CN113393946B - Medical clinical test auxiliary system and method based on virtual reality technology - Google Patents
Medical clinical test auxiliary system and method based on virtual reality technology Download PDFInfo
- Publication number
- CN113393946B CN113393946B CN202110939785.9A CN202110939785A CN113393946B CN 113393946 B CN113393946 B CN 113393946B CN 202110939785 A CN202110939785 A CN 202110939785A CN 113393946 B CN113393946 B CN 113393946B
- Authority
- CN
- China
- Prior art keywords
- clinical
- data
- virtual reality
- individual
- medical
- 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
- 238000012360 testing method Methods 0.000 title claims abstract description 80
- 238000000034 method Methods 0.000 title claims abstract description 63
- 238000005516 engineering process Methods 0.000 title claims abstract description 27
- 239000003814 drug Substances 0.000 claims abstract description 94
- 229940079593 drug Drugs 0.000 claims abstract description 55
- 230000002452 interceptive effect Effects 0.000 claims abstract description 41
- 238000013528 artificial neural network Methods 0.000 claims abstract description 36
- 230000006399 behavior Effects 0.000 claims abstract description 26
- 230000010354 integration Effects 0.000 claims abstract description 21
- 230000000857 drug effect Effects 0.000 claims abstract description 14
- 238000007405 data analysis Methods 0.000 claims abstract description 11
- 238000010801 machine learning Methods 0.000 claims abstract description 8
- 230000000638 stimulation Effects 0.000 claims description 35
- 238000013135 deep learning Methods 0.000 claims description 25
- 230000009471 action Effects 0.000 claims description 23
- 230000008569 process Effects 0.000 claims description 20
- 230000000694 effects Effects 0.000 claims description 19
- 238000013179 statistical model Methods 0.000 claims description 18
- 230000003993 interaction Effects 0.000 claims description 15
- 238000013461 design Methods 0.000 claims description 11
- 238000004458 analytical method Methods 0.000 claims description 9
- 238000013500 data storage Methods 0.000 claims description 8
- 230000003287 optical effect Effects 0.000 claims description 8
- 238000004364 calculation method Methods 0.000 claims description 7
- 238000006243 chemical reaction Methods 0.000 claims description 7
- 238000012545 processing Methods 0.000 claims description 6
- 230000003542 behavioural effect Effects 0.000 claims description 5
- 238000012549 training Methods 0.000 claims description 5
- 238000012216 screening Methods 0.000 claims description 4
- 208000019901 Anxiety disease Diseases 0.000 claims description 3
- 230000001133 acceleration Effects 0.000 claims description 3
- 230000036506 anxiety Effects 0.000 claims description 3
- 238000009223 counseling Methods 0.000 claims description 3
- 230000008921 facial expression Effects 0.000 claims description 3
- 230000001755 vocal effect Effects 0.000 claims description 2
- 238000011161 development Methods 0.000 abstract description 10
- 210000005036 nerve Anatomy 0.000 abstract description 6
- 229940121657 clinical drug Drugs 0.000 abstract description 5
- 230000006870 function Effects 0.000 abstract description 4
- 230000004044 response Effects 0.000 description 7
- 230000007613 environmental effect Effects 0.000 description 5
- 238000004891 communication Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000009509 drug development Methods 0.000 description 3
- 230000001537 neural effect Effects 0.000 description 3
- 239000002547 new drug Substances 0.000 description 3
- 239000003086 colorant Substances 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 201000010099 disease Diseases 0.000 description 2
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 230000008520 organization Effects 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 239000004576 sand Substances 0.000 description 2
- 230000009897 systematic effect Effects 0.000 description 2
- 238000001467 acupuncture Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 239000008280 blood Substances 0.000 description 1
- 210000004369 blood Anatomy 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000013144 data compression Methods 0.000 description 1
- 238000013499 data model Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000009699 differential effect Effects 0.000 description 1
- 238000007877 drug screening Methods 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 238000013100 final test Methods 0.000 description 1
- 239000011521 glass Substances 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000003340 mental effect Effects 0.000 description 1
- 230000008035 nerve activity Effects 0.000 description 1
- 210000000653 nervous system Anatomy 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 210000000056 organ Anatomy 0.000 description 1
- 230000008092 positive effect Effects 0.000 description 1
- 239000000523 sample Substances 0.000 description 1
- 230000026676 system process Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H70/00—ICT specially adapted for the handling or processing of medical references
- G16H70/40—ICT specially adapted for the handling or processing of medical references relating to drugs, e.g. their side effects or intended usage
Abstract
The invention relates to a medical clinical test auxiliary system and a method based on a virtual reality technology, wherein the system comprises a medical clinical test hardware system and medical clinical test software, the medical clinical test hardware system consists of an interactive integration unit and a data integration unit, and the clinical test software consists of an application program and a data analysis system; the method comprises the following steps: establishing a test scheme before a newly developed drug enters a clinical test link; establishing a prediction model for each clinical individual through a machine learning neural network according to the individual behaviors and the bioelectrical signals of the clinical individuals; grouping clinical individuals to implement medicaments, and obtaining clinical data through tests; the influence of non-drugs is eliminated from clinical data, and drug effect data which is only dominated by drug factors is obtained. The invention provides clinical test auxiliary function for the development of nerve drugs by adopting a virtual reality technology, obtains the key data of the effectiveness of clinical drugs and helps a medicine developer to reduce the technical risk in clinical tests.
