CN114366030B - Intelligent auxiliary system and method for anesthesia operation - Google Patents
Intelligent auxiliary system and method for anesthesia operation Download PDFInfo
- Publication number
- CN114366030B CN114366030B CN202111677791.8A CN202111677791A CN114366030B CN 114366030 B CN114366030 B CN 114366030B CN 202111677791 A CN202111677791 A CN 202111677791A CN 114366030 B CN114366030 B CN 114366030B
- Authority
- CN
- China
- Prior art keywords
- data
- physiological index
- physiological
- index
- anesthesia
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 206010002091 Anaesthesia Diseases 0.000 title claims abstract description 122
- 230000037005 anaesthesia Effects 0.000 title claims abstract description 122
- 238000000034 method Methods 0.000 title claims abstract description 50
- 208000024891 symptom Diseases 0.000 claims abstract description 126
- 230000036541 health Effects 0.000 claims abstract description 122
- 238000003745 diagnosis Methods 0.000 claims abstract description 89
- 230000003993 interaction Effects 0.000 claims abstract description 71
- 238000012544 monitoring process Methods 0.000 claims abstract description 35
- 230000004069 differentiation Effects 0.000 claims abstract description 19
- 230000005856 abnormality Effects 0.000 claims description 41
- 238000004458 analytical method Methods 0.000 claims description 38
- 238000012549 training Methods 0.000 claims description 30
- 238000001356 surgical procedure Methods 0.000 claims description 26
- 239000008280 blood Substances 0.000 claims description 21
- 210000004369 blood Anatomy 0.000 claims description 21
- 230000005540 biological transmission Effects 0.000 claims description 20
- 230000004927 fusion Effects 0.000 claims description 17
- 238000013210 evaluation model Methods 0.000 claims description 15
- 238000012360 testing method Methods 0.000 claims description 15
- 238000002360 preparation method Methods 0.000 claims description 13
- 238000013461 design Methods 0.000 claims description 12
- 239000007789 gas Substances 0.000 claims description 11
- 238000000605 extraction Methods 0.000 claims description 9
- 238000004806 packaging method and process Methods 0.000 claims description 9
- 208000005952 Amniotic Fluid Embolism Diseases 0.000 claims description 8
- 206010067010 Anaphylactoid syndrome of pregnancy Diseases 0.000 claims description 8
- 206010020751 Hypersensitivity Diseases 0.000 claims description 8
- 206010021143 Hypoxia Diseases 0.000 claims description 8
- 208000007536 Thrombosis Diseases 0.000 claims description 8
- 230000003213 activating effect Effects 0.000 claims description 8
- 208000026935 allergic disease Diseases 0.000 claims description 8
- 230000007815 allergy Effects 0.000 claims description 8
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 claims description 8
- 230000036772 blood pressure Effects 0.000 claims description 8
- 230000036760 body temperature Effects 0.000 claims description 8
- 210000004556 brain Anatomy 0.000 claims description 8
- 230000007954 hypoxia Effects 0.000 claims description 8
- 239000001301 oxygen Substances 0.000 claims description 8
- 229910052760 oxygen Inorganic materials 0.000 claims description 8
- 230000036961 partial effect Effects 0.000 claims description 8
- 201000003144 pneumothorax Diseases 0.000 claims description 8
- 230000035939 shock Effects 0.000 claims description 8
- 208000010496 Heart Arrest Diseases 0.000 claims description 6
- 230000004872 arterial blood pressure Effects 0.000 claims description 6
- 238000004891 communication Methods 0.000 claims description 6
- 238000011156 evaluation Methods 0.000 claims description 6
- 230000000004 hemodynamic effect Effects 0.000 claims description 6
- 238000011161 development Methods 0.000 abstract description 23
- 230000002159 abnormal effect Effects 0.000 abstract description 8
- 230000006870 function Effects 0.000 description 20
- 230000003862 health status Effects 0.000 description 19
- 238000005516 engineering process Methods 0.000 description 16
- 230000000670 limiting effect Effects 0.000 description 9
- 238000007726 management method Methods 0.000 description 8
- 239000008186 active pharmaceutical agent Substances 0.000 description 7
- 230000008569 process Effects 0.000 description 7
- 238000004422 calculation algorithm Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 6
- 238000010801 machine learning Methods 0.000 description 5
- 238000010606 normalization Methods 0.000 description 5
- 239000000243 solution Substances 0.000 description 5
- 230000002411 adverse Effects 0.000 description 4
- 238000006243 chemical reaction Methods 0.000 description 4
- 238000004140 cleaning Methods 0.000 description 4
- 230000002441 reversible effect Effects 0.000 description 4
- 238000001949 anaesthesia Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 3
- 230000006698 induction Effects 0.000 description 3
- 230000010354 integration Effects 0.000 description 3
- 210000005036 nerve Anatomy 0.000 description 3
- 230000002829 reductive effect Effects 0.000 description 3
- 238000012163 sequencing technique Methods 0.000 description 3
- 208000017667 Chronic Disease Diseases 0.000 description 2
- 206010049418 Sudden Cardiac Death Diseases 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 230000001364 causal effect Effects 0.000 description 2
- 230000008878 coupling Effects 0.000 description 2
- 238000010168 coupling process Methods 0.000 description 2
- 238000005859 coupling reaction Methods 0.000 description 2
- 238000013501 data transformation Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 201000010099 disease Diseases 0.000 description 2
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 2
- 230000005611 electricity Effects 0.000 description 2
- 238000005538 encapsulation Methods 0.000 description 2
- 230000001747 exhibiting effect Effects 0.000 description 2
- 208000013409 limited attention Diseases 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000035485 pulse pressure Effects 0.000 description 2
- 208000014221 sudden cardiac arrest Diseases 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 230000001131 transforming effect Effects 0.000 description 2
- 239000003994 anesthetic gas Substances 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 238000002347 injection Methods 0.000 description 1
- 239000007924 injection Substances 0.000 description 1
- 238000003780 insertion Methods 0.000 description 1
- 230000037431 insertion Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000007637 random forest analysis Methods 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 230000000241 respiratory effect Effects 0.000 description 1
- 230000029058 respiratory gaseous exchange Effects 0.000 description 1
- 238000012552 review Methods 0.000 description 1
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
- A61B5/0205—Simultaneously evaluating both cardiovascular conditions and different types of body conditions, e.g. heart and respiratory condition
- A61B5/02055—Simultaneously evaluating both cardiovascular condition and temperature
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/0002—Remote monitoring of patients using telemetry, e.g. transmission of vital signals via a communication network
- A61B5/0015—Remote monitoring of patients using telemetry, e.g. transmission of vital signals via a communication network characterised by features of the telemetry system
- A61B5/0022—Monitoring a patient using a global network, e.g. telephone networks, internet
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/01—Measuring temperature of body parts ; Diagnostic temperature sensing, e.g. for malignant or inflamed tissue
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/145—Measuring characteristics of blood in vivo, e.g. gas concentration, pH value; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid, cerebral tissue
- A61B5/14542—Measuring characteristics of blood in vivo, e.g. gas concentration, pH value; Measuring characteristics of body fluids or tissues, e.g. interstitial fluid, cerebral tissue for measuring blood gases
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/74—Details of notification to user or communication with user or patient ; user input means
- A61B5/7405—Details of notification to user or communication with user or patient ; user input means using sound
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Public Health (AREA)
- Surgery (AREA)
- Veterinary Medicine (AREA)
- General Health & Medical Sciences (AREA)
- Animal Behavior & Ethology (AREA)
- Biophysics (AREA)
- Pathology (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Medical Informatics (AREA)
- Molecular Biology (AREA)
- Cardiology (AREA)
- Physiology (AREA)
- Human Computer Interaction (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Multimedia (AREA)
- Computational Linguistics (AREA)
- Acoustics & Sound (AREA)
- Optics & Photonics (AREA)
- Pulmonology (AREA)
- Computer Networks & Wireless Communication (AREA)
- Medical Treatment And Welfare Office Work (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
Abstract
The invention discloses an intelligent auxiliary system and method for anesthesia operation, wherein the system comprises: the hardware monitoring terminal collects, transmits and fuses various physiological index data; the database module receives and stores anesthesia case data and basic information data of patients and physiological index data; the decision module acquires physiological index data, anesthesia case data and basic information data of patients, establishes physiological index differentiation standard, trains a decision model comprising physiological index health assessment, potential symptom prediction and diagnosis and treatment scheme recommendation; the doctor application terminal acquires physiological index health assessment, potential symptom prediction, diagnosis and treatment scheme recommendation and performs adjustment; the voice module provides voice interaction services of data linkage, inquiry and play. According to the voice interaction service provided by the invention, voice reminding can be given when the physiological index is abnormal or some suspicious symptoms exist; the doctor can also be provided with advice of the physiological index development condition or diagnosis and treatment scheme of the patient; effectively lightens the attention burden of doctors and improves the safety of the operation.
Description
Technical Field
The invention relates to the technical field of intelligent medical treatment, in particular to an intelligent auxiliary system and method for anesthesia operation.
Background
The medical monitor, blood gas analyzer, anaesthesia machine and other devices are common medical devices in operating rooms, anaesthesia recovery rooms, intensive care units and ambulances. The traditional monitor, blood gas analyzer, anesthesia machine and other equipment can carry out the real-time supervision of indexes such as vital sign, anesthetic gas, anesthesia degree of depth, but still have a great deal of defects in the in-service use: 1. when only one anesthesiologist is in the operating room, the vital signs of the patient are easily ignored when the operations such as anesthesia induction, puncture, nerve block and the like are performed, and the monitoring equipment has no voice reminding function, so that the anesthesia safety is reduced; 2. when emergency treatment, rescue and other conditions are met, the equipment cannot report vital signs in real time and help doctors to analyze abnormal reasons; 3. the vital sign, the respiratory parameter, the blood gas index, the thromboelastography and other anesthesia operation related indexes cannot be integrated and analyzed; 4. when a problematic case is encountered, a doctor is easy to cause decision errors due to the limitation of self knowledge. There is no report on the anesthesia operation auxiliary decision system and application in clinic.
