公开(公告)号：US20210407675A1

公开(公告)日：2021-12-30

申请号：US17294283

申请日：2019-11-18

Applicant: DEEP BIO INC.

Inventor： Sun Woo KIM

IPC: G16H50/20

Abstract: Disclosed are a supervised learning-based consensus diagnosis method and a system thereof. The supervised learning-based consensus diagnosis method includes: a step of confirming, by a consensus diagnostic system, N individual diagnosis results in which each of N (N is an integer of 2 or more) diagnostic systems receives and outputs predetermined biological data, wherein the N diagnostic systems, respectively, are systems that are each trained with learning data annotated by different annotation subjects; and a step of outputting a consensus diagnosis result of the biological data on the basis of the individual diagnosis results confirmed by the consensus diagnosis system.

公开(公告)号：US20210248745A1

公开(公告)日：2021-08-12

申请号：US17271214

申请日：2019-08-07

Applicant: DEEP BIO INC.

Inventor： Joon Young CHO ,  Sun Woo KIM

IPC: G06T7/00 ,  G06T7/11 ,  G06T7/90

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.

公开(公告)号：US20210142900A1

公开(公告)日：2021-05-13

申请号：US16972231

申请日：2019-06-04

Applicant: DEEP BIO INC.

Inventor： Sanghun LEE ,  JoonYoung CHO ,  Sun Woo KIM

IPC: G16H50/20

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.

公开(公告)号：US20240221373A1

公开(公告)日：2024-07-04

申请号：US18288380

申请日：2022-04-20

Applicant: DEEP BIO INC.

Inventor： Jun Young CHOI ,  Tae Yeong KWAK ,  Sun Woo KIM

IPC: G06V10/82 ,  G06T7/00 ,  G16H50/20

CPC classification number: G06V10/82 ,  G06T7/0012 ,  G16H50/20 ,  G06T2207/30068 ,  G06T2207/30096 ,  G06V2201/032

Abstract: A training method for training an artificial neural network capable of determining a breast cancer lesion area in consideration of both microscopic features and macroscopic features of biological tissue, and a computing system for performing same. A method is provided for training an artificial neural network, comprising steps in which: an artificial neural network training system acquires a slide image of a biological tissue slide; the artificial neural network training system acquires, from the slide image, a first high-resolution patch to an Nth high-resolution patch; the artificial neural network training system acquires an ith low-resolution patch corresponding to an ith high-resolution patch (1

公开(公告)号：US20230298753A1

公开(公告)日：2023-09-21

申请号：US18017397

申请日：2020-07-23

Applicant: DEEP BIO INC.

Inventor： Sang Hun LEE ,  Sun Woo KIM

IPC: G16H50/20 ,  G16H50/70 ,  G16H30/40 ,  G06T7/00

CPC classification number: G16H50/20 ,  G16H50/70 ,  G16H30/40 ,  G06T7/0014 ,  G06T2207/30024 ,  G06T2207/20081 ,  G06T2207/20084 ,  G06T2207/30081

Abstract: Disclosed in the present invention are a method for annotating a pathogenic site of a disease by means of semi-supervised learning. According to an aspect of the present invention, provided is the method comprising the steps in which: the diagnosis system using a neural network generates a patch-level classification neural network, which predicts a classification result relating to whether or not a predetermined disease is present in a patch, and a pixel-level classification neural network which predicts a classification result relating to the disease for each pixel constituting the patch; the diagnosis system obtains a plurality of slide images for learning, wherein each of the plurality of slide images for learning is labeled with a corresponding slide-level diagnosis result; and the diagnosis system gradually learns the patch-level classification neural network and pixel-level classification neural network by means of the plurality of slide images for learning.

公开(公告)号：US20230298752A1

公开(公告)日：2023-09-21

申请号：US18017382

申请日：2020-07-23

Applicant: DEEP BIO INC.

Inventor： Sun Woo KIM

IPC: G16H50/20

CPC classification number: G16H50/20

Abstract: Disclosed are a method and a system for providing a customized diagnosis system. A method for providing a diagnosis system, which is to achieve the technical task, comprises the steps wherein: a reference diagnosis system transmits, to a first diagnostic subject system, M (M is an integer of 2 or larger) pieces of unannotated learning data for customization; the reference diagnosis system receives the M pieces of learning data for customization, which are annotated by a first diagnostic subject side, from the first diagnostic subject system; and the received annotated M pieces of learning data for customization are reflected in the reference diagnosis system, wherein the M pieces of learning data for customization, which have been annotated by the first diagnostic subject side, are reflected in a neural network for the reference diagnosis system under a condition that a predetermined price is to be paid.

公开(公告)号：US20230229927A1

公开(公告)日：2023-07-20

申请号：US18008426

申请日：2021-06-03

Applicant: DEEP BIO INC.

Inventor： Sun Woo KIM ,  Tae Yeong KWAK ,  Hye Yoon CHANG ,  Ye Chan MUN

IPC: G06N3/09 ,  G06N3/0455 ,  G06T7/00

CPC classification number: G06N3/09 ,  G06N3/0455 ,  G06T7/0012 ,  G06T2207/30096 ,  G06T2207/20081 ,  G06T2207/20084 ,  G06T2207/20076 ,  G06T2207/20021

Abstract: The present disclosure discloses a method and system for training a neural network for determining severity, and more particularly, a method and system which may effectively learn a neural network performing patch unit severity diagnosis using a pathological slide image to which a severity indication (label) is given.

公开(公告)号：US20210391076A1

公开(公告)日：2021-12-16

申请号：US17282775

申请日：2019-10-04

Applicant: DEEP BIO INC.

Inventor： Tae Yeong KWAK ,  Sang Hun LEE ,  Sun Woo KIM

IPC: G16H50/20 ,  G06T7/00 ,  G16H30/40 ,  G06N3/08 ,  G06N3/04

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.

公开(公告)号：US20210304889A1

公开(公告)日：2021-09-30

申请号：US15734907

申请日：2019-06-04

Applicant: DEEP BIO INC.

Inventor： Sanghun LEE ,  JoonYoung CHO ,  Sun Woo KIM

IPC: G16H50/20 ,  G06N3/04 ,  G06N3/08 ,  G06K9/62 ,  G16H30/40 ,  G06T7/73

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.

公开(公告)号：US20210295997A1

公开(公告)日：2021-09-23

申请号：US17266119

申请日：2019-08-07

Applicant: DEEP BIO INC.

Inventor： Sun Woo KIM

IPC: G16H50/20 ,  G16H30/40 ,  G16H50/30

Abstract: A biometric image diagnosis system includes a first search module that retrieves, from a database (DB) storing pre-diagnosed biometric images and results with regards to the pre-diagnosed biometric images, at least one similar biometric image having similar characteristics to a predetermined biometric image to be diagnosed; a second search module configured to retrieve at least one pre-diagnosed biometric image pre-diagnosed by a diagnostician from the DB; a communications module that transmits the at least one similar biometric image and the at least one pre-diagnosed biometric image to a terminal of the diagnostician; and a storage module that stores, in the DB, a disease diagnosis result with respect to the biometric image to be diagnosed, wherein the terminal of the diagnostician displays the biometric, image to be diagnosed, the at least one similar biometric image, and the at least one pre-diagnosed biometric image.