CN116721777A - Neural network-based drug efficacy evaluation method, device, equipment and medium - Google Patents
Neural network-based drug efficacy evaluation method, device, equipment and medium Download PDFInfo
- Publication number
- CN116721777A CN116721777A CN202310999925.0A CN202310999925A CN116721777A CN 116721777 A CN116721777 A CN 116721777A CN 202310999925 A CN202310999925 A CN 202310999925A CN 116721777 A CN116721777 A CN 116721777A
- Authority
- CN
- China
- Prior art keywords
- physiological data
- body surface
- data
- module
- neural network
- 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.)
- Granted
Links
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 80
- 239000003814 drug Substances 0.000 title claims abstract description 74
- 229940079593 drug Drugs 0.000 title claims abstract description 57
- 238000011156 evaluation Methods 0.000 title claims description 24
- 241000282414 Homo sapiens Species 0.000 claims abstract description 30
- 238000000034 method Methods 0.000 claims abstract description 27
- 238000010171 animal model Methods 0.000 claims abstract description 25
- 229940000406 drug candidate Drugs 0.000 claims abstract description 25
- 238000005259 measurement Methods 0.000 claims abstract description 6
- 239000013598 vector Substances 0.000 claims description 40
- 230000006870 function Effects 0.000 claims description 31
- 230000004927 fusion Effects 0.000 claims description 21
- 238000012549 training Methods 0.000 claims description 16
- 241000283984 Rodentia Species 0.000 claims description 15
- 210000004556 brain Anatomy 0.000 claims description 7
- 238000004590 computer program Methods 0.000 claims description 7
- 206010041235 Snoring Diseases 0.000 claims description 5
- 230000003187 abdominal effect Effects 0.000 claims description 5
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 claims description 5
- 239000008280 blood Substances 0.000 claims description 5
- 210000004369 blood Anatomy 0.000 claims description 5
- 230000003183 myoelectrical effect Effects 0.000 claims description 5
- 229910052760 oxygen Inorganic materials 0.000 claims description 5
- 239000001301 oxygen Substances 0.000 claims description 5
- 230000000241 respiratory effect Effects 0.000 claims description 5
- 238000003780 insertion Methods 0.000 claims description 3
- 230000037431 insertion Effects 0.000 claims description 3
- 238000012854 evaluation process Methods 0.000 abstract description 6
- 230000009471 action Effects 0.000 description 10
- 208000013738 Sleep Initiation and Maintenance disease Diseases 0.000 description 7
- 230000008667 sleep stage Effects 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 238000002474 experimental method Methods 0.000 description 5
- 241000282412 Homo Species 0.000 description 4
- 238000003745 diagnosis Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 230000007246 mechanism Effects 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 238000011160 research Methods 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 230000000857 drug effect Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000002372 labelling Methods 0.000 description 3
- 238000013508 migration Methods 0.000 description 3
- 230000005012 migration Effects 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 238000003759 clinical diagnosis Methods 0.000 description 2
- 201000010099 disease Diseases 0.000 description 2
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 2
- 230000003285 pharmacodynamic effect Effects 0.000 description 2
- 230000000144 pharmacologic effect Effects 0.000 description 2
- 208000019116 sleep disease Diseases 0.000 description 2
- 230000003860 sleep quality Effects 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 206010067484 Adverse reaction Diseases 0.000 description 1
- 206010007269 Carcinogenicity Diseases 0.000 description 1
- 241000282693 Cercopithecidae Species 0.000 description 1
- 241001465754 Metazoa Species 0.000 description 1
- 241000283973 Oryctolagus cuniculus Species 0.000 description 1
- 102000005877 Peptide Initiation Factors Human genes 0.000 description 1
- 108010044843 Peptide Initiation Factors Proteins 0.000 description 1
- 241000009328 Perro Species 0.000 description 1
- 241000282898 Sus scrofa Species 0.000 description 1
- 230000007059 acute toxicity Effects 0.000 description 1
- 231100000403 acute toxicity Toxicity 0.000 description 1
- 230000006838 adverse reaction Effects 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000007670 carcinogenicity Effects 0.000 description 1
- 231100000260 carcinogenicity Toxicity 0.000 description 1
- 230000007665 chronic toxicity Effects 0.000 description 1
- 231100000160 chronic toxicity Toxicity 0.000 description 1
- 150000001875 compounds Chemical class 0.000 description 1
- 239000000470 constituent Substances 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000002565 electrocardiography Methods 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000000977 initiatory effect Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 230000037353 metabolic pathway Effects 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 230000001575 pathological effect Effects 0.000 description 1
- 230000001766 physiological effect Effects 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 231100000205 reproductive and developmental toxicity Toxicity 0.000 description 1
- 238000012827 research and development Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 230000001225 therapeutic effect Effects 0.000 description 1
- 231100000331 toxic Toxicity 0.000 description 1
- 230000002588 toxic effect Effects 0.000 description 1
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
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/70—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/253—Fusion techniques of extracted features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/047—Probabilistic or stochastic networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/048—Activation functions
-
- 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
- G16H10/00—ICT specially adapted for the handling or processing of patient-related medical or healthcare data
- G16H10/20—ICT specially adapted for the handling or processing of patient-related medical or healthcare data for electronic clinical trials or questionnaires
-
- 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
- G16H10/00—ICT specially adapted for the handling or processing of patient-related medical or healthcare data
- G16H10/60—ICT specially adapted for the handling or processing of patient-related medical or healthcare data for patient-specific data, e.g. for electronic patient records
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2123/00—Data types
- G06F2123/02—Data types in the time domain, e.g. time-series data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/08—Feature extraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/12—Classification; Matching
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Abstract
The invention relates to the technical field of neural networks, in particular to a method, a device, equipment and a medium for evaluating drug efficacy based on a neural network. The method comprises the following steps: acquiring first body surface physiological data generated by an experimental animal taking a candidate drug and second body surface physiological data generated by an experimental human not taking the candidate drug; respectively inputting the first body table physiological data and the second body table physiological data into a preset first neural network and a second neural network, and sequentially outputting to obtain a first output characteristic and a second output characteristic; inputting the first output characteristic and the second output characteristic into a preset loss function based on probability distribution measurement to obtain a loss value; updating network parameters of the second neural network based on the loss value; wherein the updated second neural network is used to clinically evaluate the drug candidate. The technical scheme can reduce the clinical evaluation process of candidate medicines.
