Abstract: A method for training a deep learning model of a patterning process. The method includes obtaining (i) training data comprising an input image of at least a part of a substrate having a plurality of features and a truth image, (ii) a set of classes, each class corresponding to a feature of the plurality of features of the substrate within the input image, and (iii) a deep learning model configured to receive the training data and the set of classes, generating a predicted image, by modeling and/or simulation of the deep learning model using the input image, assigning a class of the set of classes to a feature within the predicted image based on matching of the feature with a corresponding feature within the truth image, and generating, by modeling and/or simulation, a trained deep learning model by iteratively assigning weights using a loss function.

Abstract: In a lithographic process, product units such as semiconductor wafers are subjected to lithographic patterning operations and chemical and physical processing operations. Alignment data or other measurements are made at stages during the performance of the process to obtain object data representing positional deviation or other parameters measured at points spatially distributed across each unit. This object data is used to obtain diagnostic information by performing a multivariate analysis to decompose a set of vectors representing the units in the multidimensional space into one or more component vectors. Diagnostic information about the industrial process is extracted using the component vectors. The performance of the industrial process for subsequent product units can be controlled based on the extracted diagnostic information.

Abstract: A method of determining a measurement sequence for an inspection tool inspecting a structure generated by a lithographic process performed by a lithographic system is presented, the method including deriving a model for the lithographic process as performed by the lithographic system, the model including a relationship between a set of system variables describing the lithographic system and an output variable representing the structure resulting of the lithographic process, determining an observability of one or more system variables in the output variable, and determining the measurement sequence for the inspection tool, based on the observability.

Abstract: In a lithographic process, product units such as semiconductor wafers are subjected to lithographic patterning operations and chemical and physical processing operations. Alignment data or other measurements are made at stages during the performance of the process to obtain object data representing positional deviation or other parameters measured at points spatially distributed across each unit. This object data is used to obtain diagnostic information by performing a multivariate analysis to decompose a set of vectors representing the units in the multidimensional space into one or more component vectors. Diagnostic information about the industrial process is extracted using the component vectors. The performance of the industrial process for subsequent product units can be controlled based on the extracted diagnostic information.

Abstract: A method for training a deep learning model of a patterning process. The method includes obtaining (i) training data including an input image of at least a part of a substrate having a plurality of features and including a truth image, (ii) a set of classes, each class corresponding to a feature of the plurality of features of the substrate within the input image, and (iii) a deep learning model configured to receive the training data and the set of classes, generating a predicted image, by modeling and/or simulation with the deep learning model using the input image, assigning a class of the set of classes to a feature within the predicted image based on matching of the feature with a corresponding feature within the truth image, and generating, by modeling and/or simulation, a trained deep learning model by iteratively assigning weights using a loss function.

Abstract: A method for training a machine learning model configured to predict a substrate image corresponding to a printed pattern of a substrate as measured via a metrology tool. The method involves obtaining a training data set including (i) metrology data of the metrology tool used to measure the printed pattern of the substrate, and (ii) a representation of a mask pattern employed for imaging the printed pattern on the substrate; and training, based on the training data set, a machine learning model to predict the substrate image of the substrate as measured by the metrology tool such that a cost function is improved, wherein the cost function includes a relationship between the predicted substrate image and the metrology data.

Abstract: Disclosed are methods and systems for processing medical image data. The method comprising inputting, with one or more processors of one or more computation devices, medical image data into an encoder stage of an encoder-decoder pair (EDP) as a first input among one or more inputs; calculating, with the one or more processors, a latent space representation of the one or more inputs using the encoder stage of the EDP; providing, from a latent space database stored within one or more storage devices accessible by the one or more computation devices, latent space representations of other inputs; and determining, with the one or more processors, a classification based on the latent space representation of the one or more inputs and at least one latent space representation of the other inputs.

Abstract: A method and apparatus of detection, registration and quantification of an image is described. The method may include obtaining an image of a lithographically created structure, and applying a level set method to an object, representing the structure, of the image to create a mathematical representation of the structure. The method may include obtaining a first dataset representative of a reference image object of a structure at a nominal condition of a parameter, and obtaining second dataset representative of a template image object of the structure at a non-nominal condition of the parameter. The method may further include obtaining a deformation field representative of changes between the first dataset and the second dataset. The deformation field may be generated by transforming the second dataset to project the template image object onto the reference image object. A dependence relationship between the deformation field and change in the parameter may be obtained.

Abstract: Methods of measuring variation across multiple instances of a pattern on a substrate or substrates after a step in a device manufacturing process are disclosed. In one arrangement, data representing a set of images is received. Each image represents a different instance of the pattern. The set of images are registered relative to each other to superimpose the instances of the pattern. Variation in the pattern is measured using the registered set of images. The pattern comprises a plurality of pattern elements and the registration comprises applying different weightings to two or more of the plurality of pattern elements. The weightings control the extent to which each pattern element contributes to the registration of the set of images. Each weighting is based on an expected variation of the pattern element to which the weighting is applied.