The prior patent CN110232961a discloses a voice recognition system for controlling the mode of administration of an anesthesia machine. The system changes artificial anesthesia control into voice control, the output object is an anesthesia machine, and the output result is anesthesia administration parameters, such as: the type of administration, rate, concentration, etc. The prior patent CN111798956a discloses a decision determining method, device and system for artificial intelligence anesthesia, which uses the historical anesthesia decision of a user as the prior decision reference by acquiring the target user identification, acquires the vital sign information of the user to predict the anesthesia reaction, and controls the anesthesia implementing device to perform anesthesia control such as medicine dosage, category, injection speed and the like according to the reaction. There are a number of limitations: 1. no voice interaction form is provided; 2. the physical condition of the user changes along with the development of time, the types of operations implemented by the user can be possibly distinguished, and whether the historical anesthesia decision has obvious reference significance is to be questioned; 3. summarizing and fusing different monitoring hardware data, and no specific implementation scheme exists; 4. the vital sign abnormality adopts a general judgment standard, and is not suitable for patients with specific physiological indexes; the vital sign value is adopted as an algorithm expression feature, the range of the value is too wide, and the feature expression distinction degree is poor, so that the algorithm accuracy is not high; 5. the anesthesia operation is complicated in the category of the surgery, the surgery process is influenced by various sudden, urgent and uncontrollable factors, the operation process is completely separated from doctors, the algorithm output is directly used as an anesthesia control decision, and the practical application rationality, feasibility and safety are not provided.
Disclosure of Invention
Aiming at the defects in the technology, the invention provides an intelligent auxiliary system and method for anesthesia operation, which are provided with voice interaction service, and if a physiological index abnormality or a certain suspicious symptom exists in an intraoperative patient, the system can actively give out voice prompt; the voice inquiry function provides the doctor with the advice of the physiological index development condition or the diagnosis and treatment scheme of the patient at any time; the voice interaction service effectively lightens the attention burden of doctors and improves the operation safety.
To achieve these objects and other advantages and in accordance with the purpose of the invention, as embodied by the following:
the embodiment of the invention provides an intelligent auxiliary system for anesthesia operation, which comprises the following components:
the hardware monitoring terminal is used for collecting, transmitting and fusing various physiological index data;
a database module for receiving and storing anesthesia case data and patient base information data and the physiological index data;
the decision module is used for acquiring the physiological index data, the anesthesia case data and the basic information data of the patient, making physiological index differentiation standard and training a decision model; the decision model is used for providing physiological index health assessment, potential symptom prediction and diagnosis and treatment scheme recommendation;
The doctor application terminal is used for acquiring the physiological index health assessment, the potential symptom prediction, the diagnosis and treatment scheme recommendation and performing adjustment through a doctor;
and the voice module is respectively in communication connection with the database module, the decision module and the doctor application terminal and is used for data linkage and providing voice interaction services including inquiry and play for the doctor application terminal.
Preferably, the hardware monitoring terminal includes:
a monitoring unit for collecting a plurality of the physiological index data,
the transmission and fusion unit is used for unifying transmission modes of a plurality of physiological index data, formulating unified coding standards of the plurality of physiological index data, fusing and outputting;
wherein the plurality of physiological indexes at least comprise vital signs, blood gas indexes, hemodynamic indexes and anesthesia depth indexes;
the multiple physiological index data output after fusion comprise body temperature, heart rate, pulse rate, blood pressure, blood oxygen saturation, and end-of-breath CO 2 Partial pressure, central venous pressure, arterial pressure variation and brain electrical double frequency index.
Preferably, the database module includes:
the database unit is used for storing the fused multiple physiological index data, the anesthesia case data and the basic information data of the patient;
The packaging test unit is used for carrying out classified packaging, network encryption, authority setting and test analysis on all data of the database unit; the method comprises the steps of,
and the API interface unit is used for providing inquiry, storage and editing services of the data.
Preferably, the decision module includes:
a data preparation unit for acquiring the physiological index data, the anesthesia case data, and the patient base information data;
a specification making unit for making a physiological index abnormality determination specification for the data acquired by the data preparing unit; the method comprises the steps of,
the feature extraction and decision model training unit is used for extracting features of the data acquired by the data preparation unit, combining the physiological index abnormality judgment standard, training a decision model and outputting a decision, wherein the decision comprises the physiological index health assessment, the potential symptom prediction and the diagnosis and treatment scheme recommendation;
wherein the physiological index abnormality determination specification comprises a general specification and a differentiation determination specification; the universal specification includes setting a given limit for health of a plurality of the physiological indicators; the differential judgment standard comprises inserting an initial reference value set according to basic information data of a patient on the basis of the given limit value, and setting independent judgment conditions of each physiological index; the independent determination condition includes a reference limit value and a trend condition value.
Preferably, the decision model comprises:
the physiological index health evaluation model is used for carrying out physiological index health tracking evaluation on a plurality of physiological indexes based on the physiological index abnormality judgment standard and outputting a plurality of physiological index health states;
the potential symptom prediction model is used for retrieving the anesthesia case data according to the health states of the multiple physiological indexes, acquiring physiological index data of batch patients and common intraoperative symptom information, and completing label matching of corresponding physiological index information by taking the intraoperative symptom as a label; fitting a physiological index health evaluation model to the physiological index information of the patients in batches, and outputting a classification result of the physiological index health state of the patients; matching the result combination of the physical index health state with the intraoperative symptom label again, training a potential symptom prediction model for the data set, and outputting a prediction result of the potential symptom of the patient; the method comprises the steps of,
a diagnosis and treatment plan recommendation model obtained by retrieving the anesthesia case data in the database module according to the plurality of physiological index health states and the potential symptom classification result, the diagnosis and treatment plan recommendation model being used for giving a diagnosis and treatment plan recommendation;
Wherein the health states of the multiple physiological indexes comprise normal, high, low, extremely high and extremely low;
the potential symptom classification includes allergy, hypoxia, shock, pneumothorax, amniotic fluid embolism, venous air thrombosis and cardiac arrest.
Preferably, the doctor application terminal further includes:
the data acquisition and display unit is used for acquiring the anesthesia case data and the decision and display data;
the business management unit is used for designing and managing data linkage and interaction logic among the units or modules according to actual use requirements;
the first voice interaction unit is used for realizing the data linkage and interaction logic with the voice module through voice; the method comprises the steps of,
the scheme adjusting and iterating unit is used for checking, adjusting and checking the acquired health states of the multiple physiological indexes, the potential symptom classification result and the diagnosis and treatment scheme recommendation according to actual clinical experience by a doctor to form a final diagnosis and treatment scheme, and updating and iterating the final diagnosis and treatment scheme to the database module;
the data linkage and interaction logic comprises the steps of setting a patient reference value, defining an entry by a design index specification, opening or closing a physiological index trend analysis, opening or closing a diagnosis and treatment scheme suggestion, opening or closing a voice prompt, opening or closing a vital sign abnormality prompt, opening or closing a potential symptom prompt and activating voice awakening.
Preferably, the voice module further comprises:
the information interaction unit is used for respectively carrying out information data interaction with the database module and the decision module;
the second voice interaction unit is used for realizing data linkage and interaction logic with the doctor application terminal through voice;
the data linkage and interaction logic comprises the steps of setting a patient reference value, defining an entry by a design index specification, opening or closing a physiological index trend analysis, opening or closing a diagnosis and treatment scheme suggestion, opening or closing a voice prompt, opening or closing a vital sign abnormality prompt, opening or closing a potential symptom prompt and activating voice awakening.
An intelligent assistance method for anesthesia surgery, comprising the steps of:
collecting, transmitting and fusing various physiological index data;
setting a database, and receiving and storing anesthesia case data, basic information data of patients and the physiological index data;
acquiring the physiological index data, the anesthesia case data and the basic information data of the patient, and making a physiological index abnormality judgment standard and a training decision model; the decision model is used for providing physiological index health assessment, potential symptom prediction and diagnosis and treatment scheme recommendation;
The doctor adjusts the acquired physiological index health assessment, the potential symptom prediction and the diagnosis and treatment scheme recommendation;
through setting the voice interaction service, a doctor obtains the inquiry and play service;
the inquiry and play service comprises the steps of inserting patient reference values, defining an entrance by design index specification, opening or closing physiological index trend analysis, opening or closing diagnosis and treatment scheme suggestion and opening or closing voice prompt.
Preferably, the making of the physiological index abnormality determination criterion includes the steps of:
setting given limit values of the health of a plurality of physiological indexes as a general specification;
on the basis of the given limit value, inserting an initial reference value set according to basic information data of patients, and setting independent judgment conditions of each physiological index to form a differential judgment standard;
wherein the independent determination condition includes a reference limit value and a trend condition value.
Preferably, training the decision model comprises the steps of:
performing physiological index health tracking evaluation on a plurality of physiological indexes based on the differentiation judgment standard to obtain a physiological index health evaluation model, and outputting a plurality of physiological index health states;
According to the health states of the multiple physiological indexes, retrieving the anesthesia case data, acquiring physiological index data of a batch of patients and common intraoperative symptom information, and completing label matching of corresponding physiological index information by taking the intraoperative symptom as a label; fitting a physiological index health evaluation model to the physiological index information of the patients in batches, and outputting a classification result of the physiological index health state of the patients; matching the result combination of the physical index health state with the intraoperative symptom label again, training a potential symptom prediction model for the data set, and outputting a prediction result of the potential symptom of the patient;
according to the health states of the multiple physiological indexes and the potential symptom classification result, anesthesia case data in the database module are searched, a diagnosis and treatment scheme recommendation model is obtained, and diagnosis and treatment scheme recommendation is given;
wherein the health states of the multiple physiological indexes comprise normal, high, low, extremely high and extremely low;
the potential symptom classification includes allergy, hypoxia, shock, pneumothorax, amniotic fluid embolism, venous air thrombosis and cardiac arrest.
