-
公开(公告)号：US20230031698A1
公开(公告)日：2023-02-02
申请号：US17871342
申请日：2022-07-22
Applicant: SCREEN HOLDINGS CO., LTD.
Inventor： Ryo HASEBE
IPC: G06V20/69 , G06V10/26 , G06V10/764 , G06V10/776 , G06V10/774
Abstract: An image processing method according to the invention includes obtaining a ground truth image teaching a cell region occupied by a cell in an original image for each of a plurality of the original images obtained by bright-field imaging of the cell, generating a reverse image by reversing luminance of the original image at least for the cell region based on each original image, and constructing a classification model by performing machine learning using a set of the original image and the ground truth image corresponding to the original image and a set of the reverse image and the ground truth image corresponding to the original image as a basis of the reverse image respectively as training data.
-
Patent Agency Ranking
公开(公告)号：US20230230268A1
公开(公告)日：2023-07-20
申请号：US18148129
申请日：2022-12-29
Applicant: SCREEN HOLDINGS CO., LTD.
Inventor： Shogo Tokisue , Ryo HASEBE , Hiroshi OGI
CPC classification number: G06T7/571 , G06T3/0068 , G06T7/33 , G06T2207/10056 , G06T2207/10101 , G06T2207/20221 , G06T2200/24 , G06T2207/30024 , G06T2207/10148 , G06T2207/10028 , G06T2200/04
Abstract: An image display method includes the following operations (a) to (e). The (a) is of obtaining a plurality of two-dimensional images by two-dimensionally imaging a specimen, in which a plurality of objects to be observed are present three-dimensionally in the specimen, at a plurality of mutually different focus positions. The (b) is of obtaining image data representing a three-dimensional shape of the specimen. The (c) is of obtaining a three-dimensional image of the specimen based on the image data. The (d) is of obtaining the two-dimensional image selected from the plurality of two-dimensional images or a two-dimensional image generated to be focused on the plurality of objects based on the plurality of two-dimensional images as an integration two-dimensional image. The (e) is of integrating the integration two-dimensional image obtained in the (d) with the three-dimensional image obtained in the (c) and displaying an integrated image on a display unit.
-
公开(公告)号：US20240303811A1
公开(公告)日：2024-09-12
申请号：US18598539
申请日：2024-03-07
Applicant: SCREEN Holdings Co., Ltd.
Inventor： Ryo HASEBE
CPC classification number: G06T7/0012 , G06V10/44 , G06T2207/10101 , G06T2207/20081 , G06T2207/30024 , G06V2201/03
Abstract: First, a biological sample is imaged, and a photographic image in which intensity values are distributed is acquired. After that, a localization region corresponding to an unusual part is extracted from the photographic image. At that time, a region of which intensity value satisfies a predetermined requirement in the photographic image is extracted as the localization region. Alternatively, the photographic image is input to a trained model created in advance, and a localization region output from the trained model is obtained. In this manner, the localization region corresponding to the unusual part can be extracted from the photographic image of the biological sample. This enables noninvasive observation of the unusual part of the biological sample without processing a cell by staining or the like.
-
公开(公告)号：US20250157231A1
公开(公告)日：2025-05-15
申请号：US18838336
申请日：2022-12-20
Applicant: SCREEN Holdings Co., Ltd.
Inventor： Ryo HASEBE , Yasushi KUROMI
Abstract: According to this image standardization method, first, a region corresponding to a whole or a part of a biological sample in an image is extracted as an observation target region. Subsequently, an aggregate value of intensity values is calculated for each of multiple observation target regions. After that, intensity values of the multiple observation target regions are adjusted so that a difference in the aggregate value becomes smaller. This reduces variation in brightness among the multiple observation target regions. This can reduce a difference in an intensity value of the observation target region for the image of the biological sample including multiple cells. Consequently, multiple biological samples can be fairly observed and evaluated.
-
公开(公告)号：US20240420327A1
公开(公告)日：2024-12-19
申请号：US18816477
申请日：2024-08-27
Applicant: SCREEN Holdings Co., Ltd.
Inventor： Ryo HASEBE
Abstract: An image processing method comprises obtaining three-dimensional image data representing a three-dimensional image of a cell aggregate with a cavity inside obtained by optical coherence tomography imaging, obtaining a thickness distribution representing a thickness of the cell aggregate between an outer surface of the cell aggregate facing an outside space and an inner surface thereof facing the cavity at each position based on the three-dimensional image data; and detecting protrusion parts having a prominent thickness as compared to a surrounding region based on the thickness distribution.
-
公开(公告)号：US20240303813A1
公开(公告)日：2024-09-12
申请号：US18599728
申请日：2024-03-08
Applicant: SCREEN Holdings Co., Ltd. , Cyfuse Biomedical K.K.
Inventor： Ryo HASEBE , Toshihiko Maekawa , Ayu Inoue
CPC classification number: G06T7/0012 , G06T7/11 , G06T2207/30024
Abstract: In an evaluation method, first, spheroids obtained by three-dimensional culture of multiple kinds of liver-derived cells are imaged by optical coherence photography, a localization region is extracted from the photographic image, the localization region is analyzed (analysis step), and the condition of the spheroid is evaluated. The analysis step includes a first calculation step of calculating the area of the entire spheroid in the photographic image, a second calculation step of calculating the area of the localization region, a third calculation step of calculating the ratio of the localization region on the basis of the two areas, and a fourth calculation step of calculating an evaluation parameter on the basis of the ratio.
-
-
-
-
-
Search Results
6
records
(took 0.014 seconds)
