scispace - formally typeset
Search or ask a question
Author

Li Qiaoqin

Bio: Li Qiaoqin is an academic researcher. The author has contributed to research in topics: Feature extraction & Deep learning. The author has an hindex of 3, co-authored 35 publications receiving 35 citations.

Papers
More filters
Patent
11 Oct 2019
TL;DR: In this article, an exercise prescription recommendation method for the for hypertensive old people based on deep learning was proposed, which comprises the steps: collecting health data; testing the cardiac function state of the hypertensive older people in a quiet state, and evaluating the heart rate, the blood pressure and the variability of heart rate variability; identifying daily actions of the elderly people; evaluating the cardiopulmonary function of the old people; and making personalized exercise prescriptions by referring to FITT rules according to BMI, age, clinical diagnosis and exercise preference.
Abstract: The invention discloses an exercise prescription recommendation method for the for hypertensive old people based on deep learning. The method comprises the steps: collecting health data; testing the cardiac function state of the hypertensive old people in a quiet state, and evaluating the heart rate, the blood pressure and the heart rate variability; identifying daily actions of the hypertensive old people; evaluating the cardiopulmonary function of the hypertensive old people by combining heart rate variability, energy consumption and heart rate according to the exercise-blood pressure risk grade and exercise risk contribution degree, and making personalized exercise prescriptions for the old people by referring to FITT rules according to BMI, age, clinical diagnosis and exercise preference. According to the deep learning scheme provided by the invention, the characteristic representation of exercise parameters such as exercise modes and intensity and time series data such as heart rate, electrocardiogram and real-time blood pressure is better extracted, the relation between exercise and hypertension fluctuation rules is fully learned and mined in combination with population informatics, and a personalized exercise prescription is generated.

4 citations

Patent
14 Jan 2020
TL;DR: In this paper, a drug target affinity prediction method based on deep learning was proposed, which relates to the technical field of drug target affinities prediction, and consists the steps of: obtaining a drug compound and target protein data from a Davis data set and a KIBA data set; encoding the compound, and representing the protein by using a position specificity scoring matrix; inputting a compound label code into a CNN model, and performing feature extraction on the compound to obtain molecular representation of the compound; and learning an order relationship between amino acids in a protein sequence.
Abstract: The invention discloses a drug target affinity prediction method based on deep learning, and relates to the technical field of drug target affinity prediction. The method comprises the steps of: obtaining a drug compound and target protein data from a Davis data set and a KIBA data set; encoding the compound, and representing the protein by using a position specificity scoring matrix; inputting acompound label code into a CNN model, and performing feature extraction on the compound to obtain molecular representation of the compound; inputting the position specificity scoring matrix of the protein into an LSTM model, performing feature extraction on a protein sequence, and learning an order relationship between amino acids in a protein structure and a relationship between residues on the protein sequence to obtain the sequence representation of the protein; and simultaneously inputting the molecular representations of the compounds and the sequence representations of the proteins intoa fully linked layer to predict the affinity of the interaction of the compounds and the proteins. The method can predict the affinity relationship between the drug and the target more accurately.

3 citations

Patent
07 Jan 2020
TL;DR: In this paper, a fall type and injury part detection method based on feature classification is presented, which includes collecting user accelerometer and gyroscope data by a wearable sensor system; performing numerical normalization processing on the acquired sensor data; acquiring time domain and frequency domain features of the preprocessed data; carrying out feature screening by adopting principal component analysis; establishing a tumble detection model based on a random forest, and performing tumble detection and tumble category judgment; and judging a matched fall injury part according to the fall type.
Abstract: The invention belongs to the technical field of electronic information detection, and discloses a fall type and injury part detection method based on feature classification. The method includes: collecting user accelerometer and gyroscope data by a wearable sensor system; performing numerical normalization processing on the acquired sensor data; acquiring time domain and frequency domain featuresof the preprocessed data; carrying out feature screening by adopting principal component analysis; establishing a tumble detection model based on a random forest, and performing tumble detection and tumble category judgment; and judging a matched fall injury part according to the fall type. By adopting the method, the fall type can be judged. The accuracy rate reaches 91% when it is found that falling is detected to be non-falling, and the accuracy rate reaches 89% when different falling types are detected. Through comparison, it is found that the detection rate of each type of fall is higherthan the detection result of current fall direction discrimination research, and the effectiveness of the random forest model provided by the invention is verified.

