Abstract: A system for disease diagnosis includes a patch neural network for generating a patch-level diagnostic result of whether or not a disease is present in each of predetermined patches formed by dividing a slide into a predetermined size; a heat map generation module for generating a patch-level heat map image corresponding to the biometric image obtained from the slide on the basis of the patch diagnostic results of the respective multiple patches included in the slide; a tissue mask generation module for generating a tissue mask image corresponding to the biometric image obtained from the slide on the basis of a hue-saturation-value (HSV) model corresponding to the slide; and a visualization module for generating a disease diagnostic visualization image corresponding to the biometric image obtained from the slide on the basis of the patch-level heat map image and the tissue mask image.

Abstract: A system for diagnosing a disease, implemented in a system and which uses a slide of a biological image, and a neural network, includes a patch level segmentation neural network which, for each patch in which the slide is divided into a predetermined size, receives the patch through an input layer and specifies an area in which a disease exists in the patch, wherein the patch level segmentation neural network is provided with a disease diagnosis system including a patch level classification neural network which receives the patch through an input layer and outputs a patch level classification result regarding whether the disease exists in the patch, and a patch level segmentation architecture which receives a feature map generated in each of plural feature extraction layers among hidden layers included in the patch level classification neural network and specifies an area in which a disease exists in the patch.

Abstract: A disease diagnosis system uses a slide of a biological image and the neural network, the disease diagnosis system including a patch-level segmentation neural network that receives, for each predetermined patch in which the slide is divided into a predetermined size, the patch as an input layer so as to specify the area in which the disease in the patch exists, wherein the patch-level segmentation neural network comprises: a patch-level classification neural network, which receives the patch as an input layer so as to output a patch-level classification result about whether the disease exists in the patch; and a patch-level segmentation architecture, which receives a feature map generated in each of two or more feature map extraction layers from among hidden layers included in the patch-level classification neural network, so as to specify the area in which the disease in the patch exists.

Abstract: A two-phase disease diagnosis system includes a processor and a storage device storing a neural network, and using a slide including a biometric image and the neural network. The system includes a patch neural network that receives, through an input layer, a given patch segmented in a given size from the slide and outputs patch level diagnosis results indicating whether a disease is present in the patch and a slide diagnosis engine that marks a patch determined to be cancer based on the patch level diagnosis results for each of multiple patches included in the slide and outputs slide level diagnosis results indicating whether a disease is present in the slide based on the marked results. The patch neural network receives, through the input layer, four-channel information including original color information three-channels and a gray channel for the patch.

Abstract: A disease diagnosis system implemented in a system including a processor and which uses the neural network and a slide on which biological tissue is provided and from which a biological image is obtained, the system including: a pre-processing module for generating first to Kth image information corresponding to each of a plurality of patches obtained by dividing the slide into a predetermined size, wherein ith image information, 1<=i<=K, includes at least one channel value corresponding to an ith color model, and a first color model to a Kth color model are different from each other; and a patch neural network which receives the first image information to the Kth image information corresponding to each of the plurality of patches, so as to output a patch-level diagnosis result indicating whether a disease exists in each patch.

Abstract: A disease diagnosis system includes a processor and a storage device storing a neural network. The processor trains the neural network in the storage device to output a determination value corresponding to a probability having at least one of a plurality of states using a given loss function and learning data labeled so that a given unitary unit included in a biometric image is to have at least one of the plurality of states. The neural network includes a specific layer to output a plurality of feature values corresponding to a probability that the unitary unit is to be determined as each of the plurality of states. The loss function incorporates both first and second feature values corresponding to first and second states into a dual labeling unitary unit with the first state having a higher probability and a second state having lower probability.

Abstract: A system and a method that output in both a machine-readable and a human-readable format, a result obtained by performing a diagnosis of a disease through an image of living tissue. A diagnosis result generation system includes a marking information generation module for generating marking information indicating a result obtained by diagnosing whether a disease is present in biological tissue provided on a slide of which a biological image is obtained therefrom, wherein the marking information includes disease state information for each pixel of the biometric image obtained from the slide; a contour extraction module for extracting at least one contour from the marking information; and a machine-readable/human-readable generation module for generating a machine-readable/human-readable document including outline information of each of the at least one extracted contour.

Abstract: A system for searching for a pathological image includes: an autoencoder having an encoder for receiving an original pathological image and extracting a feature of the original pathological image, and a decoder for receiving the feature of the original pathological image extracted by the encoder and generating a reconstructed pathological image corresponding to the original pathological image; a diagnostic neural network for receiving the reconstructed pathological image generated by the autoencoder that has received the original pathological image, and outputting a diagnosis result of a predetermined disease; and a training module for training the autoencoder and the diagnostic neural network by inputting a plurality of training pathological images, each labeled with a diagnosis result, into the autoencoder. The autoencoder is trained by reflecting the diagnosis result of the reconstructed pathological image output from the diagnostic neural network.

Abstract: A disease diagnosis system including a processor and a storage device for storing a neural network and using a biometric image and the neural network, the disease diagnosis system including a micro-neural network for receiving a first tile included in the biometric image through a first input layer, and including a plurality of first layers and an output layer, and a macro-neural network for receiving a macro-tile including the first tile and at least one or more second tiles adjacent to the first tile through a second input layer, and including a plurality of second layers and the output layer, in which the output layer includes at least one state channel indicating a state of a disease of a biological tissue corresponding to the first tile.

Abstract: A method for searching for an endpoint of a specific data cluster and a data processing system therefor, including the steps of: a) receiving, by a search system, an input of a numerical value for each of a plurality of individual data included in a data set; b) dividing a numerical range, to which the numerical value may belong, into a plurality of bins each bin having a predetermined bin width, by using the respective numerical values received by the search system, and generating histogram data having, as a bin value, the number of individual data corresponding to each of the divided bins; and c) searching for a target bin present at an endpoint of a specific cluster on the basis of the histogram data generated by the search system.