Tumor genetic family diagram drawing method

Abstract

The invention relates to a method for drawing a tumor genetic family chart. The method comprises the following steps: collecting basic information, tumor medical history and death information of relatives in a plurality of generations of families of consultants; defining family members by using a unique member ID, and standardizing the collected data by using the member ID as a main key to obtain a family member database; on the basis of the family member database, by taking the consultant as a starting point, upwards decreasing the generation coefficient and downwards increasing the generation coefficient, polling genetic relationships of all family members and matching to obtain corresponding family member nodes, and establishing a relative association network of a tree structure; and traversing the relative association network to draw each family member node, identifying the family member nodes with the tumor medical history, and drawing a relationship line between the nodes according to the genetic relationship to obtain a tumor genetic family diagram of the consultant. According to the method, the tumor genetic family diagram can be automatically drawn according to the intelligent interaction content with the consultant, and a powerful reference is provided for later tumor screening and risk assessment.

Description

Technical Field

[0001] The present invention relates to the field of smart medical technology, and in particular to a method for drawing a tumor genetic pedigree diagram. Background Art

[0002] Oncology consultation and genetic pedigree charting rely heavily on specialized physicians, requiring systematic training, mastery of basic consultation theory, communication skills, and years of specialized knowledge in gynecological oncology. While some computer-assisted diagramming is currently available, it still requires physicians and consultants to spend significant time together collecting medical histories, conducting inquiries, and creating diagrams. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a method for drawing a tumor genetic pedigree chart, which can automatically draw a tumor genetic pedigree chart based on the intelligent interaction content with the consultant, providing a powerful reference for later tumor screening and risk assessment.

[0004] The technical solution adopted by the present invention to solve the technical problem is to provide a method for drawing a tumor genetic pedigree diagram, comprising the following steps:

[0005] Collect basic information, cancer history and death information of relatives within several generations of the consultant's family;

[0006] Define family members with unique member IDs, and standardize the collected data into a family member data structure with member ID as the primary key to obtain a family member database;

[0007] Based on the family member database, starting from the consultant, the generation coefficient is decreased upwards and increased downwards, the kinship of all family members is polled and the corresponding family member nodes are matched to establish a tree-structured kinship network;

[0008] Traverse the kinship network to draw out each family member node, identify the family member nodes with a history of cancer, and draw relationship lines between nodes based on kinship to obtain the consultant's cancer genetic pedigree.

[0009] Furthermore, polling the kinship of all family members and matching corresponding family member nodes to establish a kinship network with a tree structure includes:

[0010] Create a set of nodes to be matched and initialize them to all family member nodes;

[0011] Establish a node set for each generation and move the consultant from the node set to be matched to the node set of this generation;

[0012] According to the order of the consultant's spouse, the consultant's children, the consultant's siblings and their spouses, the consultant's siblings' children, the consultant's parents, the consultant's parents' siblings, and the consultant's parents' parents, the family member nodes that match the current kinship relationship are retrieved from the set of nodes to be matched. During the retrieval, the family members are matched according to the arrangement order of the family members in the node set of the corresponding generation, and the matched family member nodes are moved to the node set of the corresponding generation according to the matching order to obtain a tree-structured kinship network.

[0013] Furthermore, the process of searching for family member nodes that match the current kinship relationship from the set of nodes to be matched is carried out in the order of the consultant's spouse, the consultant's children, the consultant's siblings and their spouses, the consultant's siblings' children, the consultant's parents, the consultant's parents' siblings, and the consultant's parents' parents, and matching the family member nodes according to the order in which the family members are arranged in the node set of the corresponding generation during the search, and moving the matched family member nodes to the node set of the corresponding generation in the matching order, including:

[0014] Search for the spouse and children of the consultant from the set of nodes to be matched, move the spouse to the current generation node set and place it to the left of the consultant, and move the children to the child generation node set;

[0015] Find the consultant's siblings and the spouse of each sibling from the set of nodes to be matched, and move each sibling and their spouse to the current generation node set adjacent to each other and place them to the right of the consultant;

[0016] Traverse each sibling of the consultant in the current generation node set, search for the children of each sibling in the to-be-matched node set in order, place the family member nodes that are siblings of each other adjacently, and move the children of each sibling to the descendant node set in the search order and place them to the right of the consultant's children;

[0017] Find the parents of the consultant from the set of nodes to be matched, move their parents to the set of parent nodes and place the mother to the right of the father;

[0018] Find the siblings of both parents of the consultant from the set of nodes to be matched, move the siblings of the father to the parent node set and place them on the left side of the father, and move the siblings of the mother to the parent node set and place them on the right side of the mother;

[0019] The parents of both parents of the consultant are searched from the set of nodes to be matched, and their grandparents and maternal grandparents are moved to the ancestor node set in the search order.

