61 results about "Biological information processing" patented technology

The invention provides an interpretable machine learning-based giant coronary artery tumor prediction method in Kawasaki disease, and belongs to the technical field of biological information processing. Comprising the following steps: S1, collecting hospitalized KD child patient cases by adopting a retrospective queue research method, and excluding cases which do not meet requirements; s2, coronary artery lesion evaluation and classification; s3, performing variable preprocessing; s4, constructing a machine learning model; s5, performing variable screening; s6, model interpretation; s7, determining an optimal intervention threshold value; and S8, developing an online prediction tool. According to the method, an SVM kernel lab model with the optimal performance is screened out through multi-model comparison, and a group of key prediction molecules are determined in combination with an SHAP method and recursive feature elimination; based on a multi-center data verification model, the model applicability is improved; and early recognition and layered management of high-risk child patients can be realized. A webpage tool is developed, and a patient with high risk is suggested to intervene in advance.

The invention relates to the technical field of protein related data processing, in particular to a biological information processing system and method applied to synthetic biology. The method comprises the following steps: acquiring real-time proteomics data, and performing polymorphic context decoding to obtain a protein expression context matrix; analyzing interaction of behavior characteristics in the protein expression context matrix to obtain a multi-scale causal structure map; executing structure-function transformation rule extraction in the multi-scale causal structure atlas to obtain a function mapping unit set; performing evaluation based on the function mapping unit set so as to form a configuration decision diagram; simulating paths in the configuration decision diagram, and performing path adaptability scoring to obtain a path adaptability feedback table; and optimizing the path adaptability feedback table to obtain an optimal expression configuration set. According to the method, the precision, the stability and the controllability of protein information in the process of structural analysis, function recognition and regulation path construction can be improved.

The invention provides an IVIG drug resistance prediction method in KD based on an interpretable machine learning model, and belongs to the technical field of biological information processing. Comprising the following steps: S1, data acquisition; s2, data processing; s3, constructing a model; s4, a model interpretation module; and S5, webpage deployment. According to the invention, based on the large-scale electronic medical record data of the Kawasaki disease patient, a machine learning model with good interpretability is constructed for predicting the IVIG resistance, and an online webpage calculation tool is established for clinical doctors to input key indexes in real time and automatically output the IVIG resistance risk probability of the patient. According to the tool, 11.6% of the optimal prediction threshold value is set, doctors can be helped to identify high-risk individuals, treatment strategies can be adjusted in time, and coronary artery complications are reduced.

This invention relates to the field of bioinformatics processing and medical laboratory data analysis, specifically a method for classifying hepatitis B antibody patterns based on plasma protein profiles. The method includes: acquiring host hardware information of the execution environment and extracting central processing unit (CPU) cache parameters; acquiring a high-dimensional sparse one-dimensional array and extracting non-zero feature indices; performing mapping calculations using a locality-sensitive hashing (LSH) algorithm to reconstruct the data into a locally dense two-dimensional matrix; dynamically segmenting the data into independent sub-blocks according to cache parameters and initial segmentation dimensions; performing low-rank tensor decomposition on the independent sub-blocks to extract local latent feature vectors; concatenating the sub-blocks, weighting them through a single-layer attention network, and inputting the result into a classifier function to output a target classification pattern vector; and generating dimension update instructions based on a preset dynamic adjustment mechanism to adjust subsequent segmentation dimensions. This invention achieves lower memory peaks, higher cache hit rates, and stable multi-label pattern output capabilities.

This invention relates to the field of bioinformatics processing and discloses a method and system for mining the structure-activity relationship (SMR) of porcine neurotrophic peptides based on sequence features. The method includes constructing an original sample index table and fusing multi-source production data, extracting peptide sequence features to generate a sequence feature matrix, constructing a sequence-process joint graph containing peptide nodes and process state nodes, training a structure-activity relationship graph neural network to mine SMR relationships, and deriving a process control decision table based on a process response sample set generated by the network, thereby achieving online optimization of the porcine neurotrophic peptide preparation process. This invention solves the problem of SMR mining caused by the separation of process parameters, sequence information, and activity data, achieving accurate characterization of the synergistic effect of sequence and process and reverse optimization of process parameters, thus improving the targeted enrichment efficiency and bioactivity retention level of target neurotrophic peptides.

