11 results about "Child autism" patented technology

The invention discloses a children autism risk assessment method and system based on multi-modal collaborative reasoning and a vertical domain large model, and the method comprises the steps: collecting the audio data of a to-be-assessed child and a parent in a parent-child interaction link, and collecting the video data through collection equipment; analyzing and marking the collected audio data, and generating a sounding record and a transcription text with a timestamp; performing real-time analysis and labeling on the collected video data, and generating sight line, expression and action recognition records with timestamps; performing independent analysis on the preprocessed multi-modal data, performing feature fusion and collaborative analysis on the multi-modal data such as audio, text and video by using a multi-modal collaborative reasoning method, and generating a scientific data index for autism risk assessment; training a vertical domain large model special for autism risk assessment; and inputting the generated data indexes into a vertical domain large model, and outputting targeted assessment suggestions by the large model in combination with daily information of the children to be assessed obtained by interviews and conversations with parents.

The invention relates to the technical field of early screening of child autism, in particular to a child autism recognition method and system based on fNIRS signals and wearable equipment, and the method comprises the steps: collecting forehead multi-band cerebral blood flow signals and head movement data of a child through an fNIRS intelligent helmet; the method comprises the following steps: eliminating motion artifacts by using an edge computing chip, obtaining a noise-reduced fNIRS signal through adaptive filtering, extracting a time-domain signal slope, a frequency-domain wavelet energy entropy and an inter-channel function connection matrix, and processing time-frequency features and topological features by using a 1D convolution branch and a graph convolution branch by means of a double-current neural network, thereby obtaining a noise-reduced fNIRS signal. After features are fused, an autism risk probability value is output through channel attention weighting, a screening grade report is generated and transmitted to a mobile terminal, the system comprises a signal acquisition unit, an edge processing unit, an intelligent analysis unit and an interactive output unit, the system is adaptive to daily scenes of children, and the screening precision is improved.

The application relates to the technical field of image recognition, in particular to a children autism spectrum disorder stereotyped behavior recognition system based on video images. The system screens target pixel points from pixel points; determines a motion intensity index according to a trajectory point sequence of the target pixel points; converts the trajectory point sequence of the target pixel points into a motion vector sequence; determines direction similarity according to direction similarity between vectors; clusters the vectors based on the direction similarity, and identifies a target cluster; determines a motion regularity index according to the number of vectors in the target cluster and the direction similarity; determines a saliency index in combination with the motion intensity index; clusters the target pixel points based on the saliency index, and sets a region of interest; determines an enhancement weight according to the density of the target pixel points in the region and the saliency index; and determines a total enhancement demand value in combination with the gray scale and gradient features of the pixel points in the region; and reversely maps the total enhancement demand value into a gamma correction parameter, so that the accuracy of recognition is improved.

FIELD: neurology; psychiatry.SUBSTANCE: used as a non-pharmacological therapy for autism spectrum disorders in children. The method includes a set of procedures: interval hypoxic-hyperoxic training (HHT), hyperbaric oxygenation and ozone therapy. HHT consists of alternating inhalations of a hypoxic mixture with an oxygen content of 9 to 16 vol.% and a hyperoxic mixture with an oxygen content of 32 to 40 vol.% for a total duration of 45-50 minutes under the control of cardiovascular system parameters. Hyperbaric oxygenation is performed in a pressure chamber under a pressure of 1.3 atmospheres of 100% oxygen mixture, flow of 10 liters per minute, for 45 minutes using a mask. Ozone therapy is performed by rectal insufflation with an ozone-oxygen mixture with an ozone concentration of 85 μg / ml in a volume of 20–30 ml or by intravenous administration of a physiological solution enriched with ozone with an ozone concentration of 85 μg / ml at a rate of 200 ml per 100 ml of physiological solution. The procedures are carried out daily for 12 days.EFFECT: method is effective and has a positive impact on the patient’s condition.1 cl, 1 dwg, 3 ex

The present application relates to the technical field of early screening of children's autism, in particular, the present application relates to a child autism recognition method and system based on fNIRS signal and wearable device, through the fNIRS intelligent helmet, the multi-band brain blood flow signal of the forehead of children and the head movement data are collected, the motion artifact is eliminated by using the edge calculation chip, and the noise reduction fNIRS signal is obtained through the adaptive filtering, the time domain signal slope and the frequency domain wavelet energy entropy and the inter-channel functional connection matrix are extracted, with the aid of double-flow neural network, the time-frequency characteristics are processed by 1D convolution branch, and the topological characteristics are processed by graph convolution branch, after the fusion characteristics, the autism risk probability value is output by the channel attention weighted output, and the screening grade report is generated again and transmitted to the mobile terminal, the system contains a signal acquisition unit, an edge processing unit, an intelligent analysis unit and an interactive output unit, adapts to the daily scene of children, and improves the screening accuracy.

