11413 results about "Method of undetermined coefficients" patented technology

The disclosure includes pulse oximetry systems and methods for determining point-by-point saturation values by encoding photoplethysmographs in the complex domain and processing the complex signals. The systems filter motion artifacts and other noise using a variety of techniques, including statistical analysis such as correlation, or phase filtering.

The present invention involves method and apparatus for analyzing two measured signals that are modeled as containing primary and secondary portions. Coefficients relate the two signals according to a model defined in accordance with the present invention. In one embodiment, the present invention involves utilizing a transformation which evaluates a plurality of possible signal coefficients find appropriate coefficients. Alternatively, the present invention involves using statistical functions or Fourier transform and windowing techniques to determine the coefficients relating to two measured signals. Use of this invention is described in particular detail with respect to blood oximetry measurements.

A computerized annotation method achieves real-time operation and better optimization properties while preserving the architectural advantages of the generative modeling approach. A novel clustering algorithm for objects is represented by discrete distributions, or bags of weighted vectors, thereby minimizing the total within cluster distance, a criterion used by the k-means algorithm. A new mixture modeling method, the hypothetical local mapping (HLM) method, is used to efficiently build a probability measure on the space of discrete distributions. Thus, in accord with the invention every image is characterized by a statistical distribution. The profiling model specifies a probability law for distributions directly.

The present invention is an improved computational method and system of empirical induction that can be used to arrive at generalized conclusions and make predictions involving longitudinal associations between and among variables and events. Empirical induction is used to gain scientific knowledge, to develop and evaluate treatments and other interventions, and to help make predictions and decisions. The invention, which is distinct from and often complementary to the statistical method, is applied to repeated measures and multiple time-series data and can be used to quantify, discover, analyze, and describe longitudinal associations for individual real and conceptual entities. Major improvements include provisions to define Boolean independent events and Boolean dependent events and to apply analysis parameters such as episode length and episode criterion for both independent and dependent variables, persistence after independent events, and delay and persistence after Boolean independent events. These improvements are in addition to levels of independent and dependent variables, delay after independent events, and provision to quantify benefit and harm across two or more dependent variables. Additional improvements include provisions to quantify longitudinal associations as functions of period or time and to compute values of predictive indices when there are two or more independent variables. Major applications and uses of the invention include data mining, the conduct of clinical trials of treatments for the management or control of chronic disorders, health-effect monitoring, the quantification and analysis internal control in adaptive systems, analyses of serial functional images, analyses of behavior and behavior modification, and use to create computerized devices and systems whose behavior can be modified by experience. The present invention is best implemented on the Internet.

The present invention is directed toward a pulse oximetry system for the determination of a physiological parameter capable of removing motion artifacts from physiological signals comprises a hardware subsystem and a software subsystem. The software subsystem is used in conjunction with the hardware subsystem to perform a method for removing a plurality of motion artifacts from the photo-plethysmographic data and for obtaining a measure of at least one physiological parameter from the data. The method comprises acquiring the raw photo-plethysmographic data, transforming the data into the frequency domain, analyzing the transformed data to locate a series of candidate cardiac spectral peaks (primary plus harmonics), reconstructing a photo-plethysmographic signal in the time domain with only the candidate cardiac spectral peaks (primary plus harmonics), computing the second order derivative of the reconstructed photo-plethysmographic signal, analyzing the candidate second order derivative photo-plethysmographic signal to determine the absence or presence of cardiac physiologic signal characteristics, and finally selecting the best physiologic candidate from the series of potential cardiac spectral peaks (primary plus harmonics) based upon a second derivative scoring system. This scoring system is preferentially based upon second derivative processing analysis, but can be equally applied using the first, third, fourth or other similar derivative processing analysis.

Time-varying spatial signals are detected by accelerometers mounted within the patient. The signals, representative of the actual 3-D trajectory of the patient, are compared with information representative of expected trajectories retrieved from memory to identify a current patient posture, which may be either a dynamic posture such as walking or running or a change in posture such as rising from a seated position to a standing position. In this manner, a change in posture of the patient is identified based upon a full 3-D trajectory, rather than merely the orientation of the patient at the beginning and the end of the change in posture. In an example described herein, the implantable device stores information representative of expected 3-D trajectories in the form of pre-calculated comparison matrices derived from orthonormal kernels employing Laguerre functions or Lagrange functions. A technique is also described for use by an external programmer for pre-calculating comparison matrices so as to reduce the processing burden within the implanted device during posture detection.