[0020] Furthermore, when drawing a tumor genetic pedigree, the distance between any two adjacent generations is set to a fixed value.

[0021] Furthermore, when drawing a tumor genetic pedigree diagram, the node corresponding to the consultant is used as the benchmark, and the node spacing is dynamically allocated according to the number of nodes of family members of the same generation.

[0022] Furthermore, the family member data structure includes a member ID, a basic information field, a tumor history field, a death information field and a kinship field; the kinship field includes a spouse member ID, a child member ID, a parent member ID and a sibling member ID.

[0023] Furthermore, the family member node adopts the family member data structure.

[0024] Furthermore, the collection of basic information, tumor medical history and death information of relatives within several generations of the consultant's family is achieved by constructing an AI intelligent body to simulate a doctor to conduct automated medical interviews with the consultant.

[0025] Furthermore, during automated medical consultations, the AI agent dynamically analyzes whether the consultant's relatives have related disease factors through hybrid retrieval technology combined with a customized genetic knowledge base, and conducts in-depth medical consultations on the relative branches with related disease factors based on the analysis results.

[0026] Furthermore, the family member database is obtained by the AI agent through large-model JSON structured processing and Function Calling technology, combined with prompt word engineering technology, to understand professional tumor information from natural text information and extract effective fields that are conducive to structuring, and assemble them into JSON structured data for specified gynecological tumors.

[0027] Beneficial effects

[0028] Due to the adoption of the above-mentioned technical solution, the present invention has the following advantages and positive effects compared with the prior art: the present invention constructs a family member database and establishes a tree-structured kinship network based on the family member database. When constructing the kinship network, the present invention further matches family member IDs through a customizable node retrieval order and hierarchical node sorting strategy on the basis of traditional multi-level family data management. Then, the kinship network is drawn into a pedigree chart using pedigree symbols and relationship lines that meet the standards of tumor genetics, achieving rapid and automatic generation of pedigree charts and greatly improving the efficiency of medical consultations. When constructing the kinship network, the search order is designed, and the family members are matched according to the order of their arrangement in the node set of the corresponding generation during the search, and the obtained member IDs are moved to the node set of the corresponding generation in the matching order, so that the constructed tree structure is more in line with the requirements of drawing the pedigree chart, thereby simplifying the process of drawing the pedigree chart. In addition, by constructing an AI intelligent agent, machine learning is combined with large models, natural language processing and other technologies, so that the intelligent agent with a large model as the core can quickly reach the ability level of professional consulting physicians. When interacting with patients, the AI intelligent agent can identify key medical terms (such as "breast cancer", "ovarian cancer", "immediate relatives", etc.), and based on intent recognition and entity extraction technology, it automatically generates the next inquiry strategy or guiding questions to guide patients to provide family information. It can efficiently, comprehensively and accurately obtain family tumor genetic data, providing a better data foundation for drawing pedigree charts. Through self-learning and iterative training, the large model quickly mastered the questioning logic and diagnostic ideas of professional consulting physicians during consultations. The intelligent agent built with this as the core can efficiently obtain family information closely related to tumor genetics. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 is a flow chart of an embodiment of the present invention;

[0030] Figure 2 This is a schematic diagram of the process of constructing and applying a family member database according to an embodiment of the present invention;

[0031] Figure 3 This is a pedigree chart according to an embodiment of the present invention. DETAILED DESCRIPTION

[0032] Below in conjunction with specific embodiment, further set forth the present invention.Should be understood that these embodiments are only used to illustrate the present invention and are not used in limiting the scope of the present invention.In addition, should be understood that after reading the content taught by the present invention, those skilled in the art can make various changes or modifications to the present invention, and these equivalent forms fall equally within the scope limited by the appended claims of the application.

[0033] The embodiment of the present invention relates to a method for drawing a tumor genetic pedigree diagram, such as Figure 1As shown, the following steps are included:

[0034] Collect basic information, cancer history and death information of relatives within several generations of the consultant's family;

[0035] Define family members with unique member IDs, and standardize the collected data into a family member data structure with member ID as the primary key to obtain a family member database;

[0036] Based on the family member database, starting from the consultant, the generation coefficient is decreased upwards and increased downwards, the kinship of all family members is polled and the corresponding family member nodes are matched to establish a tree-structured kinship network;

[0037] Traverse the kinship network to draw out each family member node, identify the family member nodes with a history of cancer, and draw relationship lines between nodes based on kinship to obtain the consultant's cancer genetic pedigree.