The invention relates to the technical field of biological information processing and automatic sorting, and discloses a chick gender automatic identification method and system based on multi-modal fusion, and the method comprises the steps: collecting original three-dimensional point cloud data through a three-dimensional form sensing module, and calculating the key region coordinates of chicks; based on the coordinates, controlling the addressable micro sampling array to execute active sampling adaptation, and obtaining measurement response time sequence data of the target volatile matter; acquiring background response time sequence data by utilizing a background channel sensor array; and performing differential processing on the two to generate a differential biochemical response time sequence vector. And then, respectively extracting morphological feature vectors and biochemical fingerprint feature vectors, inputting the morphological feature vectors and the biochemical fingerprint feature vectors into respective independent classifiers to obtain prediction categories and confidence coefficients, finally, judging a bimodal prediction result through preset cross validation gating fusion logic, outputting a final decision, and controlling a sorting execution module to complete physical sorting. And the accuracy of biochemical sampling is improved through active sampling adaptation.

To reduce a load on evaluating persons performing pain evaluation and reduce variations in the evaluation among the evaluating persons.SOLUTION: A biological information processing device includes an acquisition unit for acquiring blood vessel kinetic information indicating a blood circulation state of a subject, and a processing unit for acquiring an index value of pain evaluation for the subject by subjecting the blood vessel kinetic information to normalization processing. The processing unit calculates a reference value on the basis of the blood vessel kinetic information acquired in a reference period before starting a treatment event on the treatment for the subject, calculates an object value on the basis of the blood vessel kinetic information acquired after starting the treatment event, and calculates the ratio of the object value to the reference value as the index value as the normalization processing.SELECTED DRAWING: Figure 2

The invention belongs to the technical field of biological information processing, and provides a cancer typing method and system based on cross-modal attention and variational representation learning, and the method comprises the steps: obtaining the multi-modal medical data of a patient, coding the multi-modal medical data, mapping the multi-modal medical data to a modal specific representation space of a unified dimension, and obtaining the feature representation of a corresponding modal; a multi-head attention mechanism is introduced, for each attention head in the multi-head attention mechanism, an independent mapping space is constructed for each mode, output of the multiple attention heads is spliced on the feature dimension, and through linear transformation, multi-mode fusion representation aggregating cross-mode complementary information is obtained; deducing statistical parameters of potential posteriori distribution represented by multi-modal fusion, and sampling to obtain potential variables; based on the potential variables and the classifier, obtaining a cancer typing result through Softmax; and end-to-end accurate classification is realized.

The application relates to the technical field of protein-related data processing, in particular to a biological information processing system and method applied to synthetic biology. The method comprises the following steps: acquiring real-time proteomics data, performing polymorphic context decoding, and obtaining a protein expression context matrix; analyzing the interaction of behavior characteristics in the protein expression context matrix, and obtaining a multi-scale causal structure graph; performing structure-function transformation rule extraction on the multi-scale causal structure graph, and obtaining a function mapping unit set; evaluating the function mapping unit set, thereby forming a configuration decision graph; simulating the path in the configuration decision graph, performing path adaptability scoring, and obtaining a path adaptability feedback table; and optimizing the path adaptability feedback table, thereby obtaining an optimal expression configuration set. The application helps to improve the accuracy, stability and controllability of protein information in the structural analysis, function identification and regulation path construction process.