Abstract: A defect prediction method for a device manufacturing process involving production substrates processed by a lithographic apparatus, the method including training a classification model using a training set including measured or determined values of a process parameter associated with the production substrates processed by the device manufacturing process and an indication regarding existence of defects associated with the production substrates processed in the device manufacturing process under the values of the process parameter, and producing an output from the classification model that indicates a prediction of a defect for a substrate.

Abstract: A method for training a deep learning model of a patterning process. The method includes obtaining (i) training data including an input image of at least a part of a substrate having a plurality of features and including a truth image, (ii) a set of classes, each class corresponding to a feature of the plurality of features of the substrate within the input image, and (iii) a deep learning model configured to receive the training data and the set of classes, generating a predicted image, by modeling and/or simulation with the deep learning model using the input image, assigning a class of the set of classes to a feature within the predicted image based on matching of the feature with a corresponding feature within the truth image, and generating, by modeling and/or simulation, a trained deep learning model by iteratively assigning weights using a loss function.

Abstract: A method and apparatus of detection, registration and quantification of an image is described. The method may include obtaining an image of a lithographically created structure, and applying a level set method to an object, representing the structure, of the image to create a mathematical representation of the structure. The method may include obtaining a first dataset representative of a reference image object of a structure at a nominal condition of a parameter, and obtaining second dataset representative of a template image object of the structure at a non-nominal condition of the parameter. The method may further include obtaining a deformation field representative of changes between the first dataset and the second dataset. The deformation field may be generated by transforming the second dataset to project the template image object onto the reference image object. A dependence relationship between the deformation field and change in the parameter may be obtained.

Abstract: Methods of measuring variation across multiple instances of a pattern on a substrate or substrates after a step in a device manufacturing process are disclosed. In one arrangement, data representing a set of images is received. Each image represents a different instance of the pattern, wherein the pattern includes a plurality of pattern elements. The set of images are registered relative to each other to superimpose the instances of the pattern. The registration includes applying different weightings to two or more of the plurality of pattern elements, wherein the weightings control the extent to which each pattern element contributes to the registration of the set of images and each weighting is based on an expected variation of the pattern element to which the weighting is applied. Variation in the pattern is measured using the registered set of images.

Abstract: Described herein is a method for quantifying uncertainty in parameterized (e.g., machine learning) model predictions. The method comprises causing a parameterized model to predict multiple posterior distributions from the parameterized model for a given input. The multiple posterior distributions comprise a distribution of distributions. The method comprises determining a variability of the predicted multiple posterior distributions for the given input by sampling from the distribution of distributions; and using the determined variability in the predicted multiple posterior distributions to quantify uncertainty in the parameterized model predictions. The parameterized model comprises encoder-decoder architecture.

Abstract: A defect prediction method for a device manufacturing process involving production substrates processed by a lithographic apparatus, the method including training a classification model using a training set including measured or determined values of a process parameter associated with the production substrates processed by the device manufacturing process and an indication regarding existence of defects associated with the production substrates processed in the device manufacturing process under the values of the process parameter, and producing an output from the classification model that indicates a prediction of a defect for a substrate.

Abstract: A method of determining a measurement sequence for an inspection tool inspecting a structure generated by a lithographic process performed by a lithographic system is presented, the method including deriving a model for the lithographic process as performed by the lithographic system, the model including a relationship between a set of system variables describing the lithographic system and an output variable representing the structure resulting of the lithographic process, determining an observability of one or more system variables in the output variable, and determining the measurement sequence for the inspection tool, based on the observability.

Abstract: A method including: obtaining a logistic mathematical model predicting the formation of a physical structure created using a patterning process; evaluating the logistic mathematical model to predict formation of a part of the physical structure and generate an output; and adapting, based on the output, an aspect of the patterning process.

Abstract: A method for determining the existence of a defect in a printed pattern may include obtaining a) a captured image of a printed pattern from an image capture device, and b) a simulated image of the printed pattern generated by a process model. The method may include generating a combined image as a weighted combination of portions of the captured image and the simulated image. The method may include determining whether a defect exists in the printed pattern based on the combined image.

Abstract: A method and a computer program product that relates to lithographic apparatuses and, processes, and more particularly to a method and computer program to inspect substrates produced by the lithographic apparatuses and processes. The method and/or computer program product includes determining contributions from independent sources from results measured from a lithography process or a substrate processed by the lithography process, wherein the results are measured using a plurality of different substrate measurement recipes.

Abstract: A method including selecting a shaped feature from a set of shaped features, each shaped feature of the set of shaped features having a set of points on a perimeter of the shape of the shaped feature, creating a plurality of shape context descriptors for the selected shaped feature, wherein each shape context descriptor provides an indication of a location in a shape context descriptor framework of a first focus point of the set of points in relation to a second point of the set of points, and identifying a shaped feature from the set of shaped features having a same or similar shape as the selected shaped feature based on data from the plurality of shape context descriptors.