[0038] Among them, data collection can be achieved by building an AI agent to simulate a doctor to conduct automated medical consultations with the consultant. Based on the large model, the AI agent can collect family information through dialogue like a doctor. For example: "Are your parents alive? Do they have a genetic disease?", "Do you have brothers and sisters? How is their health?" Intelligent guidance fills in missing information, for example: if the user mentions that "both grandparents are dead", the AI can further ask about the age of death, whether they died of a genetic disease, etc. Based on the family relationship model of the knowledge graph, the AI agent can automatically identify and supplement family data, for example: determine that the "second uncle" should belong to the level of "father's brother", while the "cousin" belongs to the maternal relative.

[0039] More specifically, AI responds to patient conversation requests, determines patient intent through inquiries, conversations, and semantic analysis, and determines whether the service content is relevant, such as: consultation on routine tumor issues, tumor risk screening assessment, and tumor pedigree charting, and guides patients to the next step of the service process.

[0040] To enable AI to quickly reach the capabilities of professional consulting physicians and better understand patient intent, a targeted oncology knowledge base can be constructed for AI retrieval, providing robust professional documentation support for AI automated consultations. The oncology knowledge base integrates relevant literature in the field of genetics, expert consensus on clinical oncology diagnosis and treatment, publications from authoritative institutions, and a large number of real-world consultation conversations and clinical diagnosis and treatment cases. Through data cleaning techniques, a series of operations such as deduplication, outlier processing, data standardization, and data verification are performed on raw, low-quality data. This can significantly improve and optimize the data quality in the knowledge base, ensuring the effectiveness and accuracy of search results. The preprocessed data is stored in a vector database to accelerate knowledge base retrieval and improve the accuracy of search results.

[0041] The present invention uses an AI agent combined with hybrid retrieval RAG technology, combined with deep semantic understanding and generative models, and through the collaborative work of retrieval and generation, accurately understands patient questions and provides relevant answers. In the retrieval stage, the AI agent uses vector space retrieval technology to quickly find relevant document blocks from the team's many years of senior experience base and the medical knowledge base of gynecological tumors in this field, and calculates semantic similarity through vector comparison, and re-ranks the retrieved results using a re-ranking model to optimize the relevance of the retrieval results. Subsequently, the AI large model is used to generate answers that are consistent with the context to ensure that the answers are both accurate and fluent. This technology, through the combination of large model RAG and professional medical literature, has particularly optimized the application in the field of gynecological oncology medicine, enhanced the agent's ability to understand and respond to complex medical problems, and greatly improved the efficiency and quality of patient consultations. In addition, based on the knowledge base of the expert team, the knowledge base was expanded using a large model. After the knowledge documents were segmented, the retrieved and recalled content was tested. For paragraphs with low recall rates but high quality, high-performance large models were used to perform content-related expansion to ensure that the corresponding paragraph documents can be quickly matched during retrieval. Combined with virtual query expansion of the query text in the query retrieval stage, the corresponding knowledge base is more likely to be matched, ensuring the accuracy of query recall and reducing the illusion problem of the large model itself.

[0042] After confirming the patient's desire to draw a pedigree chart, questions can be asked based on the characteristics of the tumor. Ask about the living and deceased relatives within 3-4 generations, and whether there are any patients with the disease. Questions should be simple, specific, and avoid technical terms, so that the consultant understands the reason and purpose of the inquiry. For example: "Does anyone in your family have or currently have a tumor? This includes your parents, siblings, grandparents, and other relatives. This information will help me make a more comprehensive and accurate assessment."

[0043] Based on the pedigree characteristics of tumor syndromes, if multiple relatives in the family have tumors, and the closer the relationship, the greater the possibility of hereditary tumors. When the AI model identifies related tumor cases in relatives on the patient's paternal or maternal branches, it can focus on asking about the health status of relatives on that branch to obtain effective and critical family history information. For example: "I know that your mother has ovarian cancer and has a younger sister. Has she (your aunt) ever had a tumor? This will help us better understand the tumor risk situation in the maternal branch."