Description
Technical Field
The invention belongs to the technical field of information, relates to a medical clinical test system, and particularly relates to a medical clinical test auxiliary system and method based on a virtual reality technology.
Background
In the process of medicine development, the development of nerve medicines is one of the most risky medicine development categories of the current medicine production enterprises. The process is full of a large amount of uncertain risks, and medicine development failure is caused by the reason that clinical test link data are not ideal enough after medical enterprises consume huge capital. Compared with the development of other common disease drugs, the neural drugs inevitably process a large amount of additional interference in the clinical test link, so that the clinical data cannot reach the expectation, besides the predictable risks of the conventional drug action mechanism, the original drug screening and the like.
In a clinical trial of drugs, after the administration of drugs to clinical individuals (diseased or healthy volunteers), very significant differential effects of the same drug at the same dose often occur on different individuals due to different preferences of individual differences with respect to light, noise, indoor environment, exposed personnel and events. For example, the same piece of information may sometimes produce very distinct psychological responses in different clinical individuals. This different psychological response is reflected not only in the clinical individual's difference in the result of a particular observation measurement, but also in the level of neuronal activity and in the difference in bioelectrical signal data.
Whether new drugs entering a clinical test link effectively interfere the neural activity of a test object or not is judged, and due to the fact that diversified identities and backgrounds exist in clinical individuals, the test result may have large discreteness, and further in the process of transverse comparison, the final test result cannot be confirmed on a statistical model due to the high discreteness of data, and finally drug development failure is caused.
Disclosure of Invention
The invention aims to overcome the defects of the prior art, provides a medical clinical test auxiliary system and method based on a virtual reality technology, and can provide more comprehensive auxiliary functions of environmental factor control, an electrical stimulation implementation scheme, feedback data collection, data analysis and processing and the like for a medical clinical test.
The invention solves the technical problems in the prior art by adopting the following technical scheme:
a medical clinical test auxiliary system based on virtual reality technology comprises a medical clinical test hardware system and medical clinical test software, wherein the medical clinical test hardware system is composed of an interactive integration unit and a data integration unit, and the medical clinical test software is composed of an application program and a data analysis system;
the interaction integration unit comprises virtual reality equipment, bioelectricity signal acquisition equipment and electric signal stimulation implementation equipment: the virtual reality equipment interacts with the clinical individual and transmits interaction data to the application program, the electric signal stimulation implementation equipment implements electric signals on a specific part of a human body, and the bioelectrical signal acquisition equipment collects bioelectrical signals generated by the clinical individual and transmits the bioelectrical signals to the data integration unit;
the data integration unit comprises a data processing circuit and a data storage device, the data processing circuit processes the biological electric signal and marks time to send the biological electric signal to the data storage device, and interactive data transmitted by the virtual reality device is coded and compressed and then is marked with time to send the interactive data to the data storage device;
the application program comprises a standard virtual reality interactive test content module and a bioelectricity stimulation signal control algorithm and is used for delivering interactive scene data to the interactive integrated unit, managing the interactive progress of clinical individuals and transmitting behavior activity data and bioelectricity signal activity data of the clinical individuals to the data analysis system;
the data analysis system comprises a database system and a machine learning neural network, wherein the database system is responsible for managing data stored on the storage equipment; the machine learning neural network is used for establishing a deep learning neural network prediction model and training clinical individual behavior data and bioelectricity signal activity data to obtain data predicted by the deep learning neural network prediction model, and the medicine effect data which is only dominated by medicine factors is obtained by adopting a item-by-item difference solving method.
Furthermore, the interactive integrated unit and the data integrated unit are of a split structure or an integrated structure, the interactive integrated unit of the split structure is worn by a user independently, and the data integrated unit and the interactive integrated unit are placed in a split mode.
Further, the interactive integrated unit further comprises a mechanical sensor and an optical sensor; the mechanical sensor is used for measuring limb action signals generated when a clinical individual interacts with the virtual reality equipment, the limb action signals comprise a motion direction, a motion acceleration, rotation and vibration, and the behavior and the action of the clinical individual are restored by using a general mechanical analysis method; the optical sensor is used for optically detecting the external environment of the clinical individual and photographing the clinical individual, and the facial expression and behavior action of the clinical individual generated in the process of using the virtual reality equipment are analyzed by adopting a computer vision analysis technology.
A medical clinical test auxiliary method based on virtual reality technology is applied to a medical clinical test auxiliary system based on virtual reality technology, and comprises the following steps:
step 1, establishing a test scheme before a newly developed drug enters a clinical test link;
step 2, establishing a deep learning neural network prediction model for each clinical individual according to the individual behaviors of the clinical individuals and the bioelectrical signals;
step 3, grouping clinical individuals to implement the medicine, and obtaining clinical data through tests;
and 4, predicting data of the clinical individuals under the non-drug action by using the deep learning neural network prediction model, and removing the data predicted by the deep learning neural network prediction model from the clinical data to obtain drug effect data which is dominated by drug factors.
