Koninklijke Philips Corporation N.v.

International Business Machines Corporation

Dimitrios Mavroeidis

Jacek Lukasz Kustra

Bart Jacob Bakker

Sergio Consoli

Monique Hendriks

Pieter Christiaan Vos

Michael Wurst

Pascal Pompey

Jakub Marecek

Stojan Trajanovski

Erik Bresch

Ralf Hoffmann

Teun Van Den Heuvel

Binyam Gebrekidan Gebre

Johan Janssen

Aris Gkoulalas-Divanis

Bryan Conroy

Michele Berlingerio

Caroline Denise Francoise Raynaud

Aki Sakari Härmä

Asif Rahman

Richard Vdovjak

Ulf Grossekathöfer

Aart Tijmen Van Halteren

Arlette Van Wissen

Vlado Menkovski

Ulf Grossekathoefer

Ralf Dieter Hoffmann

Xiaowen Dong

Axel Saalbach

Matthias Lenga

The invention provides a method for evaluating a set of input data, the input data comprising at least one of: clinical data of a subject; genomic data of a subject; clinical data of a plurality of subjects; and genomic data of a plurality of subjects, using a deep learning algorithm. The method includes obtaining a set of input data, wherein the set of input data comprises raw data arranged into a plurality of data clusters and tuning the deep learning algorithm based on the plurality of data clusters. The deep learning algorithm comprises: an input layer; an output layer; and a plurality of hidden layers. The method further includes performing statistical clustering on the raw data using the deep learning algorithm, thereby generating statistical clusters and obtaining a marker from each statistical cluster. Finally, the set of input data is evaluated based on the markers to derive data of medical relevance in respect of the subject or subjects.

Evaluating input data using a deep learning algorithm

The invention relates a computer-implemented method () of generating explainability information for explaining a model output of a trained model. The method uses one or more aspect recognition models configured to indicate a presence of respective characteristics in the input instance. A saliency method is applied to obtain a masked source representation of the input instance at a source layer of the trained model (e.g., the input layer or an internal layer), comprising those elements at the source layer relevant to the model output. The masked source representation is mapped to a target layer (e.g., input or internal layer) of an aspect recognition model, and the aspect recognition model is then applied to obtain a model output indicating a presence of the given characteristic relevant to the model output of the trained model. As explainability information, the characteristics indicated by the aspect recognition models are output.

Explaining a model output of a trained model

According to an embodiment of an aspect, there is provided a computer-implemented method for determining a sleep state of a user. The method comprising receiving (S) a physiological signal from a physiological signal detector used by the user. The method further comprising determining (S), based on the received physiological signal, the sleep state of the user. The method further comprising calculating (S) a reliability value associated with the determination. The reliability value being calculated based on a comparison of the received physiological signal with historic physiological signals of the same sleep state as the determined sleep state. There is further provided a device () and computer-readable medium (). In accordance with the present disclosure, the sleep state of a user may be determined with greater accuracy when compared with past methods.

Determining a sleep state of a user

The present invention provides a method, computer program and processing system for identifies boundaries of lesions within image data. The image data is processed using a machine learning algorithm to generate probability data and uncertainty data. The probability data provides, for each image data point of the image data, a probability data points indicating a probability that said image data point is part of a lesion. The uncertainty data provides, for each probability data point, an uncertainty data point indicating an uncertainty of the said probability data point. The uncertainty data is processed to identify or correct boundaries of the lesions.

