Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for designing proteins. In one aspect, a method comprises: generating noisy molecular structure data sampled from a noise distribution that defines, for each position in an amino acid sequence of the protein, a corresponding initial spatial position for each atom in a predefined set of possible atoms; and processing the noisy molecular structure data using a diffusion model that comprises a denoising neural network to generate denoised molecular structure data that defines a denoised version of the noisy molecular structure data.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for designing a protein to satisfy one or more design criteria. In one aspect, a method comprises, at each of a plurality of iterations: modifying a current amino acid sequence to generate an alternative amino acid sequence; providing an input comprising a representation of the alternative amino acid sequence to a structure prediction neural network that is configured to process the input to generate an output characterizing a predicted structure of a protein specified by the alternative amino acid sequence; determining, based on the predicted structure of the protein specified by the alternative amino acid sequence, a design score characterizing how well the protein specified by the alternative amino acid sequence satisfies the design criteria; and determining, based on the design score, whether to update the current amino acid sequence to the alternative amino acid sequence.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for processing input sequences using a neural network that uses a partial position encoding scheme. The neural network generally includes both global and local attention layers. In the partial position encoding scheme, while the local attention layers do use position encoding, (i) a subset of the global attention layers can apply an attention mechanism that does not use position encoding, or (ii) the subset of global attention layers can apply an attention mechanism that does not apply position encoding to one or more of the dimensions of the input to the attention mechanism.

Abstract: Much of the image processing that is applied to medical images is a form of “inverse problem”. This is a class of mathematical problems in which a “forward” model by which a signal is converted into dataset is known, to at least some degree, but where the aim is to reconstruct the signal given the resulting dataset. Thus, an inverse problem is essentially seeking to discover x given knowledge of A(x)+noise by finding an appropriate reconstruction operator A† such that A† (A(x)+noise)?x, thereby enabling us to obtain x (or a close approximation) given knowledge of an output dataset consisting of A(x)+noise. Generally, several such processes (or their equivalents) are applied to the image dataset.

Abstract: Techniques for solving a radiotherapy treatment plan optimization problem are provided. The techniques include receiving a radiotherapy treatment plan optimization problem; processing the radiotherapy treatment plan optimization problem with a machine learning model to estimate one or more optimization variables of the radiotherapy treatment plan optimization problem, wherein the machine learning model is trained to establish a relationship between the one or more optimization variables and parameters of a plurality of training radiotherapy treatment plan optimization problems; and generating a solution to the radiotherapy treatment plan optimization problem based on the estimated one or more optimization variables of the radiotherapy treatment plan optimization problem.

Abstract: Methods, systems and apparatus, including computer programs encoded on computer storage media. One of the methods includes obtaining a plurality of images of a macromolecule having a plurality of atoms, training a decoder neural network on the plurality of images, and after the training, generating a plurality of conformations for at least a portion of the macromolecule that each include respective three-dimensional coordinates of each of the plurality of atoms, wherein generating each conformation includes sampling a conformation latent representation from a prior distribution over conformation latent representations, processing a respective input including the sampled conformation latent representation using the decoder neural network to generate a conformation output that specifies three-dimensional coordinates of each of the plurality of atoms for the conformation, and generating the conformation from the conformation output.

Abstract: Techniques for generating a radiotherapy treatment plan are provided. The techniques include receiving an input parameter related to a patient, the input parameter being of a given type; processing the input parameter with a machine learning technique to estimate a realizable plan parameter of a radiotherapy treatment plan, wherein the machine learning technique is trained to establish a relationship between the given type of input parameter and a set of realizable radiotherapy treatment plan parameters to achieve a target radiotherapy dose distribution; and generating the radiotherapy treatment plan based on the estimated realizable plan parameter.

Abstract: Image reconstruction can include using a statistical or machine learning, MAP estimator, or other reconstruction technique to produce a reconstructed image from acquired imaging data. A Conditional Generative Adversarial Network (CGAN) technique can be used to train a Generator, using a Discriminator, to generate posterior distribution sampled images that can be displayed or further processed such as to help provide uncertainty information about a mean reconstruction image. Such uncertainty information can be useful to help understand or even visually modify the mean reconstruction image. Similar techniques can be used in a segmentation use-case, instead of a reconstruction use case. The uncertainty information can also be useful for other post-processing techniques.

Abstract: Systems and methods for calculating radiotherapy dose distribution are provided. The systems and methods include operations for receiving data representing at least one of particle trajectories or a dose deposition pattern in a simulated delivery of a radiotherapy plan; applying a dose calculation process to the received data to generate a first radiotherapy dose distribution having a first level of detail; and processing the first radiotherapy dose distribution using a trained machine learning technique to generate a second radiotherapy dose distribution having a second level of detail that enhances the first level of detail.