The invention at least comprises the following beneficial effects:
1. the intelligent auxiliary system and the intelligent auxiliary method for anesthesia operation are provided with voice interaction service, and if a physiological index abnormality or a certain suspicious symptom exists in an operation patient, the system can actively give out voice prompt; the voice inquiry function provides the doctor with the advice of the physiological index development condition or the diagnosis and treatment scheme of the patient at any time; the voice interaction service effectively reduces the attention burden of doctors and improves the operation safety;
2. The invention provides an intelligent auxiliary system and a method for anesthesia operation, wherein the physiological index abnormality judgment standard comprises a general standard and a differential judgment standard; the general specification includes setting given limits for the health of a variety of physiological indicators; the differentiation judgment standard comprises inserting an initial reference value set according to basic information data of a patient on the basis of a given limit value, and setting independent judgment conditions of each physiological index; the independent judgment conditions comprise a reference limit value and a trend condition value; the coverage of patients is more comprehensive, and the judgment result is more scientific and accurate;
3. according to the intelligent auxiliary system and the intelligent auxiliary method for the anesthesia operation, three decisions including physiological index health assessment, potential symptom prediction and diagnosis and treatment scheme recommendation are set, feature analysis is carried out by taking the symptom output result and the physiological index health state as data sources according to the physiological index health assessment and the potential symptom prediction results, database retrieval is carried out, diagnosis and treatment scheme suggestions with high satisfaction degree and high matching degree can be output, and the accuracy of the decisions is improved.
Additional advantages, objects, and features of the invention will be set forth in part in the description which follows and in part will become apparent to those having ordinary skill in the art upon examination of the following or may be learned from practice of the invention.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are needed in the description of the embodiments or the prior art will be briefly described, and it is obvious that the drawings in the description below are some embodiments of the present invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic diagram of a smart assistance system for anesthesia procedures according to the present invention;
FIG. 2 is a schematic diagram of the components and communication of the hardware monitoring terminal according to the present invention;
FIG. 3 is a schematic diagram of the database module according to the present invention;
FIG. 4 is a schematic diagram of the composition and communication of the decision module according to the present invention;
FIG. 5 is a schematic diagram of the composition and communication of a doctor application terminal according to the present invention;
FIG. 6 is a schematic diagram of the voice module according to the present invention;
FIG. 7 is a flow chart of the intelligent assistance method for anesthesia surgery according to the present invention;
FIG. 8 is a flowchart of a method for determining a physiological index abnormality determination criterion according to the present invention;
FIG. 9 is a flow chart of a method for training a decision model according to the present invention;
in the figure:
100. a hardware monitoring terminal; 110. a monitoring unit; 120. a transmission and fusion unit; 200. a database module; 210. a database unit; 220. a package test unit; 230. an API interface unit; 300. a decision module; 310. a data preparation unit; 320. a specification making unit; 330. the feature extraction and decision model training unit; 331. a physiological index health assessment model; 332. a model of potential symptom prediction; 333. a diagnosis and treatment scheme recommendation model; 400. a doctor application terminal; 410. a data acquisition and display unit; 420. a service management unit; 430. a first voice interaction unit; 440. a scheme adjustment and iteration unit; 500. a voice module; 510. an information interaction unit; 520. and a second voice interaction unit.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are some, but not all embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Terms such as "having," "including," and "comprising" used in various embodiments of the invention described below do not exclude the presence or addition of one or more other elements or combinations thereof; the technical features involved can be combined with one another as long as they do not conflict with one another.
< embodiment 1>
As shown in fig. 1-6. The embodiment of the invention provides an intelligent auxiliary system for anesthesia surgery, which comprises a hardware monitoring terminal 100, a database module 200, a decision module 300, a doctor application terminal 400 and a voice module 500. The hardware monitoring terminal 100 is used for collecting, transmitting and fusing various physiological index data; the database module 200 is used for receiving and storing anesthesia case data and basic information data of patients and physiological index data; the decision module 300 is used for acquiring physiological index data, anesthesia case data and basic information data of patients, making physiological index differentiation specification and training a decision model; the decision model is used for providing physiological index health assessment, potential symptom prediction and diagnosis and treatment scheme recommendation; the doctor application terminal 400 is used for acquiring physical index health assessment, potential symptom prediction, diagnosis and treatment scheme recommendation and performing adjustment by a doctor; the voice module 500 is respectively in communication connection with the database module 200, the decision module 300 and the doctor application terminal 400, and is used for data linkage and providing voice interaction services including inquiry and play for the doctor application terminal 400.
Based on the above-described embodiments, the present embodiment gives a preferred embodiment of the hardware monitoring terminal 100.
The hardware monitoring terminal 100 includes a monitoring unit 110, a transmission and fusion unit 120. The monitoring unit 110 is configured to collect various physiological index data, preferably, the monitoring unit 110 includes, but is not limited to, a multi-parameter monitor, a blood gas analyzer, an anesthesia machine and other medical devices commonly used in anesthesia operation, and is capable of collecting various physiological index information of a patient including, but not limited to, vital signs, blood gas index, hemodynamic index, anesthesia depth index and the like. The transmission and fusion unit 120 is configured to unify a transmission mode of the multiple kinds of physiological index data, formulate a unified coding specification of the multiple kinds of physiological index data, and output the fused multiple kinds of physiological index data after fusion, where the fused multiple kinds of physiological index data include body temperature, heart rate, pulse rate, blood pressure, blood oxygen saturation, end-of-breath CO2 partial pressure, central venous pressure, arterial pulse pressure variability, and brain electricity double-frequency index. Specifically, the transmission mode of the multiple physiological index data is unified, which includes, but is not limited to, adopting one of Bluetooth, wiFi and 4G/5G as a transmission technology form, and being provided with a corresponding data transmission module Bluetooth gateway or WiFi router or 4G/5G router, so that the data fusion and normalization of the multiple physiological indexes can be facilitated. The unified coding specification of the multiple kinds of physiological index data is formulated for packaging and encoding and decoding of multiple kinds of subsequent physiological indexes, for example, the unified coding specification may be a data packaging format of unified different hardware monitoring terminals 100, and mainly includes firmware information, data transmission frequency, data transmission time, data exchange format, data parameter naming and corresponding types, and the like, as shown in the following table 1.
TABLE 1 unified coding Specification
The plurality of monitoring units 110 eventually fuse the normalized set of data units according to the unified coding specification as shown in table 2 below.
TABLE 2 fusion-normalized set of data examples
It should be noted that, in table 1 and table 2, the parameter terminal is represented as a set of array data, which represents that the data of 3 hardware monitoring terminals 100 are fused, and are respectively a monitor— "jianhuyi", a blood gas analyzer — "xueqiyi", and an anesthesia machine — "mazuiji"; the array data stored by the parameter mac represents the physical mac addresses corresponding to the 3 types of hardware; the parameter freq represents the data acquisition frequency of hardware interruption, and the acquisition frequency is adjusted according to the actual application requirement; the parameter timeStamp stores unix time stamps, representing specific time of data acquisition; the parameter data represents the fused physiological index data.
Based on the above embodiments, the present embodiment gives a preferred embodiment of the database module 200.
Database module 200 includes a database unit 210, a package test unit 220, and an API interface unit 230. The database unit 210 is used for storing the fused various physiological index data, anesthesia case data and basic information data of patients. Here, the basic information data of the patient includes basic information such as the patient's inpatient number, sex, age, weight, etc., and physiological index abnormality determination criterion information corresponding to the patient; the multiple physiological index data comprise vital signs, blood and qi indexes, hemodynamic indexes, anesthesia depth indexes and the like; the fused multiple physiological index data comprise body temperature, heart rate, pulse rate, blood pressure, blood oxygen saturation, end-of-breath CO2 partial pressure, central venous pressure, arterial pressure variation and brain electricity double-frequency index BIS; the anesthesia case data is obtained by acquiring abundant anesthesia operation cases through hospital data authority. The package testing unit 220 is used for performing classified package, network encryption, authority setting and test analysis on all data of the database unit 210. Specifically, the package testing unit 220 packages different service data, ensures data security by means of network encryption technology, setting data access authority, and the like, and provides query, storage and editing services for other units through the API interface form of the API interface unit 230. In addition, the package testing unit 220 performs performance, availability, consistency and expansibility analysis on the architecture of the database module 200, so as to ensure stable and efficient operation of the database and the server.
Based on the above embodiments, the present embodiment gives a preferred embodiment of the decision module 300. The decision module 300 includes a data preparation unit 310, a specification making unit 320, a feature extraction and decision model training unit 330. The data preparation unit 310 is configured to initiate a data request to the database module 200, obtain the physiological index data, the anesthesia case data and the basic information data of the patient, implement a data preparation task before the algorithm modeling, and improve the algorithm development efficiency. Still further, the data preparation unit 310 further includes cleaning, integrating, and transforming the data. The data cleaning removes noise data and irrelevant data in the source data set, processes missing data, cleans dirty data and blank values, identifies and deletes isolated points and the like; data integration completes data matching of the same entity; the data transformation is used to find a characteristic representation of the data. The specification making unit 320 is configured to make a physiological index abnormality determination specification for the data acquired by the data preparing unit 310. The physiological index abnormality determination specification comprises a general specification and a differentiation determination specification; the general specification includes setting given limits for the health of a variety of physiological indicators; the differentiation judgment standard comprises inserting an initial reference value set according to basic information data of a patient on the basis of a given limit value, and setting independent judgment conditions of each physiological index; the independent determination condition includes a reference limit value and a trend condition value. Specifically, the specification making unit 320 performs information interaction with the doctor application terminal 400, makes a physiological index differentiation specification for patients having significant individual differences, and the subsequent unit performs feature extraction analysis with reference to the specification. The feature extraction and decision model training unit 330 is used for feature extraction of the data acquired by the data preparation unit 310, training a decision model in combination with physiological index differentiation specification, and outputting decisions including physiological index health assessment, potential symptom prediction and diagnosis and treatment scheme recommendation. Specifically, the feature extraction and decision model training unit 330 converts the original data into features by means of regular normalization or mathematical conversion for different models established by the subsequent units, so as to improve the accuracy of the training model.