Further, step 4 is followed by:
and 5, outputting the drug effect data with the drug factors occupying the leading position to a medical clinical statistical model, and obtaining a clinical test conclusion after statistics of the medical clinical statistical model.
Further, the specific implementation method of step 1 is as follows: before a newly developed medicine enters a clinical test link, a medicine developer designs a clinical test flow according to expected data and a statistical model, wherein the clinical test flow comprises a plan for determining the number of recruited healthy/patient volunteers and screening and/or control groups; determining a virtual reality scene and an electrical stimulation interference scheme which are expected to be implemented; adjusting design elements of a virtual reality scene and input parameters of an electrical stimulation scheme; the input parameters include color, sound, social pressure, level of fear, level of anxiety, difficulty of virtual tasks, applied pulse voltage, and pulse delay.
Further, the specific implementation method of step 2 is as follows: under the interference of a pre-established virtual reality scene and an electrical stimulation scheme, collecting feedback data of clinical individuals and establishing a data archive base according to individual numbers; the collected data is used as historical data to be marked and sent to a computer neural network, and weight relation calculation is carried out on the influence of input parameters on various clinical reactions; using 80% of the historical data to train the prediction model, and using the remaining 20% to test the effectiveness of the model; each trained prediction model is sent into a system storage device for storage according to the serial number of the clinical individual;
the bioelectricity signals comprise electroencephalogram signals, electromyogram signals and electrocardiosignals; the individual behaviors include questionnaires, verbal interactions, feedback from medical care, psychological counseling, daily behavioral observations, and prescribed designed limb movements.
Further, the specific implementation method of step 3 is as follows: grouping clinical individuals according to a design scheme of a clinical test; administering the medication to a group of clinical individuals; the clinical individual starts a test after being administered with the drug, interacts with the virtual reality scene and receives electrical stimulation interference according to a plan; the trial data was recorded and time-stamped to obtain clinical data.
Further, the specific implementation method of step 4 is as follows: virtual reality interactive contents delivered to clinical individuals and electric stimulation applied according to a plan are taken as input parameters and are sent to the deep learning neural network prediction model trained in the step 2; predicting the behavior response and the neurobioelectricity signals of each clinical individual one by one according to the serial number of the clinical individual; the predicted data is stored according to class classification and is used as the data of the clinical individual under the non-drug action; according to data predicted by the deep learning neural network and clinical data, the drug effect data which is solely dominated by drug factors is obtained after item-by-item difference solving is carried out by adopting an item-by-item difference solving method.
Further, the item-by-item difference solving method is realized by adopting the following formula:
drug effect data = clinical data-data predicted by deep learning neural network predictive models.
The invention has the advantages and positive effects that:
1. the invention provides a medical clinical test auxiliary function for the development of nerve drugs by adopting a virtual reality technology, can collect clinical individual behaviors and bioelectricity signals and carry out systematic analysis after a certain dose of drugs are applied to clinical individuals in a clinical test stage, calculates the effectiveness key data of the current clinical drugs, sends the effectiveness key data of the clinical drugs into a statistical model algorithm used by a drug developer, forms a final conclusion on the developed new drugs in the clinical test, and helps the drug developer reduce the technical risk in the clinical test.
2. In the actual clinical process, clinical drugs and controlled environment variables are applied to clinical individuals, meanwhile, the controlled environment variables are synchronously sent to a prediction model, the reaction of each clinical individual under the condition of not using the drugs is predicted, data predicted by the prediction model are removed from clinical data, drug effect data which are only dominated by drug factors are obtained, the final conclusion on the effectiveness of the clinical drugs is obtained, the result is accurate and reliable, and the method can be widely applied to the development work of drugs for the nervous system.
Drawings
FIG. 1 is a schematic diagram of the medical clinical trial auxiliary system connection of the present invention;
FIG. 2 is a schematic diagram of an auxiliary method for clinical trials of medicine according to the present invention;
FIG. 3(a) is a virtual reality scene input parameter column representation intent;
FIG. 3(b) is a schematic diagram of a list of electrical stimulation disturbance input parameters;
FIG. 4 is a schematic illustration of the organization and method of practice of the present invention during a medical clinical trial.
Detailed Description
The embodiments of the present invention will be described in detail with reference to the accompanying drawings.
A medical clinical trial auxiliary system based on virtual reality technology, as shown in fig. 1, includes a medical clinical trial hardware system and medical clinical trial software. The input end of the clinical trial auxiliary system is used by clinical trial volunteers designed and organized by a medicine developer. The clinical trial volunteers are persons who enter clinical trials through certain screening conditions, namely clinical individuals, and the clinical individuals refer to each volunteer who participates in clinical trials. According to the statistical model design of clinical tests of a medicine developer, clinical test volunteers are grouped and wear the auxiliary system for the medical clinical tests. The clinical test auxiliary system collects, analyzes and arranges data in the clinical test process of the medicine, so as to calculate the key data of the effectiveness of the current clinical medicine. The output end of the clinical trial auxiliary system sends a group of processed data, and the quantitative figures describe the specific influence of the medicine on clinical individuals in the clinical process. The quantified data does not contain conclusions of the medical clinical trial. The data are transmitted to a statistical model designed by a medicine developer to be calculated, and the medicine developer obtains a final clinical test conclusion.