3 citations

Patent
14 Jun 2019
TL;DR: In this paper, an automatic evaluation method of an upper limb motion function in stroke based on deep learning is presented. But the method comprises the following steps: collecting inertial sensing data andmyoelectric data of the upper limb motions of a patient based on a wearable sensor system; performing length normalization and numerical normalization preprocessing on the collected data; respectively inputting the inertial sensors data and myoelectrics data into two convolutional neural networks for feature extraction, performing fusing all characteristics to generate a motion function level based on the Brunnstrom scale
Abstract: The invention discloses an automatic evaluation method of an upper limb motion function in stroke based on deep learning. The method comprises the following steps: collecting inertial sensing data andmyoelectric data of the upper limb motion process of a patient based on a wearable sensor system; performing length normalization and numerical normalization preprocessing on the collected data; respectively inputting the inertial sensing data and myoelectric data into two convolutional neural networks for feature extraction, performing fusing all characteristics to generate a motion function level based on a Brunnstrom scale, and performing iteration on the model parameters based on a reverse direction propagation algorithm to train a deep learning network model; for patients who need to perform upper limb motion function assessment, performing data acquisition and pretreatment, inputting the data into the trained deep learning model to automatically generate the Brunnstrom staging evaluation results of the upper limb motor function for the patient. The automatic evaluation method can be applied in hospital environment, community and home environment, and can improve the accuracy ofautomatic assessment.

2 citations

Patent
31 Dec 2019
TL;DR: In this paper, a human body tumble detection method based on multi-source heterogeneous data fusion was proposed, which comprises the following steps: acquiring a behavior depth image and skeleton information through Kinect, and getting rid of constraints of a wearable sensor on selection of the sensor; secondly, solving theproblem that wearable sensors cannot be used in specific scenes such as bathrooms and toilets, and meanwhile, avoiding the problem that human privacy is invaded due to the fact that a common camera isused for monitoring.
Abstract: The invention belongs to the field of human body tumble detection, and provides a human body tumble detection method based on multi-source heterogeneous data fusion. The method comprises the followingsteps: acquiring a behavior depth image and skeleton information of a human body through Kinect, and getting rid of constraints of a wearable sensor on selection of the sensor; secondly, solving theproblem that wearable sensors cannot be used in specific scenes such as bathrooms and toilets, and meanwhile, avoiding the problem that human privacy is invaded due to the fact that a common camera isused for monitoring; meanwhile, extracting features from multi-source heterogeneous data through a deep learning model, introducing keyless attention fusion into a data fusion mode, and avoiding dataredundancy and calculation complexity generated by data-level fusion. Compared with the prior art, the accuracy of tumble detection is remarkably improved.

2 citations


Cited by
More filters
Patent
07 Feb 2020
TL;DR: In this article, the authors proposed a method for selecting positive sample data in source domain data to extend training data of a target domain by fusing semantic difference and label difference of sentences in the source domain and the target domain, so as to achieve the purpose of enhancing named entity recognition performance.
Abstract: The invention provides a method for selecting positive sample data in source domain data to extend training data of a target domain by fusing semantic difference and label difference of sentences in the source domain and the target domain, so as to achieve the purpose of enhancing named entity recognition performance of the target domain Based on a conventional Bi-LSTM+CRF model, in order to fusesemantic differences and label differences of sentences in a source domain and a target domain, semantic difference and label difference are introduced through state representation and reward settingin reinforcement learning; therefore, the trained decision network can select sentences having positive influence on the named entity recognition performance of the target domain in the data of the source domain, expand the training data of the target domain, solve the problem of insufficient training data of the target domain, and improve the named entity recognition performance of the target domain at the same time

3 citations

Patent
10 Jul 2020
TL;DR: In this paper, an adaptive decision tree fall detection method and system is presented, which consists of the following steps: 1, acquiring three-axis acceleration and threeaxis angular velocity data of falling and non-falling actions of a human body, and performing screening; 2, calculating resultant acceleration and resultant angular acceleration, dividing a training set, substituting a test set and a verification set into a TSFRESH library to calculate features, and screening and deleting useless features; 3, selecting preliminary important features by using a random forest sieve; 4, establishing a decision tree model
Abstract: The invention discloses an adaptive decision tree fall detection method and system, and belongs to the technical field of human body behavior recognition and judgment. The method comprises the following steps: 1, acquiring three-axis acceleration and three-axis angular velocity data of falling and non-falling actions of a human body, and performing screening; 2, calculating resultant accelerationand resultant angular acceleration, dividing a training set, substituting a test set and a verification set into a TSFRESH library to calculate features, and screening and deleting useless features; 3, selecting preliminary important features by using a random forest sieve; 4, establishing a decision tree model for training and verification, and testing a result; 5, continuing to obtain a new sample, repeating the steps 2 to 4, and updating the decision tree model. According to the invention, an accurate tumble judgment result can be obtained through a decision tree algorithm with a small calculation amount; after a certain number of samples are collected, the decision tree model is updated, so the judgment precision of the algorithm can be further improved.