Description
Technical Field
The invention relates to the technical field of neural networks, in particular to a method, a device, equipment and a medium for evaluating drug efficacy based on a neural network.
Background
The pharmacodynamics for short, which is the study of the biochemical and physiological effects and the mechanism of the medicine and the relation between the dosage and the effect, aims to determine the expected curative effect of the new medicine for clinical prevention, diagnosis and treatment, determine the action intensity of the new medicine, clarify the action part and mechanism of the new medicine and find the expected wide pharmacological action beyond the clinic. Pharmacodynamics includes basic types of drug actions, selectivity of drug actions, dose-effect relation of drug actions, therapeutic actions and adverse reactions of drugs and action mechanisms of drugs.
Before the candidate medicine enters the clinical test stage in the research and development process of the new medicine, the drug efficacy indexes such as acute, chronic, developmental and reproductive toxicity, carcinogenicity and the like of the candidate medicine must be comprehensively evaluated so as to ensure the effectiveness and the safety of the medicine after the medicine is applied to a human body. Since experimental animals generally possess similar molecular targets and metabolic pathways as humans, they are used for preclinical studies of drug candidates. The pharmacological action of the medicine is evaluated by different animal models and experimental methods, the action mechanism is researched, and the toxic action is observed.
In the related art, preclinical evaluation of candidate drugs usually uses a rodent (e.g., rat or mouse) and a non-rodent (e.g., rabbit, dog, pig or monkey) to sequentially perform drug experiments, wherein the test results of the non-rodent animals not only can significantly improve the success rate of clinical experiments of candidate drugs, but also can provide higher security guarantee for the drugs after clinical use. Even so, clinical evaluation of drug candidates still requires a large number of volunteers or patients to conduct drug experiments, which can result in excessively lengthy clinical evaluation of drug candidates.
Therefore, there is a need to provide a method, a device and a medium for evaluating drug efficacy based on neural network to solve the above technical problems.
Disclosure of Invention
The invention describes a drug efficacy evaluation method, device, equipment and medium based on a neural network, which can reduce the clinical evaluation process of candidate drugs.
According to a first aspect, the present invention provides a method for evaluating drug efficacy of a drug based on a neural network, comprising:
acquiring first body surface physiological data generated by an experimental animal taking a candidate drug and second body surface physiological data generated by an experimental human not taking the candidate drug; the experimental animal comprises rodents and non-rodents, the first body surface physiological data and the second body surface physiological data are the same in data type, and the first body surface physiological data and the second body surface physiological data are time sequence data;
respectively inputting the first body table physiological data and the second body table physiological data into a preset first neural network and a second neural network, and sequentially outputting to obtain a first output characteristic and a second output characteristic;
inputting the first output characteristic and the second output characteristic into a preset loss function based on probability distribution measurement to obtain a loss value;
updating network parameters of the second neural network based on the loss value; wherein the updated second neural network is used to clinically evaluate the drug candidate.
According to one embodiment, the body surface physiological data includes at least one of: respiratory pressure data, brain electrical data, eye electrical data, myoelectrical data, electrocardiographic data, chest strap data, abdominal strap data, pulse wave data, leg movement data, snore data, pulse rate data, and blood oxygen saturation data.
According to one embodiment, the probability distribution metric-based loss function comprises at least one of: KL divergence loss function, cross entropy loss function, softmax loss function, and Focal loss function.
According to one embodiment, the first neural network and the second neural network each comprise a convolution module, an intra-modality attention module, and a linear module connected in sequence;
in a model training stage, a feature fusion module is connected between the intra-mode attention module and the linear module, and an inter-mode attention module is connected between the intra-mode attention module and the feature fusion module;
the convolution module is used for extracting time domain features and frequency domain features of the body surface physiological data, the intra-mode attention module is used for adjusting weights of features of the region of interest in the body surface physiological data of the insertion position code, the inter-mode attention module is used for adjusting weights of features of the region of interest in the first body surface physiological data and the second body surface physiological data, the feature fusion module is used for carrying out feature fusion on features of the region of interest in the first body surface physiological data and the second body surface physiological data and features of the region of interest output by the inter-mode attention module respectively, and the linear module is used for adjusting dimensions of the features output by the two feature fusion modules to be the same dimension.