The invention provides a method for analyzing spatial variations in field and crop performance. The method allows analysis of the field and crop performance as a function of various treatment and / or environmental conditions. This method is an advance over the prior art in that it allows better defined, more accurate comparisons of adjacent test areas, eliminating the need for extrapolation of data and avoiding inaccuracies therefrom. Also, the more accurate comparisons allow better characterization of effects due to the environment, such as those caused by genotype by environment (GxE) interaction. The more accurately defined comparison units also lead to improved analysis when comparing to other data layers for non-spatial analysis. The invention also provides a system for carrying out this method. The system preferably includes a device for inputting spatially-referenced data, a memory, an output device, and a processor. The output data may be represented spatially or non-spatially. The information provided by the method and system of the invention will allow farmers to make more informed decisions regarding seed and farming inputs and practices, thereby maximizing agricultural productivity.

System comprised of a medical device and method for analyzing physiological and health data and representing the most significant parameters at different levels of detail which are understandable to a lay person and a medical professional. Low, intermediate and high-resolution scales can exchange information between each other for improving the analyses; the scales can be defined according to the corresponding software and hardware resources. A low-resolution Scale I represents a small number of primary elements such as intervals between the heart beats, duration of electrocardiographic PQ, QRS, and QT-intervals, amplitudes of P-, Q-, R-, S-, and T-waves. This real-time analysis is implemented in a portable device that requires minimum computational resources. The set of primary elements and their search criteria can be adjusted using intermediate or high-resolution levels. At the intermediate-resolution Scale II, serial changes in each of the said elements can be determined using a mathematical decomposition into series of basis functions and their coefficients. This scale can be implemented using a specialized processor or a computer organizer. At the high-resolution Scale III, combined serial changes in all primary elements can be determined to provide complete information about the dynamics of the signal. This scale can be implemented using a powerful processor, a network of computers or the Internet. The system can be used for personal or group self-evaluation, emergency or routine ECG analysis, or continuous event, stress-test or bed-side monitoring.

A system and method for organization and analysis of complex and dynamically interactive time series is disclosed. One example comprises a processor based system for relational analysis of physiologic signals for providing early recognition of catastrophic and pathologic events such as pathophysiologic divergence. The processor is programmed to identify pathophysiologic divergence of at least one of first and second physiologic parameters in relationship to the other and to output an indication of the divergence. An object-based method of iterative relational processing waveform fragments in the time domain is described wherein each more complex waveform object inherits the characteristics of the waveform objects from which it is derived. The first physiologic parameter can be the amplitude and frequency of the variation in chest wall impedance or nasal pressure and the second parameter can be a measure or indication of the arterial oxygen saturation.

Aspects of the invention relate generally to accessing, storing, and processing vector data to represent various geographical features such as roads, rivers, lakes, countries, continents, and oceans on one or more maps. More specifically, the vector data may be pre-simplified for rendering at different zoom levels. The simplification process is based on removing vertices from vector data in order to reduce the number of points in a given polygon or line. As this process is very expensive in terms of time and processing power, the system and method allow for estimation of the proportion of vertices which that would be removed from the original geometry. Based on this estimation, one may decide whether or not the simplification is worth the effort to compute and store the simplified data.

A system and method for calculating an engagement value by quantifying an amount that a user is acting without thinking considering brainwaves and a heart rate can be used to compare media based on an individual or a group of individuals. Events of the media can be contrasted and compared by the engagement value as well. Statistical measurements may be taken to improve media.

A method of producing noise-free images is disclosed. The method includes using machine learning incorporating a filter to output filter parameters using the training images. The machine learning may include training a neural network. The filter parameters are applied to Monte Carlo rendered training images that have noise to generate noise-free images. The training may include determining, computing and extracting features of the training images; computing filter parameters; applying an error metric; and applying backpropgation. The neural network may be a multilayer perceptron. The machine learning model is applied to new noisy Monte Carlo rendered images to create noise-free images. This may include applying the filter to the noisy Monte Carlo rendered images using the filter parameters to create the noise-free images.

Diffraction models and scatterometry are used to reconstruct a model of a microscopic structure on a substrate. A plurality of candidate structures are defined, each represented by a plurality of parameters (p1, p2, etc.)). A plurality of model diffraction signals are calculated by simulating illumination of each of the candidate structures. The structure is reconstructed by fitting one or more of the model diffraction signals to a signal detected from the structure. In the generation of the candidate structures, a model recipe is used in which parameters are designated as either fixed or variable. Among the variable parameters, certain parameters are constrained to vary together in accordance with certain constraints, such as linear constraints. An optimized set of constraints, and therefore an optimized model recipe, is determined by reference to a user input designating one or more parameters of interest for a measurement, and by simulating the reconstruction process reconstruction. The optimized model recipe can be determined automatically by a parameter advisor process that simulates reconstruction of a set of reference structures, using a plurality of candidate model recipes. In the generation of the reference structures, restrictions can be applied to exclude unrealistic parameter combinations.