The present invention relates to a system for early screening of autism spectrum disorder using eye tracking. According to the present invention, autism spectrum disorder may be diagnosed early by applying response data of a user's gaze, which reacts to images or videos outputted on a display, to an autism spectrum disorder screening model. In addition, the present invention may be used to easily check a degree of suspicion of autism spectrum disorder at home, and thus may be utilized in the fields of early diagnosis, prevention, and treatment of autism in children.

The invention discloses a children infantile autism rehabilitation training strategy generation method and system, and relates to the technical field of intelligent rehabilitation. The three core defects of high data subjectivity, strategy templating and adjustment lagging in traditional rehabilitation training are overcome. The key points of the technical scheme are as follows: a time sequence convolutional network and a gating cross-modal attention mechanism are fused to realize accurate association of symptom inducements; performing multi-step reasoning based on the dynamic knowledge graph, and identifying hidden risks; neural development logic sorting and dynamic feasibility verification are realized through a multi-level fusion algorithm; and driving continuous optimization of the knowledge graph by using the feedback data. The full-process intelligent effect from objective evaluation and precise intervention to continuous optimization is achieved, high-accuracy, high-perspectiveness, scientific and effective rehabilitation training support is provided for the autistic children, and the pertinence, predictability and long-term effectiveness of rehabilitation training are remarkably improved.

The invention discloses an intelligent screening method for autism of children, a storage medium and a system. Belongs to the field of machine screening. The method comprises the following steps: acquiring eye movement video data of left and right eyes generated when a screening object watches a three-dimensional video; processing the eye movement video data to generate an eye movement parameter set; wherein the eye movement parameter set comprises a three-dimensional gaze point sequence, a convergence angle sequence and a parallax sequence of left and right eyes; determining multi-dimensional features for screening based on the eye movement parameter set; wherein the multi-dimensional features comprise spatial statistical features, spatio-temporal dynamic features and convergence stability features, inputting the multi-dimensional features into a pre-trained screening prediction model, and outputting a screening prediction result indicating whether the screened object suffers from autism. Therefore, the specific tiny eye movement difference can be identified more comprehensively and accurately, and the accuracy of the output screening prediction result is guaranteed.

The invention discloses a child autism auxiliary diagnosis method based on a large model and fNIRS signals. The method comprises three logic chains: (1) a voice-driven large model inquiry link; (2) a brain function collection and semantic coding link based on fNIRS; and (3) reasoning a link based on a large language model and a medical knowledge base. According to the method, a large model inquiry module driven by voice interaction, an fNIRS signal semantic coding module optimized for an autism scene and a diagnosis and intervention suggestion generation module based on RAG (Retrieval Enhanced Generation) are constructed, so that the operation ability of a child guardian is limited, and an fNIRS signal is easily interfered; objective quantitative evaluation and interpretable auxiliary diagnosis of ASD risks are realized, the use threshold is reduced, and the evaluation efficiency and reliability are improved.

The invention discloses an interactive text generation method for children autism risk assessment. The method comprises the following steps: step 1, acquiring autism domain knowledge and initializing a knowledge database; step 2, capturing expressions and voices of a user in real time from a multi-modal sensor, converting the expressions and the voices into text features, and performing feature fusion; and step 3, loading the fused features into a large language model, and in combination with a knowledge base retrieval enhancement technology, completing one or more rounds of interaction through an analysis chain, an interview chain and a question chain to generate an interactive text. The interactive text generation strategy is iterated and optimized according to the dialogue context and the multi-modal features, and according to the dialogue context, a scale score is summarized through a large language model, and a user portrait and an interview result are output. The invention further provides a generation system for implementing the interactive text generation method and related applications, and the interactive text generation method has wide application value.

Provided herein are metabolites and methods of using the metabolites for diagnosing and treating Autism Spectrum Disorder (ASD) in a subject with a gastrointestinal problem, for determining a personalized treatment protocol for the subject, and for monitoring the therapeutic effect of an ASD treatment protocol in the subject being treated with the ASD treatment protocol.