According to one embodiment, in the model training phase, the convolution module is further connected with a first vector module, and the intra-modality attention module is further connected with a second vector module;
the first vector module is used for providing a one-dimensional first vector so as to fuse the first vector with the characteristics output by the convolution module; the area where the first vector is located is the area where the attention module in the mode is interested;
the second vector module is used for providing a one-dimensional second vector so as to enable the second empty vector to be fused with the characteristics of the region of interest output by the intra-mode attention module.
According to one embodiment, the linear module is a fully connected layer or a multi-layer perceptron.
According to one embodiment, when the linear module is a multi-layer perceptron, the updated second neural network specifically clinically evaluates the drug candidate by:
acquiring real body surface physiological data generated by an experimental human taking a candidate drug and body surface physiological data to be evaluated generated by an experimental human not taking the candidate drug;
inputting the body surface physiological data to be evaluated into the updated second neural network, and outputting to obtain target body surface physiological data;
and adjusting network parameters of the multi-layer perceptron based on errors of the real body surface physiological data and the target body surface physiological data until the errors are smaller than preset errors.
According to a second aspect, the present invention provides a neural network-based drug efficacy evaluation device, comprising:
an acquisition unit for acquiring first body surface physiological data generated by an experimental animal taking a candidate drug and second body surface physiological data generated by an experimental human not taking the candidate drug; the experimental animal comprises rodents and non-rodents, the first body surface physiological data and the second body surface physiological data are the same in data type, and the first body surface physiological data and the second body surface physiological data are time sequence data;
the first input unit is used for inputting the first body form physiological data and the second body form physiological data into a preset first neural network and a second neural network respectively, and sequentially outputting the first body form physiological data and the second body form physiological data to obtain a first output characteristic and a second output characteristic;
the second input unit is used for inputting the first output characteristic and the second output characteristic into a preset loss function based on probability distribution measurement to obtain a loss value;
an updating unit, configured to update a network parameter of the second neural network based on the loss value; wherein the updated second neural network is used to clinically evaluate the drug candidate.
According to a third aspect, the present invention provides an electronic device comprising a memory and a processor, the memory having stored therein a computer program, the processor implementing the method of the first aspect when executing the computer program.
According to a fourth aspect, the present invention provides a computer readable storage medium having stored thereon a computer program which, when executed in a computer, causes the computer to perform the method of the first aspect.
According to the drug efficacy evaluation method, device, equipment and medium based on the neural network, cross-domain (namely animal-human) data migration is realized by means of the neural network, so that after training of the neural network is completed, the characteristics of experimental animals containing the drug taking candidate can be output by inputting the body surface physiological data of experimental humans not taking the drug candidate, and the accuracy of a trained model can be verified by using the body surface physiological data of experimental humans taking the drug candidate with a small sample size, so that the problem of long clinical evaluation process of the drug candidate in the related technology can be solved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions of the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described, and it is obvious that the drawings described below are some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 illustrates a flow diagram of a neural network-based drug efficacy evaluation method, according to one embodiment;
FIG. 2 shows a schematic block diagram of a neural network-based drug efficacy evaluation device, according to one embodiment;
FIG. 3 illustrates a network architecture diagram of a model training process, according to one embodiment.
Detailed Description
As described above, in the related art, clinical evaluation of a candidate drug requires a large number of volunteers or patients to conduct a drug experiment, which may result in excessively long clinical evaluation of the candidate drug.
Neural networks, also known as Artificial Neural Networks (ANNs) or Simulated Neural Networks (SNNs), are a subset of machine learning and are also the core of deep learning algorithms. The name and structure are inspired by the brain of a person, and can imitate the mutual transmission mode of biological neurons. Neural networks rely on training data to learn and improve their accuracy over time. Once the learning algorithms are optimized, the accuracy is improved, and the learning algorithms become a powerful tool in the fields of computer science and artificial intelligence, so that the data can be classified and clustered rapidly.
The inventor solves the problem of long clinical evaluation process of candidate drugs in the related technology, creatively discovers that: the cross-domain (i.e. animal-human) data migration can be realized by means of the neural network, so that after the neural network is trained, the characteristics of the experimental animal containing the drug candidate can be obtained by inputting the body surface physiological data of the experimental human not taking the drug candidate, and therefore, the accuracy of the trained model can be verified by using the body surface physiological data of the experimental human taking the drug candidate with a small sample size, and the problem of long clinical evaluation process of the drug candidate in the related technology can be solved.
The scheme provided by the invention is described below with reference to the accompanying drawings.