This application discloses a method, apparatus, storage medium, and electronic device for cross-well identification, relating to the field of bioinformatics. It involves obtaining a first matrix corresponding to each well position in a sequencing chip, translating the first matrix along at least one target direction to obtain at least one second matrix, and comparing each second matrix with the first matrix to determine the sequence similarity between each well position and its adjacent well positions along the corresponding target direction. Based on the sequence similarity, cross-well behavior is identified. Thus, by mapping the sequence data of the chip well positions into a matrix structure and utilizing matrix translation operations to achieve batch alignment of adjacent well positions, this application achieves efficient identification and quantitative evaluation of cross-well behavior, solving the problems of high computational load and low efficiency in traditional point-by-point traversal methods, and improving the accuracy of cross-well behavior identification in sequencing chips.

This invention provides a method, storage medium, and electronic device for accurately and efficiently assessing gene editing efficiency based on NGS data, belonging to the field of bioinformatics processing technology. This invention integrates the following steps: splitting NGS pooled sequencing data according to barcode tag sequences; constructing a reference genome index and dictionary; aligning the NGS data of each sample with the reference genome; calculating the alignment rate; constructing an alignment result index; converting the alignment results to TSV format; and evaluating the editing efficiency of gene editing target sites in each sample. This approach meets the need for accurate and efficient assessment of gene editing efficiency based on NGS data. The method effectively avoids the limitations of existing tools on data file size, allows for custom barcode tag sequences for data splitting, utilizes mainstream alignment algorithms to obtain high-accuracy alignment results, and leverages the advantages of computing clusters for parallel computation, thereby achieving the goal of accurately and efficiently assessing the gene editing efficiency of multiple genes and a large number of samples.

To improve measurement accuracy and responsiveness without limiting the cycle of measurable biological information.SOLUTION: A control unit of a bedside terminal analyzes biological information signals of a plurality of slots having different lengths as analysis targets among biological information signals of a subject acquired from a sensor unit, and generates information on periodicity for each slot. Next, the control unit derives a reliability index indicating reliability of the information on the periodicity for each of the plurality of slots. Then, the controller selects the slot having the highest reliability index, and calculates and outputs the biological information of the subject based on the information relating to the periodicity of the selected slot.SELECTED DRAWING: Figure 4

The invention provides a biological information processing apparatus, an information processing system, a biological information processing method, a storage medium, and a program product. The biological information processing apparatus includes: a biological information acquisition unit that acquires biological information; a biological information using unit that uses the biological information and the features extracted from the biological information; an explicit agreement acquisition unit that acquires agreement of the use of the biological information of the person; a biological information use restriction unit that restricts the use of the biological information on the basis of the agreed value; and an agreement state estimation unit that estimates an agreement state indicating agreement of the use of the biological information of the person on the basis of the acquisition state when the biological information is acquired. Further, a biological information use restriction unit relaxes restrictions on the use of the biological information of the person and features extracted from the biological information on the basis of the agreement state estimated by the agreement state estimation unit.

A biometric information processing method based on biometric recognition, which method is applied to a client, wherein the client runs on a mobile terminal. The method comprises: detecting an instruction for indicating activation of at least two cameras to respectively collect biometric information (S210); in response to the instruction, activating the at least two cameras associated with a mobile terminal, in order to respectively obtain images containing biometric information that are acquired by the cameras (S220); and sending to a server side the images containing biometric information that are acquired by the cameras, such that the server side identifies the biometric information contained in the images to obtain at least two types of biometric information, and associating the at least two types of biometric information with the same user identifier (S230).

The application discloses a mechanism-driven nanobody-antigen binding prediction method and system, relates to the technical field of bio-information processing and artificial intelligence, and inputs amino acid sequences of nanobodies and antigens into a prediction system for processing, and a construction process of the prediction system comprises the following steps: after an encoder encodes the amino acid sequences of the nanobodies and the amino acid sequences of the antigens, global features and local features of respective complementarity determining regions are extracted and fused to obtain nanobody fusion features and antigen fusion features, and the nanobody fusion features and the antigen fusion features are subjected to global average pooling to obtain an encoder feature of the nanobodies and an encoder feature of the antigens; a combination prediction model models the interaction between the encoder feature of the nanobodies and the encoder feature of the antigens through Hadamard product modeling, generates an interaction feature, and predicts a combination probability based on the interaction feature. Through the cooperative work of the mechanism-driven encoder and the prediction system, high-precision and high-efficiency nanobody-antigen binding prediction is realized.