The medical clinical test hardware system is composed of an interactive integrated unit and a data integrated unit. The two hardware units can be split systems, namely the interactive integrated unit is worn by clinical individuals independently, and the data integrated unit is placed in a split mode; or a highly integrated system worn by the clinical individual.
The interaction integration unit comprises virtual reality equipment, bioelectricity signal acquisition equipment, electric signal stimulation implementation equipment, a mechanical sensor and an optical sensor. The interactive integration unit delivers scene interactive contents to the individual through a virtual reality device (such as virtual reality glasses), wherein the scene interactive contents comprise sound and visual information. During the interaction of the clinical individual with the virtual reality scene, the virtual reality device of the interaction integration unit returns data of the clinical individual interaction, including but not limited to questionnaires, voice, action, response time, and the like. The electric signal stimulation implementing equipment in the interactive integrated system can implement voltage or current stimulation including direct current or pulse on a specific part of a human body according to a system instruction in a correct wearing mode. And the bioelectrical signal acquisition equipment of the interaction integration unit collects bioelectrical signals generated by clinical individuals in the interaction process and under the action of the electric signal stimulation implementation equipment and simply processes and stores data. The mechanical sensor in the interactive integrated unit is used for measuring limb action signals generated when the clinical individual interacts with the virtual reality equipment, the limb action signals comprise movement directions, movement acceleration, rotation, vibration and the like, and the behaviors and actions of the clinical individual can be restored by using a general mechanical analysis method. The optical sensor in the interactive integrated unit is used for optically detecting the external environment of the clinical individual, photographing the surrounding environment through optical detection, and calculating the current spatial parameters of the clinical individual according to the photographing result by a computer vision analysis technology, wherein the spatial parameters comprise the distance between the clinical individual and an obstacle, the size of the space where the clinical individual is located and external light, so that the clinical individual is not collided with the external obstacle when using virtual reality equipment, the safety is ensured, and the daily living environment and habit of the clinical individual are also learned; the optical sensor can also take a picture of a clinical individual, and the facial expressions and behavior actions of the clinical individual generated in the process of using the virtual reality equipment are further learned through a computer vision analysis technology.
The data integration unit is composed of a plurality of data processing circuits and data storage equipment, and specifically comprises a noise reduction filtering unit, an analog-to-digital conversion unit, a coding buffer unit, a data compression unit, a data storage and calculation unit and a communication control unit. The data integration unit carries out noise reduction and filtering and A/D conversion on biological electric signal data, encodes and compresses signals returned by the virtual reality device, and finally sends the signals to the storage device after time marks are marked. The data integration unit realizes the communication control function among all the units through the communication control unit.
The medical clinical test software system is composed of an application program and a data analysis system.
The application program comprises a standard virtual reality interactive test content module and a bioelectricity stimulation signal control algorithm, and controls the delivery of interactive scene data to the clinical individual interactive integration unit and manages the interactive progress of the clinical individuals. According to the design of a medical developer on a clinical test flow, whether a biological electrical stimulation signal control algorithm is activated or not is selected as a selectable item, and the nerve activity data of a person participating in a test under the interference of electrical stimulation is recorded. The application program transmits behavioral activity data and bioelectrical signal activity data of the participating clinical individuals to the data analysis system.
The data analysis system is composed of a database system and a machine learning neural network. The database system is responsible for managing the data stored on the storage device. The machine learning neural network learns the marked data and performs prediction model training on the behavioral activity data and bioelectrical signal activity data of the clinical individuals. The training data for machine learning uses data generated by preclinical clinical individuals using the system. By learning the behavior of the clinical individual, the environmental impact factor weight is calculated, and the behavior or bioelectrical signal data of the clinical individual in a next round of several interactive tests is predicted. The actual effect of the drug in a clinical trial is obtained by the difference between the data predicted by the computer and the data generated by the individual following actual clinical administration under the application of a particular set of circumstances of influence.
In order to achieve accurate elimination of environmental interference in a clinical process and obtain high-quality actual effect data of a medicine, the invention also provides a medical clinical test auxiliary method based on the virtual reality technology, which is applied to a medical clinical test auxiliary system based on the virtual reality technology, and as shown in fig. 2, the method comprises the following steps:
step 1, establishing a test scheme before a newly developed drug enters a clinical test link
In this step, the pharmaceutical developer designs the clinical trial process according to expected data and statistical models before the clinical trial session. Including determining the number of recruited healthy/patient volunteers, the schedule of screening and/or control groups. At this stage, the system performs targeted functional adjustments as required and determines a number of virtual reality scenarios and electrical stimulation interference scenarios that are expected to be implemented. The interference scheme is mainly adjusted according to design factors of a virtual reality scene and input parameters of an electrical stimulation scheme. Input parameters include, but are not limited to: color, sound, social pressure, degree of fear, degree of anxiety, virtual task difficulty, application of pulse voltage, pulse delay, and the like. The number of the virtual reality scenes and the number of the electric stimulation schemes are flexibly adjusted according to the types and the number of clinical reaction parameters needing to be observed in the current clinical project and the requirements on the sufficiency and the accuracy of a computer neural network training model.