Fig. 1 shows a flow diagram of a neural network-based drug efficacy evaluation method according to one embodiment. It is understood that the method may be performed by any apparatus, device, platform, cluster of devices having computing, processing capabilities. As shown in fig. 1, the method includes:
step 100, acquiring first body surface physiological data generated by experimental animals taking candidate drugs and second body surface physiological data generated by experimental humans not taking the candidate drugs; the experimental animals comprise rodents and non-rodents, the data types of the first body surface physiological data and the second body surface physiological data are the same, and the first body surface physiological data and the second body surface physiological data are both time sequence data;
102, respectively inputting first body table physiological data and second body table physiological data into a preset first neural network and a second neural network, and sequentially outputting to obtain a first output characteristic and a second output characteristic;
104, inputting the first output characteristic and the second output characteristic into a preset loss function based on probability distribution measurement to obtain a loss value;
step 106, updating network parameters of the second neural network based on the loss value; wherein the updated second neural network is used to clinically evaluate the drug candidate.
In this embodiment, the cross-domain (i.e., animal-human) data migration is realized by using the neural network, so that after the training of the neural network is completed, the characteristics of the experimental animal including the drug candidate can be obtained by inputting the body surface physiological data of the experimental human not taking the drug candidate, and therefore, the accuracy of the trained model can be verified by using the body surface physiological data of the experimental human taking the drug candidate with a small sample size, and the problem of the long clinical evaluation process of the drug candidate in the related art can be solved.
In one embodiment of the invention, the probability distribution metric-based loss function includes at least one of: KL divergence loss function, cross entropy loss function, softmax loss function, and Focal loss function.
In this embodiment, the probability distribution metric-based loss function is used to convert the similarity between samples into the probability of occurrence of random events, i.e. by measuring the distance between the actual distribution of the samples and the distribution estimated by the actual distribution of the samples, and judging the similarity between the actual distribution of the samples and the distribution estimated by the actual distribution of the samples, and is generally used in application problems related to probability distribution or probability of occurrence of prediction categories, which is particularly common in classification problems.
In the field of data mining, time sequence classification task is an important research direction, and diagnosis and prediction of diseases can be facilitated by analyzing and mining time sequence physiological data, so that development of intelligent medical treatment is promoted.
In one embodiment of the invention, the body surface physiological data includes at least one of: respiratory pressure data, brain electrical data, eye electrical data, myoelectrical data, electrocardiographic data, chest strap data, abdominal strap data, pulse wave data, leg movement data, snore data, pulse rate data, and blood oxygen saturation data.
The first body table physiological data and the second body table physiological data have the same data type, for example, the respiratory pressure data, the brain electrical data, the eye electrical data, the myoelectrical data, the electrocardiographic data, the chest belt data, the abdominal belt data, the pulse wave data, the leg movement data, the snore data, the pulse rate data or the blood oxygen saturation data are all the same, and the specific data types of the two are not limited.
The following is presented in terms of sleep stage or sleep data (belonging to a subset of body surface physiological data).
Sleep staging is taken as a typical physiological time sequence classification task, is a basic research in the field of sleep monitoring, and is more and more widely focused by people. Sleep staging is an important means of assessing sleep quality and sleep disorders, and sleep professionals often determine sleep stages through Polysomnography (PSG), which consists of electroencephalogram (EEG), electrooculogram (EOG), electromyogram (EMG), and Electrocardiography (ECG), which can be used to diagnose sleep disorders and other common diseases. Among other things, EEG can record not only large PSG activity changes, but also drug effects during different sleep stages and awake states. In addition, sleep can be staged according to PSG waves, the control of a sleep phase is more accurate, and a PSG chart is a more objective, accurate, rapid and widely applied drug effect evaluation method in sleep drug effect research. On the other hand, the sleep PSG explores the distribution and activity of the PSG in various frequency bands and the regularity, detail change and predictability of a nonlinear system in the brain in a linear analysis and nonlinear analysis mode. The PSG analysis method is used as a research parameter for evaluating sleep quality and PSG rhythmicity change, and has certain universality, international acceptance and larger document support and reference values.
The model forming method of the insomnia disorder of the experimental animal is divided into four types by different initiation factors, namely physical factor model forming, chemical factor model forming, pathological factor model forming and compound factor model forming. The method is characterized in that a large-scale cross-domain comparison learning sample pair database containing abundant insomnia disorder priori knowledge under the initiation of each factor is constructed through experimental animal modeling, and abundant data resources are accumulated and distributed for the subsequent sleep medicine efficacy evaluation.
The existing insomnia disorder diagnosis and sleep stage model is seriously dependent on the data tag, animal sleep tag and human tag types are inconsistent, self-supervision learning can reduce the problem of realizing cross-domain self-supervision priori knowledge generalization of tag dependence, and also avoid a large number of manual labeling cost and quality consistency of data labeling. Classifying polysomnography throughout the night often relies on trained sleep professionals, which is time consuming, labor intensive, and costly. Meanwhile, in the labeling process, label classification results given by different sleep specialists are often subjective and have larger difference, so that the credibility of insomnia disorder diagnosis and sleep stage results is reduced. More importantly, the classification standard of the sleep stage of the experimental animal is not completely consistent with that of a human, so that a label based on limited quality is abandoned as supervision information, and a similarity and dissimilarity measure is adopted as supervision information, which is more beneficial to generalizing the prior knowledge of the insomnia disorder of the experimental animal to human clinical diagnosis (namely, the physiological data of a first body form and the physiological data of a second body form are respectively input into a first neural network and a second neural network which are preset and sequentially output to obtain a first output characteristic and a second output characteristic, the first output characteristic and the second output characteristic are input into a loss function which is preset and is based on probability distribution measure to obtain a loss value, the network parameters of the second neural network are updated based on the loss value), wherein the updated second neural network is used for clinically evaluating candidate medicines.