As a further preference of the above embodiment, the decision model includes a physiological index health assessment model 331, a potential symptom prediction model 332, a diagnosis and treatment plan recommendation model 333. The physiological index health evaluation model 331 is configured to perform physiological index health tracking evaluation on multiple physiological indexes based on a physiological index abnormality determination criterion, and output multiple physiological index health states, for example, five physiological index health states: normal, high, low, extremely high, extremely low. The potential symptom prediction model 332 is used for retrieving anesthesia case data according to the health states of various physiological indexes, acquiring physiological index data of batch patients and common intra-operative symptom information, and completing label matching of corresponding physiological index information by taking the intra-operative symptom as a label; fitting the physiological index health evaluation model 331 to the physiological index information of the patients in batches, and outputting the classification result of the physiological index health state of the patients; and matching the result combination of the physical index health state with the intraoperative symptom label again, so as to train a potential symptom prediction model for the data set, and outputting a prediction result of the potential symptom of the patient. The diagnosis and treatment scheme recommendation model 333 is obtained by searching the anesthesia case data in the database module 200 according to various potential symptom classification results, such as allergy, hypoxia, shock, pneumothorax, amniotic fluid embolism, venous air thrombosis and sudden cardiac arrest, the diagnosis and treatment scheme recommendation model 333 is used for giving diagnosis and treatment scheme recommendation, for example, the diagnosis and treatment scheme recommendation model 333 receives the output results of the physiological index health assessment model 331 and the intraoperative potential symptom prediction model 332, performs feature analysis by taking the symptom output results and the physiological index health state as data sources, searches the anesthesia case database, applies semantic analysis and numerical positioning technology, and gives diagnosis and treatment scheme recommendation according to satisfaction degree and matching degree sequencing.
The interaction between the three models is exemplified by 9 physiological indexes (body temperature, heart rate, pulse rate, blood pressure, blood oxygen saturation, respiration end CO2 partial pressure, central venous pressure, arterial pulse pressure variation, brain electrical double frequency index BIS), the physiological index abnormality judgment standard comprises two types, one is a general judgment standard suitable for common people, the other is a differential judgment standard suitable for special patients, whether the physiological index is abnormal or not is judged by a given limit value under the general standard, the differential judgment standard is inserted into an initial reference value according to the patient condition, an independent judgment condition is set for a plurality of physiological indexes on the basis, the given limit value is required to be judged, the reference limit value and the trend condition of different physiological indexes are required to be considered, 9 physiological index data of patients are independently defined according to the difference of the patient, and 5 physiological index health states including the normal, the high, the low, the extremely high and the extremely low potential symptom prediction models in the operation are output by using the physiological index health state of 9 physiological index evaluation models 331, namely, the 9 physiological index health state combinations are taken as characteristic feature combinations of the physiological index health index, the physiological index is obtained by the model in the model, the forest operation is matched with the corresponding physiological index information, the forest condition information is obtained, the results are matched with the corresponding physiological index data in a batch, the method is matched, the results are matched with the physiological index data is obtained, the normal operation is matched, the results is matched with the physiological index data is matched with the normal, and the results is matched with the normal results, and the results are obtained by the 5 physiological index model are obtained by the physiological index model by the model and the model is 9 by the physiological index is based, the model predicts the possible symptoms of the patient, and the potential symptom classification results comprise: allergy, hypoxia, shock, pneumothorax, amniotic fluid embolism, venous air thrombosis, cardiac arrest and the like. The model predicts symptom classifications that substantially cover common symptoms of anesthesia procedures.
The physiological index abnormality determination specification is further described in the following.
The decision criteria of the general specification are given limits for the health of a given set of physiological indicators, which are applicable to all patients of the general population. All the given limit conditions of the physiological index health can be defined in an array form, such as threlist= [ thre0, thre1, thre2, thre3], and if the current physiological index value is < thre0, the index health status is 'very low'; if thre0< = the current physiological index value < thre1, the index health status is "low"; if thre1< = the current physiological index value < = thre2, the index health status is "normal"; if thre2< the current physiological index value < = thre3, the index health status is "high"; if thre3< the current physiological index value, the index health status is "extremely high".
The differentiation judgment standard is formed by combining trend development conditions trendCond and given limit conditions baseCond, all physiological indexes can be defined according to the standard, and different patients have different trend conditions and limit conditions due to individual differences. If the trend of a certain physiological index of the patient is satisfactory, the health state is normal; the trend of the physiological index is unsatisfactory, the physiological index is further judged according to a limiting condition, and the limiting condition is set by referring to an initial reference value of a patient. The differentiated determination criteria can effectively cover a wide variety of patients in a wide variety of surgical situations.
The trend condition is defined as follows, wherein, after trend analysis is started on a certain physiological index, if the monitored value of start < = pct%is < = end, the trend development of the physiological index is satisfied; when start is null (nan), it means that if the monitored value of pct% < = end, the trend of the physiological index is satisfactory; in the case where end is a null value (denoted by nan), this kind of physiological index trend is satisfactory if the monitor value of pct =start. The following are illustrated:
1. the trend condition of the central venous pressure CVP of a certain patient is trendcon= [80%,8cm H20 and 12cm H20], which means that after trend analysis is started, if the central venous pressure monitoring value of 8cm H20< = 80% is < = 12cm H20, the central venous pressure trend is satisfactory, and the reverse trend is unsatisfactory;
2. the trend condition of the heart rate HR of a certain patient is trendcon= [70%,45 times/min, nan ], which means that after trend analysis is started, if the heart rate HR monitoring value of 45 times/min < = 70%, the central venous pressure trend is satisfactory, and the opposite trend is unsatisfactory;
3. the trend condition of Pulse rate Pulse of a certain patient is trendcon = [80%, nan,100 times/min ], which means that after trend analysis is started, if 80% Pulse rate Pulse monitoring value < = 100 times/min, pulse rate Pulse trend is satisfactory, and reverse trend is unsatisfactory.
The limiting value condition form is defined according to an array form, such as baseCond= [ base0, base1, base2, base3], and the limiting value condition baseCond fitting is carried out on the current physiological index value under the condition that certain physiological index trend management is not satisfied. If the current physiological index value is < base0, the index health state is 'extremely low'; if base0< = the current physiological index value < base1, the index health status is "low"; if base1< = the current physiological index value < = base2, the index health status is "normal"; if base2< the current physiological index value < = base3, the index health status is "higher"; if base3 is < the current physiological index value, the index health status is "extremely high".
In this embodiment, different modules focus on implementing different sub-division tasks, and causal association and data coupling exist between the tasks. The modules are mutually matched and evolve layer by layer, and the wide unordered physiological index data is finally output as decisions focused by doctors.
Based on the above embodiments, the present embodiment gives a preferred embodiment of the doctor application terminal 400. The doctor application terminal 400 further comprises a data acquisition and display unit 410, a service management unit 420, a first voice interaction unit 430, and a scheme tuning and iteration unit 440. The data acquiring and displaying unit 410 is configured to acquire and display anesthesia case data and decisions, for example, data information of interest of a doctor, such as a physiological index, predicted symptoms, diagnosis and treatment schemes, text descriptions related to patient information, data graphs related to trend analysis, and the like, can be displayed on the UI interface of the doctor application terminal 400. The service management unit 420 is configured to design and manage data linkage and interaction logic between services according to actual usage requirements. The first voice interaction unit 430 is configured to implement data linkage and interaction logic with the voice module 500 through voice, where the data linkage and interaction logic includes setting an insertion patient reference value, a design index specification definition entry, opening or closing a physiological index trend analysis, opening or closing a diagnosis and treatment plan suggestion, opening or closing a voice prompt, opening or closing a vital sign abnormality reminder, opening or closing a potential symptom reminder, and activating voice wakeup. Specifically, the first voice interaction unit 430 may be used to define related content and specific form of semantic recognition, and is responsible for determining related content and specific form of semantic stitching; the method can be used for realizing the voice function, for example, follow-up development is carried out by means of a third party development platform for the voice of the fly voice, the hundred-degree voice and the Hua voice; the system can also be used for providing inquiry and broadcasting services, and the inquiry function can provide the doctor with the development condition of the physiological index of the patient or the proposal of the diagnosis and treatment scheme at any time; the broadcasting function actively gives out voice prompt when the physiological index is abnormal or some suspicious symptoms exist. The solution adjustment and iteration unit 440 is configured to review, adjust and calibrate the obtained health status of multiple physiological indexes, the classification result of potential symptoms and the recommendation of the diagnosis and treatment solution according to the actual clinical experience, form a final diagnosis and treatment solution, and update and iterate the final diagnosis and treatment solution to the database module 200, on one hand, adjust and calibrate the final medical decision, and improve the accuracy, scientificity and safety of the final medical decision; on the other hand, iteration is conducted, the final medical scheme decision is used for reference, the medical decision is perfected, and the data corresponding to the intraoperative symptoms, the physiological index state, the patient information and the like are updated, backed up and synchronized to the database module 200. The method is beneficial to improving the performance of the decision model after multiple iterations, thereby providing more accurate and more perfect diagnosis and treatment decision suggestions for doctors.
Based on the above embodiments, the present embodiment gives a preferred embodiment of the voice module 500. The voice module 500 further comprises an information interaction unit 510, a second voice interaction unit 520. The information interaction unit 510 is configured to interact information data with the database module 200 and the decision module 300 respectively; the second voice interaction unit 520 is configured to implement data linkage and interaction logic with the doctor application terminal 400 through voice, where the data linkage and interaction logic includes inserting patient reference values, defining an entry by design index specification, opening or closing trend analysis of physiological indexes, opening or closing advice of diagnosis and treatment schemes, opening or closing voice prompts, opening or closing abnormal vital sign prompt, opening or closing potential symptom prompt, and activating voice wakeup.