The specific parameter adjustment method is shown in fig. 3(a) and 3 (b). In terms of input, a virtual reality scene and an electrical stimulation interference scheme need to be designed for a bluebook by a set of parameterization schemes. As an example of parameterization, the example does not limit the extension of the parameterization scheme during implementation. For example, a set of virtual reality scenes is designed such that, in a piece of music, a clinical individual sees in virtual reality the look of the sea sky before sunrise at sea and plays a game of casserole with a shovel while chatting with a virtual character in virtual reality. In this example, the scene parameters in the input parameters of the virtual reality scene cover elements such as games, conversations, and sports, the environment parameters are outdoor, the environment theme colors are blue (sky and sea) and yellow (sand beach), the object colors are red and brown (red plastic shovel, plastic bucket, tawny sand), the object shape (cylinder or other novel interesting shape), a set of parameters for describing music (melody, beat, timbre), and difficulty of action, etc. In electrical stimulation protocols, common parameters include, but are not limited to: voltage, pulse timing, and application at a particular location on the body of a clinical individual, etc.
And 2, establishing a deep learning neural network prediction model for each clinical individual according to the individual behaviors of the clinical individuals and the bioelectrical signals.
Clinical individuals use the system, under the interference of a preset virtual reality scene and an electrical stimulation scheme, the correctly worn sensors and other auxiliary systems collect the feedback of the clinical individuals, and a data archive is established according to individual numbers. The collected data is marked as historical data and sent to a computer neural network, and weight relation calculation is carried out on the influence of input parameters on each clinical response. 80% of the historical data was used to train the predictive model, and the remaining 20% was used to test the effectiveness of the model. And (4) sending each trained prediction model into a system storage device for storage according to the number of the clinical individual.
The bioelectric signal includes, but is not limited to: electroencephalogram (EEG), Electromyogram (EMG), Electrocardiograph (ECD), etc. signals detected by an extracorporeal device, but also includes, but is not limited to, signals detected by an invasive detection device, such as acupuncture, probes, etc.
The individual behaviors include, but are not limited to: questionnaires, language interactions, medical feedback, psychological counseling, daily behavioral observations, and prescribed designed limb movements, among others.
And 3, grouping clinical individuals to implement the medicine, and obtaining clinical data through tests.
Clinical individuals are grouped according to the design scheme of clinical trials, and the grouped clinical individuals are administered with the medicine. Clinical individuals begin wearing the system of the present invention after being administered a drug, interact with the virtual reality scene and receive electrical stimulation interference as planned. The procedure is similar to step 2, and the test data is recorded and time-stamped to obtain clinical data. In particular, the time at which the drug reaches a peak blood concentration and the clinical data corresponding thereto are marked.
And 4, predicting data of the clinical individuals under the non-drug action by using the deep learning neural network prediction model, and removing the data predicted by the deep learning neural network prediction model from the clinical data to obtain drug effect data which is dominated by drug factors.
And (3) sending the virtual reality interactive content delivered to the clinical individual and the electrical stimulation applied according to the plan as input parameters to the deep learning neural network prediction model trained in the step 2. And (4) predicting the most possible behavioral response and neurobioelectric signals of each clinical individual one by one according to the serial number of the clinical individual. The predicted data is stored according to category classification and is used as clinical response data of the clinical individual under the non-drug action. Finally, the data predicted by the deep learning neural network is compared with real clinical data containing all factors, and after item-by-item difference solving, the obtained difference value is the effect generated on the body of a clinical individual by a medicament. After each clinical individual passes through the system, a list of parameter differences is left in the system database to represent the effect of the drug on the clinical individual.
As a clinical result, one of the problems often encountered is that such clinical results are confounded by the effects of the drug and the environmental impact to which the individual is subjected. However, because of the special properties of the nerve drugs acting on human organs, the responses of the nerve drugs after the human body receives the drugs are usually greatly and complexly related to the change of the environment and the past life history of the individual. Due to the factors of the living environment, the living habits, the disease degree, the interpersonal relationship, the thinking mode and the like of each person, a great amount of uncertain factors are bound in the final clinical result data. For example, a distinct use of "you" and "you" in the questionnaire would result in a clearly different amplitude of mental activity for the tested population. In the test, although the effectiveness of the new drug in the same environment is compared by the setting up of a control group, it is further desirable to statistically suppress the weight of such uncertain influence. In the final result, the individual experience on the environment variable is very large and difficult to predict in the clinical result, so that the data is often discrete and has no obvious rule. Furthermore, due to the discrete and random nature of the data, statistical models often fail to draw conclusions on the data that are desired by medical developers. The conclusion includes the safety value and adjustment of the dosage, the test of the pharmacokinetic model theory, the potential risk of the drug, the common side effect and the like. Clinical test data cannot be validated on a statistical model, drug development is trapped in a predicament on the process, the problem frequently encountered in the development of nerve drugs is solved, and the enthusiasm of drug developers for the development project investment of the drugs is greatly frustrated.