The model training phase is described below in connection with fig. 3.
As shown in fig. 3, in one embodiment of the present invention, the first neural network and the second neural network each include a convolution module, an intra-modality attention module, and a linear module connected in sequence;
in the model training stage, a feature fusion module is connected between the intra-mode attention module and the linear module, and an inter-mode attention module is connected between the intra-mode attention module and the feature fusion module;
the convolution module is used for extracting time domain features and frequency domain features of body surface physiological data, the intra-mode attention module is used for adjusting weights of features of the region of interest in the body surface physiological data of the inserted position codes, the inter-mode attention module is used for adjusting weights of features of the region of interest in the first body surface physiological data and the second body surface physiological data, the feature fusion module is used for carrying out feature fusion on features of the region of interest in the first body surface physiological data and the second body surface physiological data and features of the region of interest output by the inter-mode attention module respectively, and the linear module is used for adjusting dimensions of the features output by the two feature fusion modules to be the same dimension.
In the embodiment, the convolution module is arranged to extract the time domain features and the frequency domain features of the body surface physiological data, so that more feature data in the body surface physiological data can be obtained, and more effective information in the data pair of the body surface physiological data can be effectively extracted; the intra-mode attention module is arranged to extract part of the aggregation characteristics with discriminant; the robustness of the data pair of the body surface physiological data can be enhanced by setting the inter-mode attention module; by setting the feature fusion module, the feature of the experimental animal taking the candidate medicine can be included in the feature of the experimental human not taking the candidate medicine and the feature of the experimental animal not taking the candidate medicine can be included in the feature of the experimental human not taking the candidate medicine.
Taking fig. 3 as an example, in the embodiment of the present invention, the region of interest may be the first column of feature vectors, but may also be feature vectors of other positions, such as the nth row or the nth column, which is not specifically limited herein.
It should be noted that, in the embodiment of the present invention, the experimental human may be a volunteer (i.e. normal human) or a patient (i.e. diseased human), and the type of the experimental human is not limited herein.
With continued reference to fig. 3, in one embodiment of the present invention, during the model training phase, the convolution module is further connected to a first vector module, and the intra-modality attention module is further connected to a second vector module;
the first vector module is used for providing a one-dimensional first vector so as to enable the first vector to be fused with the characteristics output by the convolution module; the area where the first vector is located is the area of interest of the intra-mode attention module;
the second vector module is used for providing a one-dimensional second vector so as to enable the second null vector to be fused with the characteristics of the region of interest output by the intra-mode attention module.
In this embodiment, by setting the first vector module and the second vector module, accuracy of model training may be further improved.
In one embodiment of the invention, the linear module is a fully connected layer or a multi-layer perceptron.
In one embodiment of the present invention, when the linear module is a multi-layer perceptron, the updated second neural network specifically performs clinical evaluation of drug candidates by:
acquiring real body surface physiological data generated by experimental human taking the candidate medicine and body surface physiological data to be evaluated generated by experimental human not taking the candidate medicine;
inputting the body surface physiological data to be evaluated into a second neural network which is updated, and outputting to obtain target body surface physiological data;
and adjusting network parameters of the multi-layer perceptron based on errors of the real body surface physiological data and the target body surface physiological data until the errors are smaller than preset errors.
In this embodiment, when the linear module is a full-connection layer, the updated second neural network performs clinical evaluation on the candidate drug, compared with an evaluation mode when the linear module is a multi-layer perceptron, the neural network obtained by training the former is a final model, and fine adjustment of network parameters cannot be performed; the neural network trained by the latter can use errors (see below) to fine tune the network parameters of the multi-layer perceptron, so that the accuracy of the obtained final model is higher.
In some embodiments, the error may be a sum-of-variance, a mean-square-error, or a standard deviation, which is not specifically limited herein.
Taking the sleeping field as an example, the model training stage is favorable for generalizing the priori knowledge of the insomnia disorder of the experimental animal to the clinical diagnosis of human beings, and the obtained updated second neural network has the capability of identifying various insomnia disorders of the human beings.
The foregoing describes certain embodiments of the present invention. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims can be performed in a different order than the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
According to another embodiment, the invention provides a drug efficacy evaluation device based on a neural network. Fig. 2 shows a schematic block diagram of a neural network-based drug efficacy evaluation device, according to one embodiment. It will be appreciated that the apparatus may be implemented by any means, device, platform or cluster of devices having computing, processing capabilities. As shown in fig. 2, the apparatus includes: an acquisition unit 200, a first input unit 202, a second input unit 204, and an update unit 206. Wherein the main functions of each constituent unit are as follows:
an acquisition unit 200 for acquiring first body surface physiological data generated by an experimental animal taking a candidate drug and second body surface physiological data generated by an experimental human not taking the candidate drug; the experimental animal comprises rodents and non-rodents, the first body surface physiological data and the second body surface physiological data are the same in data type, and the first body surface physiological data and the second body surface physiological data are time sequence data;
a first input unit 202, configured to input the first body table physiological data and the second body table physiological data to a preset first neural network and second neural network, and sequentially output the first body table physiological data and the second body table physiological data to obtain a first output feature and a second output feature;
a second input unit 204, configured to input the first output feature and the second output feature to a preset loss function based on a probability distribution metric, to obtain a loss value;
an updating unit 206, configured to update the network parameter of the second neural network based on the loss value; wherein the updated second neural network is used to clinically evaluate the drug candidate.