[0044] During specific execution, the system automatically identifies and marks important medical entities (such as disease names, kinship, family medical history, etc.) from the patient's answers through large-model intent recognition (Intent Recognition) and entity recognition (Entity Recognition) technologies. These technologies use named entity recognition (NER) and relationship extraction (RelationExtraction) algorithms to accurately extract tumor genetics-related data in medical conversations. These extracted key information will be converted into structured professional gynecological tumor data (JSON format) at the end of each round of conversation. After the inquiry is completed, all family member data will be summarized as the patient's complete family history, and the structured information will be saved to the database through a network request for subsequent process call queries. Contains a data set of all family members. The data structure of individual family members includes the following fields: unique ID representing the individual, gender, age, tumor type, age of diagnosis, whether the consultant, whether deceased, age of death, cause of death, spouse member ID, child member ID, parent member ID, sibling member ID. Such as Figure 2 As shown, the processed family member database can be used to draw a family tree, and can also be used for tumor screening and risk assessment.

[0045] According to the family member database, the kinship relationships of all members are polled to match the corresponding member IDs, and the family history data is reorganized into a tree structure that is easy to parse.

[0046] Taking four generations as an example, the following method can be used to build a tree-structured kinship network, where each family member in the network uses the above data structure:

[0047] a. Create four different generational collections to store corresponding family members, including: grandparents (grandparents of the consultant), parents (parents of the consultant, siblings of the consultant), current generation (consultant and spouse, siblings of the consultant and spouse), and children (children of the consultant, siblings of the consultant);

[0048] b. Starting with the consultant, first move the consultant from the To-Be-Matched Set to the Current Generation Set. Next, search for the consultant's spouse from the To-Be-Matched Set and move the spouse to the Current Generation Set, placing them to the consultant's left. (Genetic counseling primarily investigates the consultant's family tree and follows the reading and writing convention from left to right, so the consultant's spouse is placed on the left. The right side allows for expansion and is used to accommodate the consultant's family tree members.) Finally, search for all of the consultant's children and move them to the Offspring Set.

[0049] c. Find all siblings of the consultant from the To-Be-Matched Set. Also find the spouses of each sibling. Move each pair of siblings and spouses adjacent to each other to the Current Generation Set and place them to the right of the consultant. Finally, traverse each pair of children of the consultant's siblings. Siblings that are adjacent to each other are moved, in order, to the Descendant Set and placed to the right of the consultant's children. At this point, all members of the Current Generation and Descendant Generations have been sorted.

[0050] d. Find the parent of the inquirer from the To-Match Set and move them to the Parent Set, placing the inquirer's mother to the right of the inquirer's father (generally following the rule of males on the left and females on the right). Finally, find all of the siblings of the inquirer's father from the To-Match Set and move them, in order, to the Parent Set, placing them to the left of the inquirer's father. Similarly, find all of the siblings of the inquirer's mother and move them, in order, to the Parent Set, placing them to the right of the inquirer's mother. At this point, all members of the Parent Set have been sorted.

[0051] e. Find the parents of both parents of the client from the to-be-matched set, i.e., the client's grandparents and maternal grandparents. Move them to the ancestral set in order (e.g., if the client's mother is to the right of the client's father, then the client's maternal grandparents should also be to the right of the client's grandparents). At this point, all ancestral members have been sorted.

[0052] f. After traversing all the family members to be matched, four groups of family member data will be obtained, divided according to different generation relationships, and reasonably sorted according to the kinship between members.

[0053] like Figure 3As shown, the pedigree chart is constructed using standard pedigree symbols and relationship lines used in cancer genetic counseling, including healthy individuals, deceased individuals, affected individuals, consultants, individual lines, sibling lines, spouse lines, and descendant lines. A canvas drawing engine can be used to create the pedigree chart. Following the hierarchical structure of the family tree, basic information for each member, along with the pedigree symbol and health status of each individual node, such as gender, age, cancer type, age at diagnosis, and deceased status, can be traversed and drawn. After the individual nodes are drawn, the coordinate positions of each family member's individual node are calculated. A fixed horizontal spacing is set between members of the same generation, and a fixed vertical spacing is set between members of adjacent generations. Based on this layout, horizontal translation adjustments are made based on the kinship relationships between members of adjacent generations, ensuring that the center horizontal coordinates of the child member nodes (if there are multiple child nodes, the center horizontal coordinates of the sibling lines of all children are used) align with the center horizontal coordinates of their parents' spouse lines. Dynamic adjustments are made based on the family distribution, while maintaining the minimum basic spacing between members.

[0054] After the pedigree chart is drawn, it is displayed on the terminal page in canvas format, and provides screen size adaptation, free scaling, image download functions, etc.