According to the invention, through quantitative learning of behavior habits of personnel in a test link, totally-enclosed accurate control of test conditions, and a tightly designed interactive flow and quantitative system, influences of environmental factors and individual factors in a clinical test process of the medicine are screened out through artificial intelligence, and finally, an organism reaction caused by the medicine is calculated.
Through the analysis, the invention adopts a method of item-by-item difference finding to obtain the effect which is generated on the body of a clinical individual by a medicament. The item-by-item difference solving method is realized by adopting the following formula:
drug effect data = clinical data-data predicted by deep learning neural network predictive models.
And 5, outputting the drug effect data with the drug factors occupying the leading position to a medical clinical statistical model, and obtaining a clinical test conclusion after statistics of the medical clinical statistical model.
In the step, the medicine effect data obtained in the step 4 is transmitted to a statistical model designed by a medicine developer for calculation, and the medicine developer obtains a final clinical test conclusion. Since the statistical model and the clinical validity conclusion of medicine relate to the specific work of the drug development, this step does not give any details about the specific calculation process of the statistical model.
In the specific clinical trial scenario of the drug of the present invention, the systematic implementation and organization of the clinical process is shown in FIG. 4. After each clinical individual has gone through the system process, an output set of data is stored in a database representing drug effect data after the individual has been excluded from other interfering factors. After all clinical volunteer population tests are completed, the data are summarized into data which can be transversely compared, and finally the data are used by a drug developer to make a drug effectiveness conclusion.
In addition, the output of the present invention also includes other data. Other data includes raw collected data and intermediate data generated during the calculation, including but not limited to: the method comprises the steps of a virtual reality scene, an electric stimulation interference scheme and a parameterized list thereof, the serial number of a clinical individual in a system and original data generated by interaction with the system, an algorithm trained by a computer neural network model, a prediction model corresponding to the serial number of the clinical individual one to one, a prediction result of clinical behaviors and bioelectricity signals and the like.
It should be emphasized that the embodiments described herein are illustrative rather than restrictive, and thus the present invention is not limited to the embodiments described in the detailed description, but also includes other embodiments that can be derived from the technical solutions of the present invention by those skilled in the art.
Claims (10)
1. A clinical trial auxiliary system of medicine based on virtual reality technique which characterized in that: the system comprises a medical clinical test hardware system and medical clinical test software, wherein the medical clinical test hardware system consists of an interactive integration unit and a data integration unit, and the medical clinical test software consists of an application program and a data analysis system;
the interaction integration unit comprises virtual reality equipment, bioelectricity signal acquisition equipment and electric signal stimulation implementation equipment: the virtual reality equipment interacts with the clinical individual and transmits interaction data to the application program, the electric signal stimulation implementation equipment implements electric signals on a specific part of a human body, and the bioelectrical signal acquisition equipment collects bioelectrical signals generated by the clinical individual and transmits the bioelectrical signals to the data integration unit;
the data integration unit comprises a data processing circuit and a data storage device, the data processing circuit processes the biological electric signal and marks time to send the biological electric signal to the data storage device, and interactive data transmitted by the virtual reality device is coded and compressed and then is marked with time to send the interactive data to the data storage device;
the application program comprises a standard virtual reality interactive test content module and a bioelectricity stimulation signal control algorithm and is used for delivering interactive scene data to the interactive integrated unit, managing the interactive progress of clinical individuals and transmitting behavior activity data and bioelectricity signal activity data of the clinical individuals to the data analysis system;
the data analysis system comprises a database system and a machine learning neural network, wherein the database system is responsible for managing data stored on the storage equipment; the machine learning neural network is used for establishing a deep learning neural network prediction model and training clinical individual behavior data and bioelectricity signal activity data to obtain data predicted by the deep learning neural network prediction model, and after item-by-item difference solving is carried out by adopting an item-by-item difference solving method, medicine effect data which is only dominated by medicine factors is obtained;
the item-by-item difference solving method is calculated by adopting the following formula:
the medicine effect data is clinical data, which is data predicted by a deep learning neural network prediction model;
the clinical data were obtained as follows: the clinical individuals are grouped and subjected to medicine tests to obtain the medicine;
the data predicted by the deep learning neural network prediction model is obtained by adopting the following method: virtual reality interactive contents delivered by clinical individuals and electric stimulation applied according to a plan are used as input parameters and are sent to the trained deep learning neural network prediction model; predicting the behavioral response and neurobioelectricity signals of each clinical individual one by one according to the serial number of the clinical individual; and storing the predicted data according to classification of categories and serving as the data of the clinical individual under the non-drug action.
2. A medical clinical trial assistance system based on virtual reality technology according to claim 1, characterized in that: the interactive integrated unit and the data integrated unit are of a split structure or an integrated structure, the interactive integrated unit of the split structure is worn by a user independently, and the data integrated unit and the interactive integrated unit are placed in a split mode.