In one embodiment of the invention, the body surface physiological data includes at least one of: respiratory pressure data, brain electrical data, eye electrical data, myoelectrical data, electrocardiographic data, chest strap data, abdominal strap data, pulse wave data, leg movement data, snore data, pulse rate data, and blood oxygen saturation data.
In one embodiment of the present invention, the probability distribution metric-based loss function includes at least one of: KL divergence loss function, cross entropy loss function, softmax loss function, and Focal loss function.
In one embodiment of the present invention, the first neural network and the second neural network each include a convolution module, an intra-modality attention module, and a linear module connected in sequence;
in a model training stage, a feature fusion module is connected between the intra-mode attention module and the linear module, and an inter-mode attention module is connected between the intra-mode attention module and the feature fusion module;
the convolution module is used for extracting time domain features and frequency domain features of the body surface physiological data, the intra-mode attention module is used for adjusting weights of features of the region of interest in the body surface physiological data of the insertion position code, the inter-mode attention module is used for adjusting weights of features of the region of interest in the first body surface physiological data and the second body surface physiological data, the feature fusion module is used for carrying out feature fusion on features of the region of interest in the first body surface physiological data and the second body surface physiological data and features of the region of interest output by the inter-mode attention module respectively, and the linear module is used for adjusting dimensions of the features output by the two feature fusion modules to be the same dimension.
In one embodiment of the invention, in the model training stage, the convolution module is further connected with a first vector module, and the intra-mode attention module is further connected with a second vector module;
the first vector module is used for providing a one-dimensional first vector so as to fuse the first vector with the characteristics output by the convolution module; the area where the first vector is located is the area where the attention module in the mode is interested;
the second vector module is used for providing a one-dimensional second vector so as to enable the second empty vector to be fused with the characteristics of the region of interest output by the intra-mode attention module.
In one embodiment of the invention, the linear module is a fully connected layer or a multi-layer perceptron.
In one embodiment of the present invention, when the linear module is a multi-layer perceptron, the updated second neural network specifically performs clinical evaluation on the drug candidate by:
acquiring real body surface physiological data generated by an experimental human taking a candidate drug and body surface physiological data to be evaluated generated by an experimental human not taking the candidate drug;
inputting the body surface physiological data to be evaluated into the updated second neural network, and outputting to obtain target body surface physiological data;
and adjusting network parameters of the multi-layer perceptron based on errors of the real body surface physiological data and the target body surface physiological data until the errors are smaller than preset errors.
According to an embodiment of another aspect, there is also provided a computer-readable storage medium having stored thereon a computer program which, when executed in a computer, causes the computer to perform the method described in connection with fig. 1.
According to an embodiment of yet another aspect, there is also provided an electronic device including a memory having executable code stored therein and a processor that, when executing the executable code, implements the method described in connection with fig. 1.
The embodiments of the present invention are described in a progressive manner, and the same and similar parts of the embodiments are all referred to each other, and each embodiment is mainly described in the differences from the other embodiments. In particular, for the device embodiments, since they are substantially similar to the method embodiments, the description is relatively simple, and reference is made to the description of the method embodiments in part.
Those skilled in the art will appreciate that in one or more of the examples described above, the functions described in the present invention may be implemented in hardware, software, firmware, or any combination thereof. When implemented in software, these functions may be stored on or transmitted over as one or more instructions or code on a computer-readable medium.
The foregoing embodiments have been provided for the purpose of illustrating the general principles of the present invention in further detail, and are not to be construed as limiting the scope of the invention, but are merely intended to cover any modifications, equivalents, improvements, etc. based on the teachings of the invention.
Claims (10)
1. The drug efficacy evaluation method based on the neural network is characterized by comprising the following steps of:
acquiring first body surface physiological data generated by an experimental animal taking a candidate drug and second body surface physiological data generated by an experimental human not taking the candidate drug; the experimental animal comprises rodents and non-rodents, the first body surface physiological data and the second body surface physiological data are the same in data type, and the first body surface physiological data and the second body surface physiological data are time sequence data;
respectively inputting the first body table physiological data and the second body table physiological data into a preset first neural network and a second neural network, and sequentially outputting to obtain a first output characteristic and a second output characteristic;
inputting the first output characteristic and the second output characteristic into a preset loss function based on probability distribution measurement to obtain a loss value;
updating network parameters of the second neural network based on the loss value; wherein the updated second neural network is used to clinically evaluate the drug candidate.
2. The method of claim 1, wherein the body surface physiological data comprises at least one of: respiratory pressure data, brain electrical data, eye electrical data, myoelectrical data, electrocardiographic data, chest strap data, abdominal strap data, pulse wave data, leg movement data, snore data, pulse rate data, and blood oxygen saturation data.