2 citations

Patent
15 May 2020
TL;DR: The BERT pre-training model is used for text classification in this article, where the added input layer is a feature representation layer for assisting classification and recognition, and the BERT model does not need to be improved by a Google company and the like.
Abstract: The invention relates to a text classification method and system, electronic equipment and a computer readable storage medium, and the method comprises the steps: adding an input layer of a BERT pre-training model, enabling the BERT pre-training model to participate in the training, and carrying out the classification and recognition of a to-be-classified text based on a classification model obtained after training, wherein the added input layer is a feature representation layer for assisting classification and recognition. By adding the input layer of the BERT pre-training model, the reference characteristic quantity during model text classification and recognition is increased, so that the text classification accuracy can be improved. Besides, the structure of the BERT pre-training modelis not changed, so that the BERT model does not need to be improved by a Google company and the like requesting to provide the BERT pre-training model, namely, the method is not limited by a basic model providing company, and the problem of inconvenience does not exist.

2 citations

Patent
19 Jun 2020
TL;DR: In this paper, the authors proposed an epidemic situation label data processing method and system, which comprises the steps of: extracting tracking behavior characteristics of a tracking information sequence recorded by each user terminal; screening out a first tracking behavior characteristic of a epidemic situation related user and a second tracking behaviour characteristic of an to-be-tested user; comparing the second tracking behavior attributes of the to be tested user with the first tracking behaviour characteristics of each epidemic situation-related user; and determining characteristic attributes of each user under the different epidemic situation influence type labels, thereby sending corresponding epidemic situationlabel information
Abstract: The invention relates to the technical field of data processing, and relates to an epidemic situation label data processing method and system. The method comprises the steps of: extracting tracking behavior characteristics of a tracking information sequence recorded by each user terminal; screening out a first tracking behavior characteristic of a epidemic situation related user and a second tracking behavior characteristic of a to-be-tested user; comparing the second tracking behavior characteristics of the to-be-tested user with the first tracking behavior characteristics of each epidemic situation related user; and determining characteristic attributes of the to-be-tested user under the different epidemic situation influence type labels, thereby sending corresponding epidemic situationlabel information to a user terminal of the to-be-tested user according to the characteristic attributes of the to-be-tested user under the different epidemic situation influence type labels for prompting. According to the invention, differentiated epidemic situation label analysis can be realized for each individual in a crowd, so that rapid and accurate prompt that the infectious disease epidemic situation may be affected is realized, and compared with a traditional scheme, the method and the system can effectively provide the accuracy degree of epidemic situation influence prompt.

2 citations

Patent
15 Oct 2019
TL;DR: Wang et al. as mentioned in this paper proposed a traditional Chinese medicine diagnosis and treatment knowledge graph automatic construction method based on deep learning, which comprises the steps: constructing an initialized literature medical record corpus, carrying out sentence segmentation and word segmentation on a medical record, and marking a theory-law-prescription-medicine entity in the medical record; predicting the entity through a bidirectional LSTM, and automatically extracting the entity from traditional Chinese Medicine literature medical records through a deep learning model; and clustering similar entities appearing in the same medical record to form an
Abstract: The invention discloses a traditional Chinese medicine diagnosis and treatment knowledge graph automatic construction method based on deep learning. The traditional Chinese medicine diagnosis and treatment knowledge graph automatic construction method comprises the steps: constructing an initialized literature medical record corpus, carrying out sentence segmentation and word segmentation on a medical record, and marking a theory-law-prescription-medicine entity in the medical record; predicting the entity through a bidirectional LSTM, and automatically extracting the entity from traditional Chinese medicine literature medical records through a deep learning model; and clustering similar entities appearing in the same medical record to form an entity group, then forming a triple accordingto a predefined relationship between entities, and constructing a knowledge graph. According to the invention, the relationship between traditional Chinese medicine diagnosis and treatment concepts ispredefined; construction of the knowledge graph is converted into a traditional Chinese medicine diagnosis and treatment named entity recognition task; and entities are automatically extracted from traditional Chinese medicine literature medical records through a deep learning model, and the entities are clustered to form an entity set, so that the many-to-many problem between traditional Chinesemedicine diagnosis and treatment concepts is solved, and the famous and old traditional Chinese medicine diagnosis and treatment thought in the medical records is completely displayed.

1 citations