The invention relates to the technical field of biological information processing, in particular to a cattle embryo pre-implantation genetic evaluation method and system based on whole genome sequencing, and the method comprises the following steps: obtaining whole genome sequencing data of a male parent and a female parent, embryo trophoblast live detection sequencing data and culture solution free desoxyribonucleic acid sequencing data; performing variation detection and haplotype phasing on male and female parent data to form haplotype data, and calculating a haplotype transmission posterior at an anchor point based on trophoblast data; performing genome segmentation based on reading depth observation and allele frequency observation of a trophoblast and a culture solution, and performing Bayesian inference by combining haplotype transfer posteriori as priori to obtain posteriori probabilities of aneuploid, copy number variation, chimera, pathogenic homozygosis and pollution events; and outputting a passing / rechecking / elimination conclusion, adding sequencing and updating the posteriori during rechecking until ending, and outputting a final conclusion and a genome breeding value, so that the evaluation reliability and the decision interpretability are improved.

The invention belongs to the technical field of biological information processing, and particularly relates to a method and system for constructing a risk prediction model from Kawasaki disease to huge coronary artery tumor. The invention provides a Kawasaki disease MGCAA risk prediction model construction method and system, and the method comprises the steps: building a prediction model based on six conventional clinical variables through employing a random forest (RF ranger) algorithm; an SHAP explanation mechanism is introduced, so that the contribution of each variable to a prediction result can be visually displayed in global and individual levels, and the transparency and clinical interpretability of the model are improved; through external verification and intercept-only recalibration, the reliability and the applicability of the model in different groups of people can be ensured; according to the method, an online webpage tool (Shiny App) is deployed, and a doctor can input clinical data of a patient in real time and immediately obtain individualized risk prediction and explanation, so that clinical early recognition of an MGCAA high-risk child patient is realized, and formulation of an individualized treatment scheme and early intervention measures is assisted.

The invention relates to the technical field of biological information processing and artificial intelligence, and provides a dual-channel drug interaction prediction method of a positive definite non-commutative polynomial filter, which comprises the following steps: constructing a heterogeneous drug knowledge graph, and constructing a heterogeneous sub-graph by adopting a sampling and random walk strategy; a two-channel graph representation learning model is constructed, semantic embedded drug representation is obtained through a semantic channel, and topological embedded drug representation is obtained through a topological channel by adopting a frequency domain graph convolution method based on a positive definite non-exchangeable polynomial filter; alignment and splicing fusion are carried out through a mutual information alignment mechanism, and fusion features are obtained; nonlinear transformation is carried out on the fusion features through MLP, and prediction vectors are mapped into probability distribution of the drug interaction relation through a softmax function. According to the method, the limitation of traditional spatial domain convolution on high-order semantic capture and heterogeneous structure adaptability is made up, and the calculation efficiency and the structure expression ability of the model are remarkably improved while the prediction precision is ensured.

The present disclosure relates to a de-identification method, a bio-information processing method based on de-identification, and a device thereof. It relates to a technology for protecting visual features within an image, and a de-identification method receives an image including a face region, extracts feature information about the face region from the input image, transforms the extracted feature information to provide guidance for a part to be generated within the face region, and performs de-identification based on the provided guidance.

To reduce the burden on the subject when measuring blood pressure. [Solution] The biological information processing device 1 includes a blood pressure detection unit 13 that identifies a reference pulse wave from pulse waves measured in the past and detects a first reference blood pressure value, a second reference blood pressure value, and a third reference blood pressure value, based on the amplitude of the reference pulse wave; and a blood pressure estimation unit 14 that, when a part of the waveform including the rising edge of a newly measured pulse wave corresponds to a part of the waveform including the rising edge of the reference pulse wave, estimates at least one value or range of the first blood pressure value, second blood pressure value, and third blood pressure value based on the new pulse wave, based on the corresponding value among the first reference blood pressure value, second reference blood pressure value, and third reference blood pressure value.