3. The method of claim 1, wherein the probability distribution metric-based loss function comprises at least one of: KL divergence loss function, cross entropy loss function, softmax loss function, and Focal loss function.
4. A method according to any one of claims 1-3, wherein the first and second neural networks each comprise a convolution module, an intra-modality attention module, and a linearity module connected in sequence;
in a model training stage, a feature fusion module is connected between the intra-mode attention module and the linear module, and an inter-mode attention module is connected between the intra-mode attention module and the feature fusion module;
the convolution module is used for extracting time domain features and frequency domain features of the body surface physiological data, the intra-mode attention module is used for adjusting weights of features of the region of interest in the body surface physiological data of the insertion position code, the inter-mode attention module is used for adjusting weights of features of the region of interest in the first body surface physiological data and the second body surface physiological data, the feature fusion module is used for carrying out feature fusion on features of the region of interest in the first body surface physiological data and the second body surface physiological data and features of the region of interest output by the inter-mode attention module respectively, and the linear module is used for adjusting dimensions of the features output by the two feature fusion modules to be the same dimension.
5. The method of claim 4, wherein in a model training phase, the convolution module is further coupled to a first vector module, and wherein the intra-modality attention module is further coupled to a second vector module;
the first vector module is used for providing a one-dimensional first vector so as to fuse the first vector with the characteristics output by the convolution module; the area where the first vector is located is the area where the attention module in the mode is interested;
the second vector module is used for providing a one-dimensional second vector so as to enable the second empty vector to be fused with the characteristics of the region of interest output by the intra-mode attention module.
6. The method of claim 4, wherein the linear module is a fully connected layer or a multi-layer perceptron.
7. The method according to claim 6, wherein when the linear module is a multi-layer perceptron, the updated second neural network is specifically a clinical evaluation of the drug candidate by:
acquiring real body surface physiological data generated by an experimental human taking a candidate drug and body surface physiological data to be evaluated generated by an experimental human not taking the candidate drug;
inputting the body surface physiological data to be evaluated into the updated second neural network, and outputting to obtain target body surface physiological data;
and adjusting network parameters of the multi-layer perceptron based on errors of the real body surface physiological data and the target body surface physiological data until the errors are smaller than preset errors.
8. A neural network-based drug efficacy evaluation device, comprising:
an acquisition unit for acquiring first body surface physiological data generated by an experimental animal taking a candidate drug and second body surface physiological data generated by an experimental human not taking the candidate drug; the experimental animal comprises rodents and non-rodents, the first body surface physiological data and the second body surface physiological data are the same in data type, and the first body surface physiological data and the second body surface physiological data are time sequence data;
the first input unit is used for inputting the first body form physiological data and the second body form physiological data into a preset first neural network and a second neural network respectively, and sequentially outputting the first body form physiological data and the second body form physiological data to obtain a first output characteristic and a second output characteristic;
the second input unit is used for inputting the first output characteristic and the second output characteristic into a preset loss function based on probability distribution measurement to obtain a loss value;
an updating unit, configured to update a network parameter of the second neural network based on the loss value; wherein the updated second neural network is used to clinically evaluate the drug candidate.
9. An electronic device comprising a memory and a processor, the memory having stored therein a computer program, the processor implementing the method of any of claims 1-7 when the computer program is executed.
10. A computer readable storage medium, having stored thereon a computer program which, when executed in a computer, causes the computer to perform the method of any of claims 1-7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310999925.0A CN116721777B (en) | 2023-08-10 | 2023-08-10 | Neural network-based drug efficacy evaluation method, device, equipment and medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310999925.0A CN116721777B (en) | 2023-08-10 | 2023-08-10 | Neural network-based drug efficacy evaluation method, device, equipment and medium |
Publications (2)
Publication Number | Publication Date |
---|---|
CN116721777A true CN116721777A (en) | 2023-09-08 |
CN116721777B CN116721777B (en) | 2023-10-27 |
Family
ID=87866495
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310999925.0A Active CN116721777B (en) | 2023-08-10 | 2023-08-10 | Neural network-based drug efficacy evaluation method, device, equipment and medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116721777B (en) |
Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20160217267A1 (en) * | 2015-01-20 | 2016-07-28 | Northwestern University | Systems and methods to derive models to evaluate behavior outcomes based on brain responses to complex sounds |