Claims

1. A method for drawing a tumor genetic pedigree, characterized in that: The following steps are involved: Collect basic information, cancer history and death information of relatives within several generations of the consultant's family; Define family members with unique member IDs, and standardize the collected data into a family member data structure with member ID as the primary key to obtain a family member database; Based on the family member database, starting from the consultant, the generation coefficient is decreased upwards and increased downwards, the kinship of all family members is polled and the corresponding family member nodes are matched to establish a tree-structured kinship network; Traverse the kinship network to draw out each family member node, identify the family member nodes with a history of cancer, and draw relationship lines between nodes based on kinship to obtain the consultant's cancer genetic pedigree.

2. The method according to claim 1, characterized in that The polling of the kinship relationships of all family members and matching corresponding family member nodes to establish a kinship network with a tree structure includes: Create a set of nodes to be matched and initialize them to all family member nodes; Establish a node set for each generation and move the consultant from the node set to be matched to the node set of this generation; According to the order of the consultant's spouse, the consultant's children, the consultant's siblings and their spouses, the consultant's siblings' children, the consultant's parents, the consultant's parents' siblings, and the consultant's parents' parents, the family member nodes that match the current kinship relationship are retrieved from the set of nodes to be matched. During the retrieval, the family members are matched according to the arrangement order of the family members in the node set of the corresponding generation, and the matched family member nodes are moved to the node set of the corresponding generation according to the matching order to obtain a tree-structured kinship network.

3. The method according to claim 2, characterized in that The method comprises: searching for family member nodes that match the current kinship relationship from the set of nodes to be matched in the order of the consultant's spouse, the consultant's children, the consultant's siblings and their spouses, the consultant's siblings' children, the consultant's parents, the siblings of both consultant's parents, and the parents of both consultant's parents; performing matching according to the arrangement order of the family members in the node set of the corresponding generation during the search; and moving the matched family member nodes to the node set of the corresponding generation in the matching order, including: Search for the spouse and children of the consultant from the set of nodes to be matched, move the spouse to the current generation node set and place it to the left of the consultant, and move the children to the child generation node set; Find the consultant's siblings and the spouse of each sibling from the set of nodes to be matched, and move each sibling and their spouse to the current generation node set adjacent to each other and place them to the right of the consultant; Traverse each sibling of the consultant in the current generation node set, search for the children of each sibling in the to-be-matched node set in order, place the family member nodes that are siblings of each other adjacently, and move the children of each sibling to the descendant node set in the search order and place them to the right of the consultant's children; Find the parents of the consultant from the set of nodes to be matched, move their parents to the set of parent nodes and place the mother to the right of the father; Find the siblings of both parents of the consultant from the set of nodes to be matched, move the siblings of the father to the parent node set and place them on the left side of the father, and move the siblings of the mother to the parent node set and place them on the right side of the mother; The parents of both parents of the consultant are searched from the set of nodes to be matched, and their grandparents and maternal grandparents are moved to the ancestor node set in the search order.

4. The method according to claim 1, wherein When drawing a tumor genetic pedigree, the distance between any two adjacent generations is set to a fixed value.

5. The method according to claim 1, wherein When drawing a tumor genetic pedigree diagram, the node corresponding to the consultant is used as the benchmark, and the node spacing is dynamically allocated according to the number of nodes of family members of the same generation.

6. The method according to claim 1, characterized in that The family member data structure includes a member ID, a basic information field, a tumor history field, a death information field, and a kinship field; the kinship field includes a spouse member ID, a child member ID, a parent member ID, and a sibling member ID.

7. The method according to claim 1, characterized in that The family member node adopts the family member data structure.

8. The method according to claim 1, characterized in that The collection of basic information, tumor medical history and death information of relatives within several generations of the consultant's family is achieved by constructing an AI intelligent body to simulate a doctor to conduct automated medical consultations with the consultant.

9. The method according to claim 8, characterized in that When conducting automated medical consultations, the AI agent uses hybrid retrieval technology combined with a customized genetic knowledge base to dynamically analyze whether the consultant's relatives have related disease factors, and conducts in-depth medical consultations on the relative branches with related disease factors based on the analysis results.

10. The method according to claim 1, characterized in that The family member database is obtained by the AI agent through large-model JSON structured processing and Function Calling technology, combined with prompt word engineering technology, to understand professional tumor information from natural text information and extract effective fields that are conducive to structuring, and assemble them into JSON structured data for specified gynecological tumors.

Images