3. A medical clinical trial assistance system based on virtual reality technology according to claim 1, characterized in that: the interaction integration unit further comprises a mechanical sensor and an optical sensor; the mechanical sensor is used for measuring limb action signals generated when a clinical individual interacts with the virtual reality equipment, the limb action signals comprise a motion direction, a motion acceleration, rotation and vibration, and the behavior and the action of the clinical individual are restored by using a general mechanical analysis method; the optical sensor is used for optically detecting the external environment of the clinical individual and photographing the clinical individual, and the facial expression and behavior action of the clinical individual generated in the process of using the virtual reality equipment are analyzed by adopting a computer vision analysis technology.
4. A medical clinical trial assisting method based on a virtual reality technology, which is applied to the medical clinical trial assisting system based on a virtual reality technology according to any one of claims 1 to 3, and is characterized in that: the method comprises the following steps:
step 1, establishing a test scheme before a newly developed drug enters a clinical test link;
step 2, establishing a deep learning neural network prediction model for each clinical individual according to the individual behaviors of the clinical individuals and the bioelectrical signals;
step 3, grouping clinical individuals to implement the medicine, and obtaining clinical data through tests;
and 4, predicting data of the clinical individuals under the non-drug action by using the deep learning neural network prediction model, and removing the data predicted by the deep learning neural network prediction model from the clinical data to obtain drug effect data which is dominated by drug factors.
5. The virtual reality technology-based medical clinical trial assistance method according to claim 4, wherein: after the step 4, the method further comprises the following steps:
and 5, outputting the drug effect data with the drug factors occupying the leading position to a medical clinical statistical model, and obtaining a clinical test conclusion after statistics of the medical clinical statistical model.
6. A medical clinical trial assistance method based on virtual reality technology according to claim 4 or 5, characterized in that: the specific implementation method of the step 1 comprises the following steps: before a newly developed medicine enters a clinical test link, a medicine developer designs a clinical test flow according to expected data and a statistical model, wherein the clinical test flow comprises a plan for determining the number of recruited healthy/patient volunteers and screening and/or control groups; determining a virtual reality scene and an electrical stimulation interference scheme which are expected to be implemented; adjusting design elements of a virtual reality scene and input parameters of an electrical stimulation scheme; the input parameters include color, sound, social pressure, level of fear, level of anxiety, difficulty of virtual tasks, applied pulse voltage, and pulse delay.
7. A medical clinical trial assistance method based on virtual reality technology according to claim 4 or 5, characterized in that: the specific implementation method of the step 2 comprises the following steps: under the interference of a pre-established virtual reality scene and an electrical stimulation scheme, collecting feedback data of clinical individuals and establishing a data archive base according to individual numbers; the collected data is used as historical data to be marked and sent to a computer neural network, and weight relation calculation is carried out on the influence of input parameters on various clinical reactions; using 80% of the historical data to train the prediction model, and using the remaining 20% to test the effectiveness of the model; each trained prediction model is sent into a system storage device for storage according to the serial number of the clinical individual;
the bioelectricity signals comprise electroencephalogram signals, electromyogram signals and electrocardiosignals; the individual behaviors include questionnaires, verbal interactions, feedback from medical care, psychological counseling, daily behavioral observations, and prescribed designed limb movements.
8. A medical clinical trial assistance method based on virtual reality technology according to claim 4 or 5, characterized in that: the specific implementation method of the step 3 is as follows: grouping clinical individuals according to a design scheme of a clinical test; administering the medication to a group of clinical individuals; the clinical individual starts a test after being administered with the drug, interacts with the virtual reality scene and receives electrical stimulation interference according to a plan; the trial data was recorded and time-stamped to obtain clinical data.
9. A medical clinical trial assistance method based on virtual reality technology according to claim 4 or 5, characterized in that: the specific implementation method of the step 4 comprises the following steps: virtual reality interactive contents delivered to clinical individuals and electric stimulation applied according to a plan are taken as input parameters and are sent to the deep learning neural network prediction model trained in the step 2; predicting the behavioral response and neurobioelectricity signals of each clinical individual one by one according to the serial number of the clinical individual; the predicted data is stored according to class classification and is used as the data of the clinical individual under the non-drug action; according to data predicted by the deep learning neural network and clinical data, the drug effect data which is solely dominated by drug factors is obtained after item-by-item difference solving is carried out by adopting an item-by-item difference solving method.