Disclosed herein are methods and systems for creating a complex model of the human anatomy. The anatomy may be modeled using multiple geometries. Generally, a plurality of line-of-sight models may be combined into a composite model. Multiple clouds of points may be used to create surface models which may then be merged into a common volume. The resulting composite model may include portions that are not within a line of sight of a mean center point of the composite model. The surface models may be modeled using polygons, including for example, triangles. Disclosed herein are also ways in which electrophysiology data and / or other information may be mapped from a measurement point to a point on the composite model.

Time-varying spatial signals are detected by accelerometers mounted within the patient. The signals, representative of the actual 3-D trajectory of the patient, are compared with information representative of expected trajectories retrieved from memory to identify a current patient posture, which may be either a dynamic posture such as walking or running or a change in posture such as rising from a seated position to a standing position. In this manner, a change in posture of the patient is identified based upon a full 3-D trajectory, rather than merely the orientation of the patient at the beginning and the end of the change in posture. In an example described herein, the implantable device stores information representative of expected 3-D trajectories in the form of pre-calculated comparison matrices derived from orthonormal kernels employing Laguerre functions or Lagrange functions. A technique is also described for use by an external programmer for pre-calculating comparison matrices so as to reduce the processing burden within the implanted device during posture detection.

A method and apparatus to monitor the neurologic state of a patient undergoing general anesthesia is provided. Previous automated systems to monitor the neurologic state of a patient undergoing general anesthesia involve a significant time delay between the patient's true hypnotic state and the computed indices. The present invention reduces this time delay by using a different analysis technique applied to spontaneous EEG. A wavelet decomposition and statistical analysis of the observed EEG is conducted and compared to reference data to provide a numerical indicator. In addition, this indicator is more consistent with the patient's loss of consciousness indicated by the loss of count event than previous systems.

A method for removing motion artifacts from devices for sensing bodily parameters and apparatus and system for effecting same that includes analyzing segments of measured data representing bodily parameters and possibly noise from motion artifacts. Each data segment is frequency analyzed to determine up to three candidate peaks for further analysis. Up to three candidate frequencies may be filtered and various parameters associated with each candidate frequency are calculated. A pulse-estimate input may also be accepted from an external source. The best frequency, if one exists, is determined by arbitrating the candidate frequencies and the pulse-estimate input using the calculated parameters according to predefined criteria. If a best frequency is found, a pulse rate and SpO2 may be output. If a best frequency is not found, other, conventional techniques for calculating pulse rate and SpO2 may be used.

The present invention provides a rolling bearing fault diagnosis method in various working conditions based on feature transfer learning, and relates to the field of fault diagnosis. The objective ofthe invention is to solve the problem that a rolling bearing, especially to various working conditions, is low in accuracy of diagnosis. The method comprise the steps of: employing a VMD (VariationalMode Decomposition) to perform decomposition of vibration signals of a rolling bearing in each state to obtain a series of intrinsic mode functions, performing singular value decomposition of a matrixformed by the intrinsic mode functions to solve a singular value or a singular value entropy, combining time domain features and frequency domain features of the vibration signals to construct a multi-feature set; introducing a semisupervised transfer component analysis method to perform multinuclear construction of a kernel function thereof, sample features of different working conditions are commonly mapped to a shared reproducing kernel Hilbert space so as to improve the data intra-class compactness and the inter-class differentiation; and employing the maximum mean discrepancy embedding to select more efficient data as a source domain, inputting source domain feature samples into a SVM (Support Vector Machine) for training, and testing target domain feature samples after mapping. Therolling bearing fault diagnosis method in various working conditions has higher accuracy in the rolling bearing multi-state classification in various working conditions.

Apparatus, methods, and other embodiments associated with NMR fingerprinting are described. One example NMR apparatus includes an NMR logic configured to repetitively and variably sample a (k, t, E) space associated with an object to acquire a set of NMR signals. Members of the set of NMR signals are associated with different points in the (k, t, E) space. Sampling is performed with t and / or E varying in a non-constant way. The varying parameters may include flip angle, echo time, RF amplitude, and other parameters. The NMR apparatus may also include a signal logic configured to produce an NMR signal evolution from the NMR signals, a matching logic configured to compare a signal evolution to a known, simulated or predicted signal evolution, and a characterization logic configured to characterize a resonant species in the object as a result of the signal evolution comparisons.

The invention provides in one aspect methods and apparatus for use with C-arm and other CT systems, e.g., with non-rigid geometries. In such systems, by way of example, calibration can be performed to determine the exact position of the x-ray source and the exact orientation of the detector where each projection measurement is made. Next, a weighting coefficient can be determined for the voxels in each plane of a reconstruction volume at every possible projection. Finally, the order in which to process the voxels during image reconstruction can be determined. Following an actual CT scan procedure in which scans are obtained of a volume to be constructed, a system according to these and related aspects of the invention can use an optimal, pre-calculated processing method, while utilizing offsets and weighting coefficients determined during calibration, for performing backprojection image reconstruction.