The invention relates to the technical field of biological information processing, in particular to a specimen biological information analysis and verification system based on big data, which comprises a big data acquisition module for acquiring specimen morphological data, gene sequence data and associated ecological environment data; the biological information analysis module mines a specimen feature association rule based on a big data algorithm to generate an analysis result; the multi-dimensional verification module performs cross verification on the analysis result according to form comparison, gene sequence verification and ecological habit matching dimensionality, and outputs a verification report; the species adaptation module autonomously optimizes and analyzes algorithm parameters and verifies weight distribution according to different class group specimen characteristics, storage states and verification reports; and the cloud co-processing module stores the specimen multi-dimensional data and the analysis and verification result to the cloud, monitors the data quality in real time, dynamically adjusts the storage resource allocation and returns the storage resource allocation to the user side. Therefore, the problems of data dispersion, low verification accuracy and the like in the prior art are solved.

A weight measurement apparatus of the embodiment includes processing circuitry. A weight measurement apparatus includes the processing circuitry configured to acquire a body weight of a subject and biological information other than the body weight, calculate a degree of abnormality of the body weight based on a predetermined threshold, calculate a degree of abnormality of a correlation based on a correlation relationship between the body weight and the biological information, and output information indicating the degree of abnormality in the body weight and information indicating a degree of deviation between the correlation relationship used in the calculating of the degree of abnormality in the correlation and a reference correlation relationship.

To provide a biometric information processing apparatus capable of reducing mismatch, and a biometric information processing system.SOLUTION: A biometric information processing apparatus includes a derivation unit and a classification unit. The derivation unit derives emotion information on an object living body based on at least one of biometric information and behavior information which are obtained from the object living body executing a specific task. The classification unit classifies the emotion information obtained by the derivation unit on the basis of a predetermined classification index.SELECTED DRAWING: Figure 2

The invention belongs to the technical field of biological information processing and application, and particularly relates to an autism detection model training method based on self-supervised space-time contrast learning, and the method comprises the steps: obtaining the resting-state functional magnetic resonance images of the brains of a plurality of subjects; for the functional magnetic resonance image of the brain resting state of each subject, extracting sub-images of a plurality of selected areas from the functional magnetic resonance image, and constructing a brain function nuclear magnetic network of the subject according to the sub-images so as to obtain an unlabeled sample of model training; obtaining at least two different enhancement expressions of a label-free sample, and carrying out self-supervised contrast learning training on a time encoder and a space encoder in the autism detection model by using the enhancement expressions, so as to optimize fusion analysis of the autism detection model on sample features from a space-time dimension; and randomly extracting part of samples from all the unlabeled samples, and labeling the unlabeled samples according to the illness condition of the subject, so as to finely adjust the trained autism detection model by using the labeled samples.

A radar (R) outputs a plurality of detection signals obtained by emitting radar waves in a room (E), which is a measurement region. In a biological information processing device (100), an acquisition unit (101) acquires the detection signals to identify an area in which a human U as the subject is present. The acquisition unit (101) selects and acquires, within the identified area, a plurality of detection signals at a plurality of positions in which body movement and vital information of the human U have values indicating activity. A signal correction unit (102a) corrects the signal intensity of the detection signals on the basis of the signals in the most recent time range. A feature amount extraction unit (102) extracts feature amounts of the body movement and the vital information from the plurality of detection signals after correction. A determination unit (103) determines a biological state of the human U, for example, a respiratory state, on the basis of the extracted feature amounts. For example, when the respiratory state continuously decreases over time, the determination unit (103) determines that the human U is in a dangerous state and outputs an alarm.

