Abstract: This disclosure relates to detecting prostate cancer. A processor receives image data indicative of a computed tomography scan comprising a spatial sequence of two-dimensional cross-sectional images. The processor then applies a neural network to the image data by calculating three-dimensional convolutions in one or more convolutional layers of the neural network and evaluating an output layer. A first dimension of the convolutions is defined in space along the spatial sequence of the two-dimensional cross-sectional images. The three-dimensional convolutions comprise convolutions along the first dimension. The output layer is connected to the one or more convolutional layers in the neural network, and has an output value that is indicative of a probability of detecting prostate cancer in the computed tomography scan. Three-dimensional convolutions in the neural network provide a more accurate probability of detecting prostate cancer.

Abstract: This disclosure relates to detecting prostate cancer. A processor receives image data indicative of a computed tomography scan comprising a spatial sequence of two-dimensional cross-sectional images. The processor then applies a neural network to the image data by calculating three-dimensional convolutions in one or more convolutional layers of the neural network and evaluating an output layer. A first dimension of the convolutions is defined in space along the spatial sequence of the two-dimensional cross-sectional images. The three-dimensional convolutions comprise convolutions along the first dimension. The output layer is connected to the one or more convolutional layers in the neural network, and has an output value that is indicative of a probability of detecting prostate cancer in the computed tomography scan. Three-dimensional convolutions in the neural network provide a more accurate probability of detecting prostate cancer.