10. A medical clinical trial assistance method based on virtual reality technology according to claim 9, wherein: the item-by-item difference solving method is realized by adopting the following formula:
the drug effect data is clinical data, which is predicted by a deep learning neural network prediction model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110939785.9A CN113393946B (en) | 2021-08-17 | 2021-08-17 | Medical clinical test auxiliary system and method based on virtual reality technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110939785.9A CN113393946B (en) | 2021-08-17 | 2021-08-17 | Medical clinical test auxiliary system and method based on virtual reality technology |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113393946A CN113393946A (en) | 2021-09-14 |
| CN113393946B true CN113393946B (en) | 2021-11-09 |
Family
ID=77622888
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110939785.9A Active CN113393946B (en) | 2021-08-17 | 2021-08-17 | Medical clinical test auxiliary system and method based on virtual reality technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113393946B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115312152B (en) * | 2022-09-01 | 2023-06-20 | 北京舒曼德医药科技开发有限公司 | Medicine early warning system based on clinical test platform |
| CN116364309B (en) * | 2023-03-28 | 2024-02-02 | 绍兴珂西生物科技有限公司 | Clinical trial risk assessment method and system based on neural network |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106777861A (en) * | 2016-01-11 | 2017-05-31 | 北京视友科技有限责任公司 | A kind of cognitive and brain TV university data system |
| CN108836285A (en) * | 2018-05-09 | 2018-11-20 | 陕西数字人信息科技有限公司 | The autonomous diagnostic model of human body diseases and artificial intelligence examination method and system and health evaluating visually demonstrate method |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| AU2002952938A0 (en) * | 2002-11-27 | 2002-12-12 | Ilifedata Pty Ltd | System for Conducting Clinical Trials |
| CN103164823B (en) * | 2011-12-13 | 2017-12-29 | 天津阿贝斯努科技有限公司 | Clinical drug trial monitoring management system |
| CN106665496A (en) * | 2017-02-23 | 2017-05-17 | 连云港市东方医院 | Experimental method for medicine used for treating spinal cord injury (SCI) |
| CN111916176A (en) * | 2019-05-07 | 2020-11-10 | 西安康博瑞生命科学技术有限公司 | Double-blind medicine clinical test method and system |
-
2021
- 2021-08-17 CN CN202110939785.9A patent/CN113393946B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106777861A (en) * | 2016-01-11 | 2017-05-31 | 北京视友科技有限责任公司 | A kind of cognitive and brain TV university data system |
| CN108836285A (en) * | 2018-05-09 | 2018-11-20 | 陕西数字人信息科技有限公司 | The autonomous diagnostic model of human body diseases and artificial intelligence examination method and system and health evaluating visually demonstrate method |
Non-Patent Citations (2)
| Title |
|---|
| "基于虚拟现实技术手功能康复训练系统的设计及临床应用";肖喜玲等;《中国康复医学杂志》;20140630;第29卷(第6期);全文 * |
| "探讨药物国际多中心临床试验设计的一种新方法SGDDP";黄钦等;《中国新药杂志》;20150731;第24卷(第7期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113393946A (en) | 2021-09-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Janssen et al. | Recognition of emotions in gait patterns by means of artificial neural nets | |
| Pioggia et al. | Human-robot interaction in autism: FACE, an android-based social therapy | |
| CN111225612A (en) | Neural obstacle identification and monitoring system based on machine learning | |
| CN113393946B (en) | Medical clinical test auxiliary system and method based on virtual reality technology | |
| Becedas | Brain–machine interfaces: basis and advances | |
| CN115177208A (en) | Augmented reality system and method for user health analysis | |
| CN104287747A (en) | Exercise rehabilitation robot interactive control method based on emotion perception | |
| Kouris et al. | HOLOBALANCE: An Augmented Reality virtual trainer solution forbalance training and fall prevention | |
| Randolph | Not all created equal: individual-technology fit of brain-computer interfaces | |
| CN110111873B (en) | VR clue craving response measurement system for inducing addiction based on addiction memory-motivation incentive model | |
| Funabashi et al. | Agar: A vr serious game to support the recovery of post-stroke patients | |
| Aguilar-Lazcano et al. | Interaction modalities used on serious games for upper limb rehabilitation: a systematic review | |
| CN115376695A (en) | Method, system and device for neuropsychological assessment and intervention based on augmented reality | |
| CN115691804A (en) | Social cognition assessment training system based on digital twin-neuron universe | |
| Górriz et al. | Computational approaches to Explainable Artificial Intelligence: Advances in theory, applications and trends | |
| Leonard et al. | Cortical encoding of manual articulatory and linguistic features in American Sign Language | |
| CN109126045A (en) | intelligent motion analysis and training system | |
| Han | A virtual reality algorithm for the study of clinical efficacy of sports injury rehabilitation training | |
| Frasson et al. | Zoo therapy for Alzheimer’s disease with real-time speech instruction and neurofeedback system | |
| Stănică et al. | Parkinson’s Disease–How Virtual Reality Exercises can Improve the Neurorehabilitation Process | |
| Georgoulis et al. | Epione: An innovative pain management system using facial expression analysis, biofeedback and augmented reality-based distraction | |
| Serino et al. | Immersive virtual reality for assessment and rehabilitation of deficits of body representations and cognitive functions | |
| Pioggia et al. | Facial Automaton for Conveying Emotions as a Social Rehabilitation Tool for People with Autism | |
| Manson | Examining the sensorimotor integration processes prior to and during movements to somatosensory targets | |
| Torres | Inherent Noise Hidden in Nervous Systems’ Rhythms Leads to New Strategies for Detection and Treatments of Core Motor Sensing Traits in ASD |
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 | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20231106 Address after: Rehabilitation Industry Incubator 301-31, Block D, No. 100 Torch Road, High tech Zone, Qingdao, Shandong Province, 266000 Patentee after: Yinuoke Rehabilitation Medical Technology (Qingdao) Co.,Ltd. Address before: 300480 room 209, floor 2, area C, animation building, No. 126, animation Middle Road, eco city, Binhai New Area, Tianjin (No. 1903 of Tianjin Haobang business secretary Co., Ltd.) Patentee before: Aoluo Technology (Tianjin) Co.,Ltd. |