Embodiment 1 provides an intelligent auxiliary system for anesthesia surgery, which can realize data acquisition and fusion of multiple hardware monitoring terminals 100 and multiple types of physiological indexes, wherein the hardware monitoring terminals 100 are easy to expand, the data acquisition, integration and analysis of the multiple hardware monitoring terminals 100 and the multiple types of physiological indexes can be realized according to surgery demands, and the comprehensive surgery indexes have important significance for analyzing symptoms of patients in surgery and determining diagnosis and treatment schemes. Database module 200 enables data relating to anesthesia procedures to be stored in update iterations. The decision module 300 trains and completes models such as physical index health assessment, potential symptom prediction, diagnosis and treatment scheme recommendation and the like of patients based on anesthesia operation cases by applying a machine learning method, and outputs decisions. The physiological index abnormality judgment standard based on the individual reference value is provided with two standards, namely a common standard and a differential standard, so that the patient is covered more comprehensively, and the judgment result is more scientific and accurate. Anesthesia surgery involves multiple types of patients, multiple types of surgery, and different patients may have differences in physical quality, or may be affected by certain chronic diseases, or may have a history of some past disease, exhibiting significant individual variability. There are respective criteria for whether different physiological indicators of the patient are in a normal state during the operation. The physiological index abnormality judgment criterion provided by the invention is not only suitable for common patients, but also suitable for patients with obvious individual differences of certain physiological indexes. In addition, the decision model is based on abundant anesthesia operation cases, a machine learning method is applied, models such as health assessment, potential symptom prediction, diagnosis and treatment scheme recommendation and the like of physiological indexes of patients are trained and completed, and the health development condition of the physiological indexes of the patients in operation is assessed in real time; predicting potential symptoms of the patient, and giving out a detailed diagnosis and treatment proposal; the diagnosis and treatment scheme recommendation model 333 can be fed back to the database module 200 for iterative update after being adjusted according to the experience of the doctor. The decision of the decision model provides real-time intra-operative decision reference for doctors, assists the doctors to make more efficient, safer and more accurate surgical decisions, avoids decision errors caused by self knowledge limitation or attention loss of the doctors, and further avoids adverse events such as medical accidents, medical disputes and the like. The intelligent voice module 500 completes the technical realization of the voice interaction function; the doctor application terminal 400 is used as a doctor operation interaction terminal, and specific service functions are realized by calling data, voice and decision. Under the condition that only one anesthesiologist performs anesthesia induction, puncture, nerve block and other operations in an operating room, or under the condition of complex operations such as emergency treatment, rescue and the like, the doctor has limited attention and easily ignores the index development condition and potential suspicious symptoms of the patient. Based on the interactive form of the intelligent voice module 500, if the physiological index of the patient in operation is abnormal or some suspicious symptoms exist, the patient actively gives out voice prompt; in addition, the voice query function provides advice for doctors on the development condition of the physiological index of the patient or the diagnosis and treatment scheme at any time. The interaction form can effectively lighten the attention burden of doctors and improve the safety of operations.
In summary, the anesthesia surgery is complicated in coverage, and involves various patients, and the surgery process is affected by various sudden, urgent and uncontrollable factors. The intelligent auxiliary system for anesthesia surgery provided by the invention applies a data fusion technology, an intelligent voice technology, an artificial intelligent technology and an internet technology, so that the attention burden of the doctor in surgery is reduced to the greatest extent, a real-time decision reference is provided for doctors, and the doctors are assisted to make more efficient, safer and more accurate intra-surgery decisions, so that adverse events such as medical accidents, medical disputes and the like are avoided as much as possible.
< embodiment 2>
On the basis of embodiment 1, the embodiment of the invention provides an intelligent auxiliary method for anesthesia operation, as shown in fig. 7, comprising the following steps:
s10, collecting, transmitting and fusing various physiological index data;
s20, setting a database, and receiving and storing anesthesia case data, basic information data of patients and physiological index data;
s30, acquiring physiological index data, anesthesia case data and basic information data of a patient, and making a physiological index abnormality judgment standard and training a decision model; the decision model is used for providing physiological index health assessment, potential symptom prediction and diagnosis and treatment scheme recommendation;
S40, the doctor adjusts the health assessment, the prediction of potential symptoms and the recommendation of the diagnosis and treatment scheme of the acquired physiological indexes;
s50, a doctor obtains inquiry and play services by setting voice interaction services;
the inquiry and play service comprises the steps of inserting patient reference values, defining an entrance by design index specification, opening or closing physiological index trend analysis, opening or closing diagnosis and treatment scheme suggestion and opening or closing voice prompt.
In the above embodiment, in step S10, the hardware monitoring terminal for collecting various physiological index data includes, but is not limited to, medical devices such as a multi-parameter monitor, a blood gas analyzer, and an anesthesia machine, which are commonly used in anesthesia operation, and can collect various physiological index information of a patient including, but not limited to, vital signs, blood gas index, hemodynamic index, and anesthesia depth index. The transmission and fusion of the data refer to the transmission mode of unifying various physiological index data, formulating unified coding standards of various physiological index data, outputting after fusion, and outputting various physiological index data after fusion, including body temperature, heart rate, pulse rate, blood pressure, blood oxygen saturation and end-of-breath CO 2 Partial pressure, central venous pressure, arterial pressure variation and brain electrical double frequency index. Specifically, the transmission mode of the multiple physiological index data is unified, which includes, but is not limited to, adopting one of Bluetooth, wiFi and 4G/5G as a transmission technology form, and being provided with a corresponding data transmission module Bluetooth gateway or WiFi router or 4G/5G router, so that the data fusion and normalization of the multiple physiological indexes can be facilitated. The unified coding specification of the multiple physiological index data is formulated for packaging and encoding and decoding the multiple physiological indexes, for example, the unified coding specification can be The data package formats of different hardware monitoring terminals are unified, and mainly comprise firmware information, data transmission frequency, data transmission time, data exchange format, data parameter naming, corresponding types and the like, as shown in the following table 1.
TABLE 1 unified coding Specification
The plurality of monitoring units eventually fuse the normalized set of data units according to the unified coding specification as shown in table 2 below.
TABLE 2 fusion-normalized set of data examples
It should be noted that, in table 1 and table 2, the parameter terminal is represented as a set of array data, which represents that the data of 3 hardware monitoring terminals are fused, namely, a monitor "-" jianhuyi ", a blood gas analyzer" - "xueqiyi", and an anaesthesia machine "-" mazuiji "; the array data stored by the parameter mac represents the physical mac addresses corresponding to the 3 types of hardware; the parameter freq represents the data acquisition frequency of hardware interruption, and the acquisition frequency is adjusted according to the actual application requirement; the parameter timeStamp stores unix time stamps, representing specific time of data acquisition; the parameter data represents the fused physiological index data.
In the above embodiment, in step 20, the receiving and storing of the anesthesia case data, the patient basic information data and the physiological index data is implemented by setting a database. Specifically, the database has the functions of receiving and storing, packaging test, API interface and the like. The receiving and storing function is mainly used for receiving and storing the fused multiple physiological index data, anesthesia case data and basic information number of patients According to the above. Here, the basic information data of the patient includes basic information such as the patient's inpatient number, sex, age, weight, etc., and physiological index abnormality determination criterion information corresponding to the patient; the multiple physiological index data comprise vital signs, blood and qi indexes, hemodynamic indexes, anesthesia depth indexes and the like; the fused multiple physiological index data comprise body temperature, heart rate, pulse rate, blood pressure, blood oxygen saturation, and end-of-breath CO 2 Partial pressure, central venous pressure, arterial pressure variation and brain electrical double frequency index BIS; the anesthesia case data is obtained by acquiring abundant anesthesia operation cases through hospital data authority. The encapsulation test function is used for carrying out classified encapsulation, network encryption, authority setting and test analysis on all data in the database. Specifically, the packaging test packages different service data, ensures the data security by means of network encryption technology, data access authority setting and the like, and provides inquiry, storage and editing services in an API interface mode. In addition, the package test function analyzes the performance, availability, consistency and expansibility of the architecture of the database, and ensures the stable and efficient operation of the database.
In the above embodiment, in step S30, a data request is initiated to the database to obtain the physiological index data, the anesthesia case data and the basic information data of the patient, so as to implement the data preparation work before the decision modeling and improve the development efficiency of the decision algorithm. Still further preferably, the data preparation further includes cleaning, integrating, and transforming the data. The data cleaning removes noise data and irrelevant data in the source data set, processes missing data, cleans dirty data and blank values, identifies and deletes isolated points and the like; data integration completes data matching of the same entity; the data transformation is used to find a characteristic representation of the data.
As still further preferable in step S30, as shown in fig. 8, the step of preparing a physiological index abnormality determination criterion includes the steps of:
s31, setting given limit values of health of various physiological indexes as a general specification;
s32, inserting an initial reference value set according to basic information data of a patient on the basis of a given limit value, and setting independent judgment conditions of each physiological index to form a differential judgment standard;
wherein the independent determination condition includes a reference limit value and a trend condition value.
Specifically, the standardization making process needs to perform information interaction, such as voice interaction, for making a physiological index differentiation standard for patients with obvious individual differences, extracting features of the data acquired by the data preparation unit, combining the physiological index differentiation standard, training a decision model, outputting a decision, and making decisions including physiological index health assessment, potential symptom prediction and diagnosis and treatment scheme recommendation. Preferably, the original data can be converted into the features by the feature extraction through a mode of regular normalization or mathematical conversion, and the accuracy of the training model is improved.
As a still further preference of step S30, training the decision model, as shown in fig. 9, comprises the steps of:
S33, performing physiological index health tracking evaluation on various physiological indexes based on the differentiation judgment standard to obtain a physiological index health evaluation model, and outputting various physiological index health states;
s34, searching anesthesia case data according to the health states of various physiological indexes, acquiring physiological index data of a batch of patients and common intraoperative symptom information, and completing label matching of corresponding physiological index information by taking the intraoperative symptom as a label; fitting a physiological index health evaluation model to the physiological index information of the patients in batches, and outputting a classification result of the physiological index health state of the patients; matching the result combination of the physical index health state with the intraoperative symptom label again, training a potential symptom prediction model for the data set, and outputting a prediction result of the potential symptom of the patient;
s35, searching anesthesia case data in a database module according to the health states of various physiological indexes and the potential symptom classification results, obtaining a diagnosis and treatment scheme recommendation model, and giving out diagnosis and treatment scheme recommendation;
specifically, the physiological index health assessment model may be configured to perform physiological index health tracking assessment on multiple physiological indexes based on a physiological index abnormality determination criterion, and output multiple physiological index health states, for example, five physiological index health states: normal, high, low, extremely high, extremely low. A potential symptom prediction model can be set for searching anesthesia case data according to the health states of various physiological indexes, acquiring physiological index data of batch patients and common intraoperative symptom information, and the intraoperative symptom is used as a label to complete label matching of corresponding physiological index information; fitting a physiological index health evaluation model to the physiological index information of the patients in batches, and outputting a classification result of the physiological index health state of the patients; and matching the result combination of the physical index health state with the intraoperative symptom label again, so as to train a potential symptom prediction model for the data set, and outputting a prediction result of the potential symptom of the patient. The diagnosis and treatment scheme recommendation model can be set to search anesthesia case data in the database module according to various physical index health states and the potential symptom classification results, such as allergy, hypoxia, shock, pneumothorax, amniotic fluid embolism, venous air thrombosis and sudden cardiac arrest, and give diagnosis and treatment scheme recommendation, for example, the diagnosis and treatment scheme recommendation model receives the output results of the physical index health assessment model and the intraoperative potential symptom prediction model, takes the symptom output results and the physical index health states as data sources to develop feature analysis, searches the anesthesia case database, applies semantic analysis and numerical positioning technology, and gives diagnosis and treatment scheme recommendation according to satisfaction degree and matching degree sequencing.