Abstract: Much of the image processing that is applied to medical images is a form of “inverse problem”. This is a class of mathematical problems in which a “forward” model by which a signal is converted into dataset is known, to at least some degree, but where the aim is to reconstruct the signal given the resulting dataset. Thus, an inverse problem is essentially seeking to discover x given knowledge of A(x)+noise by finding an appropriate reconstruction operator A† such that A† (A(x)+noise)?x, thereby enabling us to obtain x (or a close approximation) given knowledge of an output dataset consisting of A(x)+noise. Generally, several such processes (or their equivalents) are applied to the image dataset.

Abstract: Techniques for solving a radiotherapy treatment plan optimization problem are provided. The techniques include receiving a radiotherapy treatment plan optimization problem; processing the radiotherapy treatment plan optimization problem with a machine learning model to estimate one or more optimization variables of the radiotherapy treatment plan optimization problem, wherein the machine learning model is trained to establish a relationship between the one or more optimization variables and parameters of a plurality of training radiotherapy treatment plan optimization problems; and generating a solution to the radiotherapy treatment plan optimization problem based on the estimated one or more optimization variables of the radiotherapy treatment plan optimization problem.

Abstract: Techniques for generating a radiotherapy treatment plan are provided. The techniques include receiving an input parameter related to a patient, the input parameter being of a given type; processing the input parameter with a machine learning technique to estimate a realizable plan parameter of a radiotherapy treatment plan, wherein the machine learning technique is trained to establish a relationship between the given type of input parameter and a set of realizable radiotherapy treatment plan parameters to achieve a target radiotherapy dose distribution; and generating the radiotherapy treatment plan based on the estimated realizable plan parameter.

Abstract: Image reconstruction can include using a statistical or machine learning, MAP estimator, or other reconstruction technique to produce a reconstructed image from acquired imaging data. A Conditional Generative Adversarial Network (CGAN) technique can be used to train a Generator, using a Discriminator, to generate posterior distribution sampled images that can be displayed or further processed such as to help provide uncertainty information about a mean reconstruction image. Such uncertainty information can be useful to help understand or even visually modify the mean reconstruction image. Similar techniques can be used in a segmentation use-case, instead of a reconstruction use case. The uncertainty information can also be useful for other post-processing techniques.

Abstract: Techniques for the operation and use of a model that learns the general representation of multimodal images is disclosed. In various examples, methods from representation learning are used to find a common basis for representation of medical images. These include aspects of encoding, fusion, and downstream tasks, with use of the general representation and model. In an example, a method for generating a modality-agnostic model includes receiving imaging data, encoding the imaging data by mapping data to a latent representation, fusing the encoded data to conserve latent variables corresponding to the latent representation, and training a model using the latent representation. In an example, a method for processing imaging data using a trained modality-agnostic model includes receiving imaging data, encoding the data to the defined encoding, processing the encoded data with a trained model, and performing imaging processing operations based on output of the trained model.

Abstract: Systems and methods for calculating radiotherapy dose distribution are provided. The systems and methods include operations for receiving data representing at least one of particle trajectories or a dose deposition pattern in a simulated delivery of a radiotherapy plan; applying a dose calculation process to the received data to generate a first radiotherapy dose distribution having a first level of detail; and processing the first radiotherapy dose distribution using a trained machine learning technique to generate a second radiotherapy dose distribution having a second level of detail that enhances the first level of detail.

Abstract: Image reconstruction can include using a statistical or machine learning, MAP estimator, or other reconstruction technique to produce a reconstructed image from acquired imaging data. A Conditional Generative Adversarial Network (CGAN) technique can be used to train a Generator, using a Discriminator, to generate posterior distribution sampled images that can be displayed or further processed such as to help provide uncertainty information about a mean reconstruction image. Such uncertainty information can be useful to help understand or even visually modify the mean reconstruction image. Similar techniques can be used in a segmentation use-case, instead of a reconstruction use case. The uncertainty information can also be useful for other post-processing techniques.

Abstract: Techniques for the operation and use of a model that learns the general representation of multimodal images is disclosed. In various examples, methods from representation learning are used to find a common basis for representation of medical images. These include aspects of encoding, fusion, and downstream tasks, with use of the general representation and model. In an example, a method for generating a modality-agnostic model includes receiving imaging data, encoding the imaging data by mapping data to a latent representation, fusing the encoded data to conserve latent variables corresponding to the latent representation, and training a model using the latent representation. In an example, a method for processing imaging data using a trained modality-agnostic model includes receiving imaging data, encoding the data to the defined encoding, processing the encoded data with a trained model, and performing imaging processing operations based on output of the trained model.

Abstract: Image reconstruction can include using a statistical or machine learning, MAP estimator, or other reconstruction technique to produce a reconstructed image from acquired imaging data. A Conditional Generative Adversarial Network (CGAN) technique can be used to train a Generator, using a Discriminator, to generate posterior distribution sampled images that can be displayed or further processed such as to help provide uncertainty information about a mean reconstruction image. Such uncertainty information can be useful to help understand or even visually modify the mean reconstruction image. Similar techniques can be used in a segmentation use-case, instead of a reconstruction use case. The uncertainty information can also be useful for other post-processing techniques.