Identifying boundaries of lesions within image data

Aspects and embodiments relate to a method of providing a representation of a feature identified by a deep neural network as being relevant to an outcome, a computer program product and apparatus configured to perform that method. The method comprises: providing the deep neural network with a training library comprising: a plurality of samples associated with the outcome; using the deep neural network to recognise a feature in the plurality of samples associated with the outcome; creating a feature recognition library from an input library by identifying one or more elements in each of a plurality of samples in the input library which trigger recognition of the feature by the deep neural network; using the feature recognition library to synthesise a plurality of one or more elements of a sample which have characteristics which trigger recognition of the feature by the deep neural network; and using the synthesised plurality of one or more elements to provide a representation of the feature identified by the deep neural network in the plurality of samples associated with the outcome. Accordingly, rather than visualising a single instance of one or more elements in a sample which trigger a feature associated with an outcome, it is possible to visualise a range of samples including elements which would trigger a feature associated with an outcome, thus enabling a more comprehensive view of operation of a deep neural network in relation to a particular feature.

Deep neural network visualisation

Described herein are systems and methods for training first and second neural network models. A system comprises a memory comprising instruction data representing a set of instructions and a processor configured to communicate with the memory and to execute the set of instructions. The set of instructions, when executed by the processor, cause the processor to set a weight in the second model based on a corresponding weight in the first model, train the second model on a first dataset, wherein the training comprises updating the weight in the second model and adjust the corresponding weight in the first model based on the updated weight in the second model.

Training first and second neural network models

A method of clustering or grouping subjects that are similar to one another. A dataset contains, for each subject, a set of quantitative values which each represent a respective clinical or pathological feature of that subject. A principal component analysis, PCA, is performed on the dataset. Loadings of one of the first two principal components identified by the PCA are used to generate a respective dataset of weighting values. These weighting values are used to weigh or modify each set of quantitative values in the dataset. A clustering algorithm is performed on the weighted sets of subject data. The process may be iterated until user-defined stopping conditions are satisfied.

Subject clustering method and apparatus

A concept for training a neural network model. The concept comprises receiving training data and test data, each comprising a set of annotated images. A neural network model is trained using the training data with an initial regularization parameter. Loss functions of the neural network for both the training data and the test data are used to modify the regularization parameter, and the neural network model is retrained using the modified regularization parameter. This process is iteratively repeated until the loss functions both converge. A system, method and a computer program product embodying this concept are disclosed.

Training a neural network model

A method, system and computer-program product for identifying neural network inputs for a neural network that may have been incorrectly processed by the neural network. A set of activation values (of a subset of neurons of a single layer) associated with a neural network input is obtained. A neural network output associated with the neural network input is also obtained. A determination is made as to whether a first and second neural network input share similar sets of activation values, but dissimilar neural network outputs or vice versa. In this way a prediction can be made as to whether one of the first and second neural network inputs has been incorrectly processed by the neural network.

Using a neural network

There is provided an apparatus and a method of operating the apparatus for providing feedback to a participant directing a communication to one or more other participants. The apparatus () comprises a processor () configured to acquire, from one or more physiological characteristic sensors (), one or more physiological characteristic signals from at least one participant to which the communication is directed as the communication is received by the at least one participant. The processor () is also configured to determine a measure of the quality of the communication based on a comparison of the one or more physiological characteristic signals acquired from the at least one participant with one or more expected physiological characteristic signals and control a user interface () to provide feedback of the determined quality measure of the communication to the participant directing the communication to the at least one participant.

Apparatus and method for providing feedback to a participant of a communication

Presented are concepts for feature identification in medical imaging of a subject. One such concept processes a medical image with a Bayesian deep learning network to determine a first image feature of interest and an associated uncertainty value, the first image feature being located in a first sub-region of the image. It also processes the medical image with a generative adversarial network to determine a second image feature of interest within the first sub-region of the image and an associated uncertainty value. Based on the first and second image features and their associated uncertainty values, the first sub-region of the image is classified.

Feature identification in medical imaging

The cost of data-mining is estimated where data-mining services are delivered via a distributed computing system environment. System requirements are estimated for a particular data-mining task for an input data set having specified properties. Estimating system requirements includes applying a partial learning tool to operate on sample data from the input data set.

Growing community of inventors

Utrecht, Netherlands

Dimitrios Mavroeidis