The following are carried out according to 9 physiological indexes (body temperature, heart rate, pulse rate, blood pressure, blood oxygen saturation and end-of-breath CO 2 The interactions between the three models are exemplified by partial pressure, central venous pressure, arterial pressure variability, and brain electrical double frequency index BIS. The physiological index abnormality determination specification includes two types: one is a general judgment standard suitable for common people; the other is a differential judgment standard, which is suitable for special patients. Under the general standard, judging whether the physiological index is abnormal or not through a given limit value; the differentiation judgment standard is inserted into the initial standard value according to the condition of the patient, and on the basis of the initial standard value, independent judgment conditions are set for various physiological indexes, so that not only is the given limit value required to be judged, but also the trend development of the indexes is required to be considered, and the reference limit values and the trend conditions of different physiological indexes are independently defined according to the difference of the patient. Patient 9 physiological index numbersAccording to the model fitting, outputting the health state of 5 types of physiological indexes, including: normal, high, low, extremely high, extremely low. The intraoperative potential symptom prediction model takes the output result of the physiological index health assessment model, namely the physiological index health state combination of 9 physiological indexes as input to carry out feature analysis. Searching an anesthesia operation case database to obtain physiological index data of a batch of patients and common intraoperative symptom information, and completing label matching of corresponding physiological index information by taking the intraoperative symptom as a label; fitting a physiological index health evaluation model to the batch of physiological index information, and outputting a potential symptom classification result; the result combination of the 9 physiological index health states is subjected to label matching with the intraoperative symptoms to form a batch data set; training a model by using a random forest, a decision tree, an SVM and other machine learning methods, wherein the model predicts possible symptoms of a patient, and the symptom classification result comprises: allergy, hypoxia, shock, pneumothorax, amniotic fluid embolism, venous air thrombosis, cardiac arrest and the like. The model predicts symptom classifications that substantially cover common symptoms of anesthesia procedures.
The physiological index abnormality determination specification is further described in the following.
The decision criteria of the general specification are given limits for the health of a given set of physiological indicators, which are applicable to all patients of the general population. All the given limit conditions of the physiological index health can be defined in an array form, such as threlist= [ thre0, thre1, thre2, thre3], and if the current physiological index value is < thre0, the index health status is 'very low'; if thre0< = the current physiological index value < thre1, the index health status is "low"; if thre1< = the current physiological index value < = thre2, the index health status is "normal"; if thre2< the current physiological index value < = thre3, the index health status is "high"; if thre3< the current physiological index value, the index health status is "extremely high".
The differentiation judgment standard is formed by combining trend development conditions trendCond and given limit conditions baseCond, all physiological indexes can be defined according to the standard, and different patients have different trend conditions and limit conditions due to individual differences. If the trend of a certain physiological index of the patient is satisfactory, the health state is normal; the trend of the physiological index is unsatisfactory, the physiological index is further judged according to a limiting condition, and the limiting condition is set by referring to an initial reference value of a patient. The differentiated determination criteria can effectively cover a wide variety of patients in a wide variety of surgical situations.
The trend condition is defined as follows, wherein, after trend analysis is started on a certain physiological index, if the monitored value of start < = pct%is < = end, the trend development of the physiological index is satisfied; when start is null (nan), it means that if the monitored value of pct% < = end, the trend of the physiological index is satisfactory; in the case where end is a null value (denoted by nan), this kind of physiological index trend is satisfactory if the monitor value of pct =start. The following are illustrated:
1. the trend condition of the central venous pressure CVP of a certain patient is trendcon= [80%,8cm H20 and 12cm H20], which means that after trend analysis is started, if the central venous pressure monitoring value of 8cm H20< = 80% is < = 12cm H20, the central venous pressure trend is satisfactory, and the reverse trend is unsatisfactory;
2. the trend condition of the heart rate HR of a certain patient is trendcon= [70%,45 times/min, nan ], which means that after trend analysis is started, if the heart rate HR monitoring value of 45 times/min < = 70%, the central venous pressure trend is satisfactory, and the opposite trend is unsatisfactory;
3. the trend condition of Pulse rate Pulse of a certain patient is trendcon = [80%, nan,100 times/min ], which means that after trend analysis is started, if 80% Pulse rate Pulse monitoring value < = 100 times/min, pulse rate Pulse trend is satisfactory, and reverse trend is unsatisfactory.
The limiting value condition form is defined according to an array form, such as baseCond= [ base0, base1, base2, base3], and the limiting value condition baseCond fitting is carried out on the current physiological index value under the condition that certain physiological index trend management is not satisfied. If the current physiological index value is < base0, the index health state is 'extremely low'; if base0< = the current physiological index value < base1, the index health status is "low"; if base1< = the current physiological index value < = base2, the index health status is "normal"; if base2< the current physiological index value < = base3, the index health status is "higher"; if base3 is < the current physiological index value, the index health status is "extremely high".
In this embodiment, different modules focus on implementing different sub-division tasks, and causal association and data coupling exist between the tasks. The modules are mutually matched and evolve layer by layer, and the wide unordered physiological index data is finally output as decisions focused by doctors.
As a further preferable mode of step S30, training the decision model, and training a diagnosis and treatment scheme recommendation model, wherein the decision model is used for receiving the output results of the physiological index health assessment model and the intraoperative potential symptom prediction model, performing feature analysis by taking the symptom output results and the physiological index health state as data sources, searching an anesthesia case database, applying semantic analysis and numerical positioning technology, and sequencing according to satisfaction degree and matching degree to give a diagnosis and treatment scheme recommendation.
In the above embodiment, in step S40, the doctor may implement functions of data acquisition and display, service management, voice interaction, scheme adjustment, iteration, and the like through the doctor application terminal. The data acquisition and display are used for acquiring anesthesia case data and decisions by a doctor and displaying, for example, data information of attention of the doctor, such as a physiological index, predicted symptoms, diagnosis and treatment schemes, text description related to patient information, a data graph related to trend analysis and the like, can be displayed on a UI (user interface) of an application terminal of the doctor. The business management is used for designing and managing data linkage and interaction logic between businesses according to actual use requirements. The voice interaction is used for realizing data linkage and interaction logic through intelligent voice interaction, and the data linkage and interaction logic comprises setting an inserted patient reference value, defining an inlet according to design index specification, opening or closing physiological index trend analysis, opening or closing diagnosis and treatment scheme suggestion, opening or closing voice prompt, opening or closing vital sign abnormality prompt, opening or closing potential symptom prompt and activating voice awakening. Specifically, the voice interaction can be used for defining related content and specific forms of semantic recognition and is responsible for determining related content and specific forms of semantic stitching; the method can be used for realizing the voice function, for example, follow-up development is carried out by means of a third party development platform for the voice of the fly voice, the hundred-degree voice and the Hua voice; the system can also be used for providing inquiry and broadcasting services, and the inquiry function can provide the doctor with the development condition of the physiological index of the patient or the proposal of the diagnosis and treatment scheme at any time; the broadcasting function actively gives out voice prompt when the physiological index is abnormal or some suspicious symptoms exist. The scheme adjustment and iteration are used for checking, adjusting and checking the acquired health states of various physiological indexes, potential symptom classification results and diagnosis and treatment scheme recommendation according to actual clinical experience by a doctor to form a final diagnosis and treatment scheme, and updating and iterating the final diagnosis and treatment scheme to a database, so that on one hand, the accuracy, the scientificity and the safety of a final medical decision are improved; on the other hand, iteration is conducted, the final medical scheme decision is used for reference, the update of the medical decision corresponding to the intraoperative symptom, the physiological index state, the patient information and other data is perfected, and the data is backed up and synchronized to a database. The method is beneficial to improving the performance of the decision model after multiple iterations, thereby providing more accurate and more perfect diagnosis and treatment decision suggestions for doctors.
In the above embodiment, in step S50, the preferred embodiment of the voice module is given in this embodiment. The voice module also comprises an information interaction unit and a second voice interaction unit. The information interaction unit is used for respectively carrying out information data interaction with the database module and the decision module; the second voice interaction unit is used for realizing data linkage and interaction logic with the doctor application terminal through voice, wherein the data linkage and interaction logic comprises the steps of inserting patient reference values, defining an entrance by design index standards, opening or closing physiological index trend analysis, opening or closing diagnosis and treatment scheme suggestion, opening or closing voice prompt, opening or closing vital sign abnormality prompt, opening or closing potential symptom prompt and activating voice awakening.
Embodiment 2 provides an intelligent auxiliary method for anesthesia operation, which can realize data acquisition and fusion normalization analysis of multiple types of physiological indexes according to operation requirements, and the comprehensive operation indexes have important significance for analysis of symptoms of patients in operation and determination of diagnosis and treatment schemes. Based on anesthesia operation cases, a machine learning method is applied to train and complete models such as physical index health assessment, potential symptom prediction, diagnosis and treatment scheme recommendation and the like of patients, and a decision is output. The physiological index abnormality judgment standard based on the individual reference value is provided with two standards, namely a common standard and a differential standard, so that the patient is covered more comprehensively, and the judgment result is more scientific and accurate. Anesthesia surgery involves multiple types of patients, multiple types of surgery, and different patients may have differences in physical quality, or may be affected by certain chronic diseases, or may have a history of some past disease, exhibiting significant individual variability. There are respective criteria for whether different physiological indicators of the patient are in a normal state during the operation. The physiological index abnormality judgment criterion provided by the invention is not only suitable for common patients, but also suitable for patients with obvious individual differences of certain physiological indexes. In addition, the decision model is based on abundant anesthesia operation cases, a machine learning method is applied, models such as health assessment, potential symptom prediction, diagnosis and treatment scheme recommendation and the like of physiological indexes of patients are trained and completed, and the health development condition of the physiological indexes of the patients in operation is assessed in real time; predicting potential symptoms of the patient, and giving out a detailed diagnosis and treatment proposal; the diagnosis and treatment scheme recommendation can be fed back to the database module for iterative update after being adjusted according to the experience of doctors. The decision of the decision model provides real-time intra-operative decision reference for doctors, assists the doctors to make more efficient, safer and more accurate surgical decisions, avoids decision errors caused by self knowledge limitation or attention loss of the doctors, and further avoids adverse events such as medical accidents, medical disputes and the like. Through the voice interaction function, doctors can realize calling data, voice and decision to realize specific service functions. Under the condition that only one anesthesiologist performs anesthesia induction, puncture, nerve block and other operations in an operating room, or under the condition of complex operations such as emergency treatment, rescue and the like, the doctor has limited attention and easily ignores the index development condition and potential suspicious symptoms of the patient. In the voice interaction mode, if a physiological index abnormality occurs or a certain suspicious symptom exists in a patient in operation, a voice prompt is actively given; in addition, the voice query function provides advice for doctors on the development condition of the physiological index of the patient or the diagnosis and treatment scheme at any time. The interaction form can effectively lighten the attention burden of doctors and improve the safety of operations.