The invention discloses a target detection method based on a dense connection characteristic pyramid network, and relates to the image processing and computer vision technology. The method comprises the steps of collecting an image data set labeled with a target bounding box and category information; constructing a dense connection feature pyramid network containing a Squeeze-Exciton structure capable of learning a dependency relationship between feature channels as a feature extraction backbone network; alternately training the RPN subnet and the R-FCN subnet to obtain a target detection model; and detecting a specific target in the image by using the model. The squeeze-Exciton structure and a dense connection structure are introduced into a feature extraction trunk network, characterization capability of the model is enhanced, adaptability of the model to targets of different sizes is enhanced through the feature pyramid structure, calculation sharing of the whole network model is achieved to the maximum degree through the R-FCN detection head, calculation resources are saved, and performance of the whole target detection model is improved. .

This patent concerns novel technology for detection of zero-day data classes for domains with high-dimensional mixed continuous / discrete feature spaces, including Internet traffic. Assume there is a known-class database available for learning a null hypothesis that a given new batch of unlabeled data does not contain any data from unknown / anomalous classes. A novel and effective generalization of previous parsimonious mixture and topic modeling methods is developed. The novel unsupervised anomaly detector (AD) acts on a new unlabeled batch of data to either identify the statistically significant anomalous classes latently present therein or reject the alternative hypothesis that the new batch contains any anomalous classes. The present AD invention can be applied in an on-line setting. Labeling (by a human expert or by other means) of anomalous clusters provides new supervised data that can be used to adapt an actively learned classifier whose objective is to discriminate all the classes.

The invention relates to an extraction method for early failure sensitive characteristics based on ensemble empirical mode decomposition (EEMD) and wavelet packet transform. The extraction method for the early failure sensitive characteristics based on the EEMD and the wavelet packet transform includes the following steps: (1), collected original vibration signals of mechanical and electrical equipment are decomposed according to the EEMD, white noise is added, and intrinsic mode function (IMF) components are obtained through decomposition; (2), the sensitive IMF components closely related to failure are chosen, and other irrelative IMF components are ignored; (3), the sensitive IMF components chosen through step (2) are decomposed in an orthogonal wavelet packet mode, and a wavelet coefficient of each node is obtained; and (4), envelopes are extracted from the obtained wavelet packet coefficients by adoption of the Hilbert transform and the Fourier transform, power spectrums are calculated, the power spectrum corresponding to each wavelet packet coefficient is obtained and serves as the early failure sensitive characteristic , and the sensitive characteristics are automatically obtained. Self-adapting signals can be decomposed, the sensitive characteristics can be convenient to obtain automatically, diagnosis precision and speed are improved, and a mechanical and electrical system can be diagnosed quickly, accurately and stably. The extraction method for the early failure sensitive characteristics based on the EEMD and the wavelet packet transform can be applied to the field of mechanical and electrical equipment failure diagnosis.

The present invention relates to a method and apparatus to objectively measure stress and fatigue using measurement tools, record stress and fatigue related data in a standardized database, create automated prompts and alerts based upon pre-defined stress and fatigue thresholds (which are derived based upon individual end-user and task performance), provide a number of interventions (which can be preferentially selected by the individual end-user), create data-driven best practice guidelines though meta-analysis of the database, and provide an objective tool for comparative technology assessment.

The invention belongs to the technical field of cross-media correlation learning, and particularly relates to a cross-modal subject correction modeling method based on deep learning.The method includes two main algorithms of multi-modal file expression based on deep vocabularies and correlation subject model modeling fusing cross-modal subjection correction learning.A deep learning technology is utilized for constructing deep semantic vocabularies and deep vision vocabularies to describe a semantic description part and an image part in a multi-modal file.Based on multi-modal file expression, a cross-modal correlation subject model is constructed to model a whole multi-modal file set, so that the generation process of the multi-modal file and the correlation between different modals are described.The accuracy is high, and adaptability is high.The cross-modal subject correction modeling method has important meaning for efficient cross-media information retrieval in consideration of multi-modal semantic information on the basis of the large-scale multi-modal file (a text and an image), can improve retrieval correlation and promote user experience, and has great application value in the field of cross-media information retrieval.

In one embodiment, a method is provided for classifying cognitive activity in an individual. In the method, a candidate time interval is identified from a first type of physiological data within which cognitive processing is expected to occur for the individual. In addition, a second type of physiological data is obtained that comprises data representative of a cognitive state of the individual. Further, the data representative of a cognitive state of the individual is extracted from the second type of physiological data based on the identified candidate time interval.