segmentation and parameter extraction on a bone seam region in an image, the bone seam analysis submodule comprises a bone seam width SW (t), an edge bone density EBD (t) and a healing symmetry index SI (t), and the parameters and bone fragments Liang Canshu are input into an enhanced space-time diagram neural network together to realize joint dynamic modeling of a bone seam and a bone trabecula structure.
  4. 4. The bone microstructure variation real-time tracking decision-making management system according to claim 1, wherein the space-time diagram neural network model comprises a diagram construction layer, a space-time convolution layer, an attention mechanism layer and an output layer; the map construction layer uses bone microstructure parameters of each time point as node characteristics and uses time step and parameter correlation as edge weight to construct a dynamic multi-relation map; The space-time convolution layer adopts a graph convolution network to extract space characteristics and combines a time convolution network or a cyclic neural network to capture time sequence dependence; the attention mechanism layer draws time attention and node attention, and focuses on time points and key structural parameters which are obviously changed; And the output layer outputs a structural degradation index, an abnormal evolution trend label and a future evolution prediction.
  5. 5. The bone microstructure variation real-time tracking decision management system according to claim 1, wherein the mechanical property prediction module is further configured to Calculation of compressive Strength The calculation relation is as follows: ; Wherein the method comprises the steps of And Is the material constant calibrated by in vitro compression test.
  6. 6. The bone microstructure variation real-time tracking decision management system according to claim 1, wherein the bone metabolism intervention rule library called by the intervention policy generation module is stored in a bone metabolism knowledge graph constructed based on an ontology, and the knowledge graph comprises diseases, symptoms, medicines, nutrients, exercise types, examination items, crowd labels, treatment relationships, tabu relationships, enhancement relationships, side effect relationships and evidence-based grade relationships Logical interventions generate rules that support multi-hop reasoning and collision detection.
  7. 7. The bone microstructure variation real-time tracking decision-making management system according to claim 6, wherein the intervention strategy generation module combines the user's age, gender, race, menopausal status, renal function, liver function, vitamin D level, fracture history, medication history, allergy history, exercise habits, dietary structure, smoking drinking history, and professional load type when generating the intervention plan.
  8. 8. The bone microstructure variation real-time tracking decision-making management system according to claim 1, wherein the visual interface of the interactive decision-making output module comprises a left area displaying a graph of bone microstructure parameters over time, a middle area displaying a current value of elastic modulus and compressive strength and a percentile thereof in a normal population distribution, and a right area listing recommended intervention strategies and evidence-based grades thereof.
  9. 9. The bone microstructure variation real-time tracking decision management system of claim 1, wherein said virtual health assistant comprises a natural language understanding module, a knowledge retrieval module, a text generation module, and a speech synthesis module; the natural language understanding module is used for analyzing user intention and entity based on the pre-training language model; the knowledge retrieval module is used for retrieving related information from the output of the dynamic structure analysis unit and the bone metabolism knowledge graph; The text generation module is used for generating colloquially and emotionally adapted interpretation text and supporting multiple rounds of dialogue and inquiry guidance; The voice synthesis module is used for converting the text into natural voice output and supporting multilingual and dialect adaptation.
  10. 10. The bone microstructure variation real-time tracking decision management system of claim 1 wherein said predetermined threshold is a composite degradation index derived from a weighted fusion of a plurality of bone microstructure parameters.

Description

Bone microstructure change real-time tracking decision management system Technical Field The invention belongs to the field of artificial intelligence, and particularly relates to a real-time tracking decision-making management system for bone microstructure change. Background In the fields of orthopaedics medicine, osteoporosis research and bone health monitoring, the method for accurately and real-timely tracking and evaluating the dynamic change of the bone microstructure has important clinical and scientific research values. Bone is used as a highly dynamic tissue, and the microstructure (such as bone trabecular thickness, porosity, connection density and the like) of the bone can continuously change along with factors such as age, disease progression, pharmaceutical intervention or mechanical load and the like; The traditional evaluation means mainly depend on imaging technologies such as a dual-energy X-ray absorption method, high-resolution peripheral quantitative CT or micro CT, and the like, although the methods can provide more accurate static structural information, the problems of radiation exposure, high cost, incapability of continuous real-time monitoring, data processing lag and the like are generally existed, the timely intervention requirement of clinic on a dynamic evolution process is difficult to meet, in addition, the traditional system focuses on image acquisition and post-processing, and the capability of integrating and analyzing multi-source image data, biomechanical parameters and clinical indexes and generating intelligent decision support based on the multi-source image data and biomechanical parameters is lacking. For this reason, a real-time tracking decision management system for bone microstructure changes is proposed by those skilled in the art to solve the problems presented in the background art. Disclosure of Invention In order to solve the technical problems, the invention provides a bone microstructure change real-time tracking decision management system to solve the problems in the prior art. A real-time tracking decision management system for bone microstructure change comprises an image data access module, a dynamic structure analysis unit and an interactive decision output module; The image data access module is used for receiving bone three-dimensional time sequence data streams from the high-resolution microscopic imaging equipment and the clinical imaging equipment in real time; The dynamic structure analysis unit comprises a bone microstructure evolution identification module, a mechanical property prediction module and an intervention strategy generation module; The bone microstructure evolution recognition module is used for extracting bone trabecula thickness, bone trabecula spacing, porosity and connection density parameters based on bone image data of continuous time points, and recognizing abnormal evolution trend of a microstructure through a space-time diagram neural network model; The mechanical property prediction module is used for predicting the thickness of the bone trabecula Spacing of bone trabeculaeConnection densityBy the formula: ; calculating the equivalent elastic modulus at time t WhereinThe learning weight coefficient is obtained through clinical data training; the intervention strategy generation module is used for calling a bone metabolism intervention rule base according to the mechanical property result output by the mechanical property prediction module and the structure evolution trend output by the bone microstructure evolution identification module to generate nutrition supplement, medicine intervention and mechanical stimulation schemes; The interactive decision output module is used for visually displaying skeleton microstructure evolution tracks, current mechanical property evaluation results and recommended intervention strategies, and the integrated virtual health assistant interprets risk levels and countermeasures to a user in a natural language interactive mode. Further, the skeleton three-dimensional time sequence data stream received by the image data access module is derived from at least one device of micro-CT, confocal microscope and dual-energy X-ray absorber DXA, and the image data access module further comprises an image preprocessing unit for performing at least one operation of uniform format, motion artifact correction, gray level normalization, automatic bone tissue segmentation and time sequence registration on the received original image data to generate standardized three-dimensional time sequence skeleton structure data. Further, the bone microstructure parameters extracted by the bone microstructure evolution recognition module further comprise the number of bone trabeculaeThe bone microstructure evolution recognition module further comprises a bone seam analysis submodule, wherein the bone seam analysis submodule is used for carrying out fine segmentation and parameter extraction on a bone seam region in a