US20200168310A1 (en) * | 2017-05-31 | 2020-05-28 | Jerker Westin | Systems for evaluating dosage parameters |
CN111432720A (en) * | 2017-10-06 | 2020-07-17 | 梅约医学教育与研究基金会 | ECG-based cardiac ejection fraction screening |
US11263749B1 (en) * | 2021-06-04 | 2022-03-01 | In-Med Prognostics Inc. | Predictive prognosis based on multimodal analysis |
US20220165359A1 (en) * | 2020-11-23 | 2022-05-26 | Peptilogics, Inc. | Generating anti-infective design spaces for selecting drug candidates |
CN114830138A (en) * | 2019-10-31 | 2022-07-29 | 伟摩有限责任公司 | Training trajectory scoring neural networks to accurately assign scores |
CN115298534A (en) * | 2019-11-17 | 2022-11-04 | 伯克利之光生命科技公司 | System and method for analysis of biological samples |
CN115398550A (en) * | 2020-03-23 | 2022-11-25 | 基因泰克公司 | Estimating pharmacokinetic parameters using deep learning |
CN115495654A (en) * | 2022-09-26 | 2022-12-20 | 北京邮电大学 | Click rate estimation method and device based on subspace projection neural network |
US11564591B1 (en) * | 2021-11-29 | 2023-01-31 | Physcade, Inc. | System and method for diagnosing and treating biological rhythm disorders |
CN116204831A (en) * | 2022-12-29 | 2023-06-02 | 南京邮电大学通达学院 | Road-to-ground analysis method based on neural network |
-
2023
- 2023-08-10 CN CN202310999925.0A patent/CN116721777B/en active Active
Patent Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20160217267A1 (en) * | 2015-01-20 | 2016-07-28 | Northwestern University | Systems and methods to derive models to evaluate behavior outcomes based on brain responses to complex sounds |
US20200168310A1 (en) * | 2017-05-31 | 2020-05-28 | Jerker Westin | Systems for evaluating dosage parameters |
CN111432720A (en) * | 2017-10-06 | 2020-07-17 | 梅约医学教育与研究基金会 | ECG-based cardiac ejection fraction screening |
CN114830138A (en) * | 2019-10-31 | 2022-07-29 | 伟摩有限责任公司 | Training trajectory scoring neural networks to accurately assign scores |
CN115298534A (en) * | 2019-11-17 | 2022-11-04 | 伯克利之光生命科技公司 | System and method for analysis of biological samples |
CN115398550A (en) * | 2020-03-23 | 2022-11-25 | 基因泰克公司 | Estimating pharmacokinetic parameters using deep learning |
US20220165359A1 (en) * | 2020-11-23 | 2022-05-26 | Peptilogics, Inc. | Generating anti-infective design spaces for selecting drug candidates |
US11263749B1 (en) * | 2021-06-04 | 2022-03-01 | In-Med Prognostics Inc. | Predictive prognosis based on multimodal analysis |
US11564591B1 (en) * | 2021-11-29 | 2023-01-31 | Physcade, Inc. | System and method for diagnosing and treating biological rhythm disorders |
CN115495654A (en) * | 2022-09-26 | 2022-12-20 | 北京邮电大学 | Click rate estimation method and device based on subspace projection neural network |
CN116204831A (en) * | 2022-12-29 | 2023-06-02 | 南京邮电大学通达学院 | Road-to-ground analysis method based on neural network |
Also Published As
Publication number | Publication date |
---|---|
CN116721777B (en) | 2023-10-27 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Ay et al. | Automated depression detection using deep representation and sequence learning with EEG signals | |
Faust et al. | Deep learning for healthcare applications based on physiological signals: A review | |
Maheshwari et al. | Automated accurate emotion recognition system using rhythm-specific deep convolutional neural network technique with multi-channel EEG signals | |
Mumtaz et al. | A machine learning framework involving EEG-based functional connectivity to diagnose major depressive disorder (MDD) | |
Alić et al. | Machine learning techniques for classification of diabetes and cardiovascular diseases | |
Udhaya Kumar et al. | PSO-based feature selection and neighborhood rough set-based classification for BCI multiclass motor imagery task | |
Kass et al. | Computational neuroscience: Mathematical and statistical perspectives | |
Rivera et al. | Diagnosis and prognosis of mental disorders by means of EEG and deep learning: a systematic mapping study | |
Van Steenkiste et al. | Transfer learning in ECG classification from human to horse using a novel parallel neural network architecture | |
Yuan et al. | A hybrid self-attention deep learning framework for multivariate sleep stage classification | |
Hassan et al. | Fusion of multivariate EEG signals for schizophrenia detection using CNN and machine learning techniques | |
Kaya et al. | A new approach for congestive heart failure and arrhythmia classification using angle transformation with LSTM | |
Koçer et al. | Classifying neuromuscular diseases using artificial neural networks with applied Autoregressive and Cepstral analysis | |
Calhas et al. | On the use of pairwise distance learning for brain signal classification with limited observations | |
Cimr et al. | Automatic seizure detection by convolutional neural networks with computational complexity analysis | |
Jose et al. | Adaptive rag-bull rider: A modified self-adaptive optimization algorithm for epileptic seizure detection with deep stacked autoencoder using electroencephalogram | |
Dutta et al. | Deep learning-based multi-head self-attention model for human epilepsy identification from EEG signal for biomedical traits | |
Kim et al. | Fast automatic artifact annotator for EEG signals using deep learning | |
Liu et al. | A novel recognition strategy for epilepsy EEG signals based on conditional entropy of ordinal patterns | |
Shpigelman et al. | Spikernels: predicting arm movements by embedding population spike rate patterns in inner-product spaces | |
Veena et al. | An automated pre-term prediction system using EHG signal with the aid of deep learning technique | |
Erdoğan et al. | Objective pain assessment using vital signs | |
Patel et al. | CNN-FEBAC: A framework for attention measurement of autistic individuals | |
Taori et al. | Cross-task cognitive load classification with identity mapping-based distributed CNN and attention-based RNN using gabor decomposed data images | |
CN116721777B (en) | Neural network-based drug efficacy evaluation method, device, equipment and medium |
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 |