To provide a technology to associate a person to be measured with biological information.SOLUTION: A biological information processing device includes: a signal reception unit for receiving a signal on biological information reflected from at least one person to be measured; a candidate region specification unit for calculating an arrival direction of the signal and / or a distance to the person to be measured from the received signal, and specifying a candidate region of the person to be measured using the calculated arrival direction and / or the distance; an information generation unit for generating biological information corresponding to the candidate region of the person to be measured from the received signal; a position information acquisition unit for acquiring position information on the person to be measured; and a biological information association unit for associating the person to be measured with the generated biological information on the basis of the acquired position information.SELECTED DRAWING: Figure 3

The invention relates to the technical field of biological information processing, in particular to a sex cell automatic classification and recording method and system.The method comprises the following steps that the sex cell DNA synthesis rate is obtained through a pulse labeling method, the S period starting time and the G2 / M transition duration are extracted, the cell nucleus area and chromatin distribution are combined, XIST and SRY genes are screened, sex specific changes are analyzed, and the sex cell DNA synthesis rate is obtained. And matching a gender track, calculating an S period proportion and a G2 / M duration, outputting a gender classification label, and integrating an appearance modification data update record. According to the method, nucleotide and fluorescence microscopic imaging are introduced, the DNA synthesis rate is quantified, a characteristic trajectory is established in combination with cell cycle key nodes and nuclear morphology, gender specificity is analyzed in combination with XIST and SRY gene expression and chromosome signals, classification basis comprehensiveness is improved, a screening threshold value is set to be matched with the gender trajectory, and the accuracy of gender detection is improved. The cycle ratio is calculated, the classification accuracy is improved in combination with the confidence coefficient, genetic and epigenetic features are integrated during data storage, and subsequent analysis is supported.

The present invention relates to the technical field of bioinformatics processing, and specifically to a method and system for preparing CIK cells, comprising the following steps: collecting cell density data in a time series, calculating autocorrelation function values and partial autocorrelation function values, extracting first-order and second-order difference sequences, calculating the ADF statistic and the KPSS statistic based on the difference sequences, comparing the statistic ranges to determine the stationarity of the time series, and generating stationary cell density sequence data. In the present invention, through the extraction and optimization of lag correlation values and partial correlation values, the accuracy of model parameters is enhanced, providing reliable support for the dynamic correction and joint analysis of the future cell proliferation rate. Based on the analysis of difference means and weighted cumulative values, abnormal points are accurately identified, significantly improving the efficiency of anomaly detection. Combining the abnormal point index and the drift rate calculation of environmental parameters, the dynamic adjustment of culture conditions is realized, reducing the impact of environmental fluctuations on the quality of cell proliferation, and ensuring the preparation quality and efficiency.

This invention provides a method and system for cancer subtype identification based on multi-omics data, belonging to the field of bioinformatics processing technology. The method includes: constructing a DILCORE model that integrates a multi-branch variational autoencoder, contrastive learning, and cross-view attention mechanisms. The multi-branch variational autoencoder decomposes omics observation data into common components for subtype classification and view-specific noise components, achieving noise suppression; the InfoNCE contrastive loss is introduced to bring common representations of the same sample closer across different views, enhancing cross-omics consistency; the cross-view residual self-attention mechanism is used to adaptively weight and fuse common vectors; finally, a self-supervised clustering fine-tuning optimization strategy is introduced to jointly improve representation quality and clustering performance in the latent space. This achieves deep and effective integration of multi-omics data, significantly improving the accuracy of cancer subtype identification and providing a powerful tool for personalized cancer treatment and prognostic assessment.

A biological information processing device acquires biological information including a volume pulse wave and a heart sound, performs second-order differentiation on the volume pulse wave to derive an acceleration pulse wave, determines in the acceleration pulse wave an offset position occurring an offset time after the beginning of a grade II sound, backtracks from a local minimum point subsequent to the offset position to find a first local maximum point, determines a point in the volume pulse wave corresponding to the first local maximum point to be a dicrotic notch, and stores information indicating the dicrotic notch in association with or as part of the biological information.