In summary, the anesthesia surgery is complicated in coverage, and involves various patients, and the surgery process is affected by various sudden, urgent and uncontrollable factors. The intelligent auxiliary method for anesthesia surgery provided by the invention applies the data fusion technology, the intelligent voice technology, the artificial intelligence technology and the Internet technology, so that the attention burden of the doctor in surgery is reduced to the greatest extent, a real-time decision reference is provided for the doctor, the doctor is assisted to make more efficient, safer and more accurate intra-surgery decisions, and adverse events such as medical accidents, medical disputes and the like are avoided as much as possible.
It is apparent that the above examples are given by way of illustration only and are not limiting of the embodiments. Other variations or modifications of the above teachings will be apparent to those of ordinary skill in the art. It is not necessary here nor is it exhaustive of all embodiments. And obvious variations or modifications thereof are contemplated as falling within the scope of the present invention.
Claims (9)
1. An intelligent assistance system for anesthesia procedures, comprising:
the hardware monitoring terminal is used for collecting, transmitting and fusing various physiological index data;
A database module for receiving and storing anesthesia case data and patient base information data and the physiological index data;
the decision module is used for acquiring the physiological index data, the anesthesia case data and the basic information data of the patient, making physiological index differentiation standard and training a decision model; the decision model is used for providing physiological index health assessment, potential symptom prediction and diagnosis and treatment scheme recommendation;
the doctor application terminal is used for acquiring the physiological index health assessment, the potential symptom prediction, the diagnosis and treatment scheme recommendation and performing adjustment through a doctor;
the voice module is respectively in communication connection with the database module, the decision module and the doctor application terminal and is used for data linkage and providing voice interaction services including inquiry and play for the doctor application terminal;
the decision module comprises:
a data preparation unit for acquiring the physiological index data, the anesthesia case data, and the patient base information data;
a specification making unit for making a physiological index abnormality determination specification for the data acquired by the data preparing unit; the method comprises the steps of,
The feature extraction and decision model training unit is used for extracting features of the data acquired by the data preparation unit, combining the physiological index abnormality judgment standard, training a decision model and outputting a decision, wherein the decision comprises the physiological index health assessment, the potential symptom prediction and the diagnosis and treatment scheme recommendation;
wherein the physiological index abnormality determination specification comprises a general specification and a differentiation determination specification; the universal specification includes setting a given limit for health of a plurality of the physiological indicators; the differential judgment standard comprises inserting an initial reference value set according to basic information data of a patient on the basis of the given limit value, and setting independent judgment conditions of each physiological index; the independent determination condition includes a reference limit value and a trend condition value.
2. The intelligent assistance system for anesthesia surgery of claim 1 wherein the hardware monitoring terminal comprises:
a monitoring unit for collecting a plurality of the physiological index data,
the transmission and fusion unit is used for unifying transmission modes of a plurality of physiological index data, formulating unified coding standards of the plurality of physiological index data, fusing and outputting;
Wherein the plurality of physiological indexes at least comprise vital signs, blood gas indexes, hemodynamic indexes and anesthesia depth indexes;
the multiple physiological index data output after fusion comprise body temperature, heart rate, pulse rate, blood pressure, blood oxygen saturation, and end-of-breath CO 2 Partial pressure, central venous pressure, arterial pressure variation and brain electrical double frequency index.
3. The intelligent assistance system for anesthesia surgery of claim 1 wherein the database module comprises:
the database unit is used for storing the fused multiple physiological index data, the anesthesia case data and the basic information data of the patient;
the packaging test unit is used for carrying out classified packaging, network encryption, authority setting and test analysis on all data of the database unit; the method comprises the steps of,
and the API interface unit is used for providing inquiry, storage and editing services of the data.
4. The intelligent assistance system for an anesthesia procedure of claim 1 wherein the decision model comprises:
the physiological index health evaluation model is used for carrying out physiological index health tracking evaluation on a plurality of physiological indexes based on the physiological index abnormality judgment standard and outputting a plurality of physiological index health states;
The potential symptom prediction model is used for retrieving the anesthesia case data, acquiring physiological index data of a batch of patients and common intraoperative symptom information, and completing label matching of corresponding physiological index information by taking the intraoperative symptom as a label; fitting the physiological index health evaluation model to the physiological index information of the patients in batches, and outputting the classification result of the physiological index health state of the patients; matching the result combination of the physical index health state with the intraoperative symptom label again, training a potential symptom prediction model for the data set, and outputting a prediction result of the potential symptom of the patient; the method comprises the steps of,
a diagnosis and treatment plan recommendation model obtained by retrieving the anesthesia case data in the database module according to the plurality of physiological index health states and the potential symptom classification result, the diagnosis and treatment plan recommendation model being used for giving a diagnosis and treatment plan recommendation;
wherein the health states of the multiple physiological indexes comprise normal, high, low, extremely high and extremely low;
the potential symptom classification includes allergy, hypoxia, shock, pneumothorax, amniotic fluid embolism, venous air thrombosis and cardiac arrest.
5. The intelligent assistance system for anesthesia surgery of claim 1 wherein the doctor application terminal further comprises:
The data acquisition and display unit is used for acquiring the anesthesia case data and the decision and display data;
the business management unit is used for designing and managing data linkage and interaction logic among the units or modules according to actual use requirements;
the first voice interaction unit is used for realizing the data linkage and interaction logic with the voice module through voice; the method comprises the steps of,
the scheme adjusting and iterating unit is used for checking, adjusting and checking the acquired health states of the multiple physiological indexes, the potential symptom classification result and the diagnosis and treatment scheme recommendation according to actual clinical experience by a doctor to form a final diagnosis and treatment scheme, and updating and iterating the final diagnosis and treatment scheme to the database module;
the data linkage and interaction logic comprises setting an inserted patient reference value, defining an inlet by a design index specification, opening or closing a physiological index trend analysis, opening or closing a diagnosis and treatment scheme suggestion, opening or closing a voice prompt, opening or closing a vital sign abnormality prompt, opening or closing a potential symptom prompt and activating voice awakening.
6. The intelligent assistance system for an anesthesia procedure of claim 1 wherein the voice module further comprises:
The information interaction unit is used for respectively carrying out information data interaction with the database module and the decision module;
the second voice interaction unit is used for realizing data linkage and interaction logic with the doctor application terminal through voice;
the data linkage and interaction logic comprises setting an inserted patient reference value, defining an inlet by a design index specification, opening or closing a physiological index trend analysis, opening or closing a diagnosis and treatment scheme suggestion, opening or closing a voice prompt, opening or closing a vital sign abnormality prompt, opening or closing a potential symptom prompt and activating voice awakening.
7. A method of performing anesthesia procedure assistance using the intelligent assistance system for anesthesia procedures of any one of claims 1-6, comprising the steps of:
collecting, transmitting and fusing various physiological index data;
setting a database, and receiving and storing anesthesia case data, basic information data of patients and the physiological index data;
acquiring the physiological index data, the anesthesia case data and the basic information data of the patient, and making a physiological index abnormality judgment standard and a training decision model; the decision model is used for providing physiological index health assessment, potential symptom prediction and diagnosis and treatment scheme recommendation;
The doctor adjusts the acquired physiological index health assessment, the potential symptom prediction and the diagnosis and treatment scheme recommendation;
through setting the voice interaction service, a doctor obtains the inquiry and play service;
the inquiry and play service comprises the steps of inserting patient reference values, defining an entrance by design index specification, opening or closing physiological index trend analysis, opening or closing diagnosis and treatment scheme suggestion and opening or closing voice prompt.
8. The intelligent assistance method for performing an anesthesia procedure of claim 7 wherein establishing a physiological index abnormality determination specification comprises the steps of:
setting given limit values of the health of a plurality of physiological indexes as a general specification;
on the basis of the given limit value, inserting an initial reference value set according to basic information data of patients, and setting independent judgment conditions of each physiological index to form a differential judgment standard;
wherein the independent determination condition includes a reference limit value and a trend condition value.
