8 results about "Retrospective analysis" patented technology

This invention discloses a trial auxiliary data management system and method based on judicial case quality evaluation, belonging to the field of data processing technology. By decomposing and reorganizing core and auxiliary cases, a cache of related cases is constructed. This allows for further mining of related cases from both core and auxiliary case perspectives, outputting the most similar related cases to the submitted cases to be evaluated. Simultaneously, using the optimal solution as a reference, cases requiring random checks can be quickly identified, significantly reducing errors and inaccuracies caused by manual operation, improving trial quality and efficiency. Furthermore, the optimal solution found can provide positive or negative preventative references for trial retrospective analysis.

The invention discloses a digital twinning modeling and evaluation method for typhoon resistance toughness of a culture platform, and relates to the technical field of ocean engineering digital twinning and risk evaluation.According to the scheme, a high-fidelity digital twinning body integrating multiple physical field models is constructed, and real-time data bidirectional interaction with a physical platform is established; on the basis, anti-typhoon toughness quantitative indexes including robustness, redundancy and restorability are systematically defined; by using the digital twinborn body, dynamic deduction, real-time evaluation and backtracking analysis of the whole process of the typhoon can be carried out, and the toughness state of the platform can be accurately observed; and finally, based on an evaluation result, simulating, optimizing and recommending an optimal toughness enhancement strategy in a digital space. According to the method, the anti-typhoon capability of the culture platform is predicted from static evaluation to dynamic prediction and from a single safety coefficient to comprehensive toughness management, and a core technical support is provided for safe operation of deep and far sea culture.

Systems, methods and computer-readable media for providing explainable selective attention in multi-source or multi-speaker environments. A selective attention module receives multimodal sensor data including audio, video, gaze, text, and physiological signals from a plurality of sources. An attention inference engine generates attention distributions over the sources and fuses them into a probabilistic belief state. An explainability module produces interpretable outputs corresponding to the fused belief, including attention matrices, confidence scores, reliability measures, margin-based differentiators, and natural language rationales. The explainability outputs are rendered through visual, auditory, or augmented / virtual reality interfaces to indicate the attended source, suppressed sources, and reasoning for the selection. The system enables user interaction by providing justifications in real time, logging explanations for retrospective analysis, and supporting adaptation of thresholds and model weights based on feedback. The disclosed technology improves transparency, interpretability, and trust in selective attention systems, while maintaining real-time performance in dynamic multi-speaker environments.

The invention relates to the technical field of clinical diagnosis, and particularly discloses a construction method and application of a primary liver cancer diagnosis model GAAAD. According to the method, clinical information of patients with primary hepatocellular carcinoma, patients with other liver diseases and healthy people is subjected to retrospective analysis, a multi-factor Logistic regression method is utilized, and a GAAAD combined diagnosis model composed of five variables including gender, age, AFP, AKR1B10 and PVIKA-II is constructed. Clinical data verifies that the model is excellent in prediction distinction degree and accuracy, can accurately evaluate the risk of an individual suffering from primary hepatocellular carcinoma, has important clinical value in early prevention, diagnosis and treatment and prognosis evaluation of diseases, and can assist in formulating a scientific diagnosis and treatment scheme.

The invention provides a puerpera and fetus two-dimensional synergistic painless delivery anesthesia dynamic regulation and control monitoring method, and belongs to the technical field of painless delivery anesthesia. The method comprises the following steps: constructing a puerpera-fetus two-dimensional monitoring system, and synchronously collecting multidimensional signals such as puerpera side pain perception, circulatory respiration and drug metabolism and core signals such as fetus side heart rate and blood oxygen saturation; performing adaptive filtering preprocessing and attention mechanism driven multi-source data fusion analysis to obtain a puerpera anesthesia state, a fetus safety state and a collaborative correlation coefficient of the two states; constructing a regulation and control model by taking a parturient painless threshold-a fetus safety threshold as double constraints, outputting an initial anesthesia scheme, and performing real-time dynamic adjustment in combination with a delivery process and drug interaction; meanwhile, full-process data recording and three-level early warning are achieved, and the scheme is optimized through backtracking analysis after delivery. The problems that in the prior art, puerpera pain relieving is emphasized, and fetus safety cooperative monitoring and regulation lack dynamics are neglected are solved, and anesthesia precision and mother and infant safety are improved.

Methods of reliably estimating genomic fraction (e.g., fetal fraction) from polymorphisms such as small base variations or insertions-deletions are disclosed. Sequenced data from a multigenomic source is used to determine allele counts for one or more of the polymorphisms. For one or more of the polymorphisms, zygosity is assigned, and genomic fraction is determined from the zygosity and allele counts. Certain embodiments employ SNPs as the relevant polymorphism. The disclosed methods can be applied as part of an intentional, pre-designed re-sequencing study targeted against known polymorphisms or can be used in a retrospective analysis of variations found by coincidence in overlapping sequences generated from maternal plasma (or any other setting where a mixture of DNA from several people are present).

A robotic surgical system includes a robotic instrument coupled to a controller and configured to interact with biological tissue. A physiologic sensing subsystem has one or more sensors configured to detect real-time physiological parameters. The sensors include at least one of: tissue oxygenation sensors, perfusion sensors, bioimpedance sensors, or neural conductivity sensors. A boundary mapping engine includes an artificial intelligence (AI) module trained to: generate dynamic boundary zones within a surgical field based on sensed physiological thresholds and contextual procedural data; classify zones as high-risk based on learned thresholds derived from patient-specific models, historical surgical outcomes, or intraoperative signal patterns; and generate an instrument constraint map corresponding to the boundary zones and continuously refine it based on ongoing sensor inputs and AI-inferred trends; A control module configured to: receive the instrument constraint map; restrict, dampen, or modify robotic instrument movement preventing intrusion into AI-classified high-risk boundary zones; and log override attempts, constraint violations, and AI confidence scores for real-time or retrospective analysis. A visualization interface presents real-time physiological boundary overlays, constraint zones, and AI-generated risk indicators to the surgeon.

This invention belongs to the field of medical data analysis and provides a retrospective analysis method and system for calculating physiological data thresholds based on mathematical inflection points. The method includes: acquiring the patient's historical physiological data and the patient's outcome data; using the patient's historical physiological data as the independent variable and the patient's outcome data as the dependent variable, constructing a regression analysis model using a machine learning classification algorithm; performing second-order differentiation on the regression model to obtain the point where the second derivative is zero, which is the threshold; and predicting the occurrence of postoperative abnormalities based on the threshold. This invention calculates the corresponding threshold using mathematical means, providing reliable mathematical support and theoretical basis for the origin of the threshold. Furthermore, it allows for automated calculation of the corresponding threshold by designing computer programs, ensuring the uniqueness of the results for each calculation and eliminating interference from human factors.