9. The intelligent assistance method for conducting an anesthesia procedure of claim 7 wherein training a decision model comprises the steps of:
performing physiological index health tracking evaluation on a plurality of physiological indexes based on the differentiation judgment standard to obtain a physiological index health evaluation model, and outputting a plurality of physiological index health states;
According to the health states of the multiple physiological indexes, retrieving the anesthesia case data, acquiring physiological index data of a batch of patients and common intraoperative symptom information, and completing label matching of corresponding physiological index information by taking the intraoperative symptom as a label; fitting a physiological index health evaluation model to the physiological index information of the patients in batches, and outputting a classification result of the physiological index health state of the patients; matching the result combination of the physical index health state with the intraoperative symptom label again, training a potential symptom prediction model for the data set, and outputting a prediction result of the potential symptom of the patient;
according to the health states of the multiple physiological indexes and the potential symptom classification result, anesthesia case data in the database module are searched, a diagnosis and treatment scheme recommendation model is obtained, and diagnosis and treatment scheme recommendation is given;
wherein the health states of the multiple physiological indexes comprise normal, high, low, extremely high and extremely low;
the potential symptom classification includes allergy, hypoxia, shock, pneumothorax, amniotic fluid embolism, venous air thrombosis and cardiac arrest.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111677791.8A CN114366030B (en) | 2021-12-31 | 2021-12-31 | Intelligent auxiliary system and method for anesthesia operation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111677791.8A CN114366030B (en) | 2021-12-31 | 2021-12-31 | Intelligent auxiliary system and method for anesthesia operation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114366030A CN114366030A (en) | 2022-04-19 |
| CN114366030B true CN114366030B (en) | 2024-04-09 |
Family
ID=81141343
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111677791.8A Active CN114366030B (en) | 2021-12-31 | 2021-12-31 | Intelligent auxiliary system and method for anesthesia operation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114366030B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115148327B (en) * | 2022-07-08 | 2023-07-28 | 徐州医科大学 | Multi-party collaborative prototype system oriented to intelligent anesthesia decision and method thereof |
| CN115910323B (en) * | 2022-11-16 | 2023-09-29 | 山东仲雅信息技术有限公司 | Treatment scheme selection method, device, equipment and storage medium |
| CN116312958B (en) * | 2023-05-24 | 2023-09-15 | 成都市龙泉驿区中医医院 | Anesthesia risk early warning system, emergency management system and method |
| CN116530943B (en) * | 2023-07-05 | 2023-09-22 | 深圳市益心达医学新技术有限公司 | Anesthesia depth detection device based on blood gas data |
| CN117766152A (en) * | 2023-07-20 | 2024-03-26 | 重庆市妇幼保健院(重庆市妇产科医院、重庆市遗传与生殖研究所) | Method for intelligently detecting link drop of general anesthesia intravenous pipeline |
Citations (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101826136A (en) * | 2004-01-06 | 2010-09-08 | 株式会社东芝 | Health care support system and health care support device |
| JP2018175840A (en) * | 2017-11-02 | 2018-11-15 | 芙蓉開発株式会社 | Software, health condition determination device, and health condition determination method |
| CN109310345A (en) * | 2016-09-05 | 2019-02-05 | 深圳迈瑞生物医疗电子股份有限公司 | Medical monitoring device, patient monitoring information display method and system |
| CN110415831A (en) * | 2019-07-18 | 2019-11-05 | 天宜(天津)信息科技有限公司 | A kind of medical treatment big data cloud service analysis platform |
| CN110974201A (en) * | 2019-12-26 | 2020-04-10 | 南通大学附属医院 | Wireless medical monitoring system |
| CN111477314A (en) * | 2020-06-04 | 2020-07-31 | 郭文军 | Terminal medical equipment data acquisition device, acquisition method and acquisition system |
| CN111481174A (en) * | 2020-04-15 | 2020-08-04 | 中科搏锐（北京）科技有限公司 | Anesthesia and consciousness depth monitoring system and method |
| CN111798956A (en) * | 2020-06-28 | 2020-10-20 | 深圳市美格尔医疗技术研究院 | Decision determination method, device and system for artificial intelligence anesthesia |
| CN111863236A (en) * | 2019-04-24 | 2020-10-30 | 通用电气精准医疗有限责任公司 | Medical machine composite data and corresponding event generation |
| CN111839471A (en) * | 2020-06-02 | 2020-10-30 | 刘顺利 | Bedside intelligent accurate medical auxiliary system |
| CN112768044A (en) * | 2021-01-25 | 2021-05-07 | 深圳市即达健康医疗科技有限公司 | Intelligent operation anesthesia management system |
| CN112807542A (en) * | 2021-01-11 | 2021-05-18 | 南通市第二人民医院 | Method and device for controlling dosage of anesthetic |
| WO2021110879A1 (en) * | 2019-12-06 | 2021-06-10 | École Normale Superieure Paris-Saclay | System and method for predicting depth of anesthesia |
| CN113100709A (en) * | 2021-03-26 | 2021-07-13 | 攀枝花市中心医院 | Anesthesia depth monitoring system and anesthesia depth monitoring method |
| CN113284576A (en) * | 2021-06-30 | 2021-08-20 | 南通市妇幼保健院 | Intelligent collection method and system for physical sign data of emergency patients |
| CN113795190A (en) * | 2019-03-27 | 2021-12-14 | 麻省总医院 | Intraoperative clinical decision support system |
| WO2021260171A2 (en) * | 2020-06-26 | 2021-12-30 | Intelligence Anesthesia | System and method for peri-anaesthetic risk evaluation |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8990135B2 (en) * | 2010-06-15 | 2015-03-24 | The Regents Of The University Of Michigan | Personalized health risk assessment for critical care |
| CA2888185A1 (en) * | 2012-10-12 | 2014-04-17 | The Cleveland Clinic Foundation | Monitoring severity and duration of abberant physiological parameters during a procedure |
| US20180082036A1 (en) * | 2016-09-22 | 2018-03-22 | General Electric Company | Systems And Methods Of Medical Device Data Collection And Processing |
| US20180211011A1 (en) * | 2017-01-25 | 2018-07-26 | General Electric Company | Anesthesia assessment system and method for enhanced recovery after surgery qualification |
-
2021
- 2021-12-31 CN CN202111677791.8A patent/CN114366030B/en active Active
Patent Citations (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101826136A (en) * | 2004-01-06 | 2010-09-08 | 株式会社东芝 | Health care support system and health care support device |
| CN109310345A (en) * | 2016-09-05 | 2019-02-05 | 深圳迈瑞生物医疗电子股份有限公司 | Medical monitoring device, patient monitoring information display method and system |
| JP2018175840A (en) * | 2017-11-02 | 2018-11-15 | 芙蓉開発株式会社 | Software, health condition determination device, and health condition determination method |
| CN113795190A (en) * | 2019-03-27 | 2021-12-14 | 麻省总医院 | Intraoperative clinical decision support system |
| CN111863236A (en) * | 2019-04-24 | 2020-10-30 | 通用电气精准医疗有限责任公司 | Medical machine composite data and corresponding event generation |
| CN110415831A (en) * | 2019-07-18 | 2019-11-05 | 天宜(天津)信息科技有限公司 | A kind of medical treatment big data cloud service analysis platform |
| WO2021110879A1 (en) * | 2019-12-06 | 2021-06-10 | École Normale Superieure Paris-Saclay | System and method for predicting depth of anesthesia |
| CN110974201A (en) * | 2019-12-26 | 2020-04-10 | 南通大学附属医院 | Wireless medical monitoring system |
| CN111481174A (en) * | 2020-04-15 | 2020-08-04 | 中科搏锐（北京）科技有限公司 | Anesthesia and consciousness depth monitoring system and method |
| CN111839471A (en) * | 2020-06-02 | 2020-10-30 | 刘顺利 | Bedside intelligent accurate medical auxiliary system |
| CN111477314A (en) * | 2020-06-04 | 2020-07-31 | 郭文军 | Terminal medical equipment data acquisition device, acquisition method and acquisition system |
| WO2021260171A2 (en) * | 2020-06-26 | 2021-12-30 | Intelligence Anesthesia | System and method for peri-anaesthetic risk evaluation |
| CN111798956A (en) * | 2020-06-28 | 2020-10-20 | 深圳市美格尔医疗技术研究院 | Decision determination method, device and system for artificial intelligence anesthesia |
| CN112807542A (en) * | 2021-01-11 | 2021-05-18 | 南通市第二人民医院 | Method and device for controlling dosage of anesthetic |
| CN112768044A (en) * | 2021-01-25 | 2021-05-07 | 深圳市即达健康医疗科技有限公司 | Intelligent operation anesthesia management system |
| CN113100709A (en) * | 2021-03-26 | 2021-07-13 | 攀枝花市中心医院 | Anesthesia depth monitoring system and anesthesia depth monitoring method |
| CN113284576A (en) * | 2021-06-30 | 2021-08-20 | 南通市妇幼保健院 | Intelligent collection method and system for physical sign data of emergency patients |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114366030A (en) | 2022-04-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114366030B (en) | Intelligent auxiliary system and method for anesthesia operation | |
| US20220037038A1 (en) | Operating room checklist system | |
| CN105792731A (en) | Patient care surveillance system and method | |
| EP3380963B1 (en) | Tracking usage of a pulse oximeter via a network system | |
| US20190392952A1 (en) | Computer-implemented methods, systems, and computer-readable media for diagnosing a condition | |
| US20190108264A1 (en) | System prototype of intelligent regional mobile healthcare integrated data center based on metadata | |
| CN114420231B (en) | Interpretable continuous early warning method and system for acute kidney injury, storage medium and electronic equipment | |
| Gøeg et al. | Content analysis of physical examination templates in electronic health records using SNOMED CT | |
| CN113113136A (en) | Medical aid decision-making system based on mixed reality | |
| CN114023423A (en) | Intelligent shift switching method, device and equipment | |
| US20220310219A1 (en) | Medical record digest | |
| WO2022141925A1 (en) | Intelligent medical service system and method, and storage medium | |
| CN117475447A (en) | Intelligent severe infection early warning and auxiliary decision making system based on multi-modal data fusion federal learning | |
| CN116597954B (en) | Intelligent home-based remote diagnosis and treatment appointment consultation system | |
| CN104873171B (en) | A kind of medical system and medical information transmission method | |
| Liu | Remote intelligent medical monitoring system based on Internet of Things | |
| CN115171901A (en) | Intelligent system and method for difficult airway assessment and intraoperative airway management assistance | |
| CN111768856B (en) | Bidirectional referral automatic judgment and aftereffect evaluation system based on data driving | |
| Zhang | Real-time analysis of physiological data and development of alarm algorithms for patient monitoring in the intensive care unit | |
| US20160371457A1 (en) | System and Method for Data Analyzing of Health-Related Data | |
| US20200035361A1 (en) | Method and electronic device for artificial intelligence (ai)-based assistive health sensing in internet of things network | |
| US11361864B2 (en) | Tracking usage of a pulse oximeter via a network system | |
| US20210151178A1 (en) | System, apparatus and method for monitoring clinical state of a user | |
| CN115862870B (en) | Chronic disease management system and method based on artificial intelligence | |
| Bianchi et al. | Biomedical data integration–capturing similarities while preserving disparities |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |