Seoul National University

Osong Medical Innovation Foundation, Korea University Research and Business Foundation

Planit Healthcare Inc.

Won-Ki Jeong

Jinwook Choi

Jeong Hwan Park

Youngkwan Kim

Hyunsoo Lee

Kyoungbun Lee

Kyung Chul Moon

Kyu Sang Lee

Gyeong Hoon Kang

Se Young Chun

Gheeyoung Choe

A cell image segmentation method using scribble labels includes iteratively pre-training via an image segmentation network (U-Net) using a cell image and scribble labels indicating a cell region and a background region as training data, calculating an exponential moving average (EMA) of image segmentation prediction probabilities at a predetermined interval during the pre-training, self-training by assigning the cell region and the background region for which the EMA of image segmentation prediction probabilities is over a preset threshold to be a pseudo-label, and iteratively refining the image segmentation prediction probability based on a scribbled loss (L) obtained through a result of the training and an unscribbled loss (L). Accordingly, it is possible to achieve cell image segmentation with high reliability using only scribble labels.

Method for cell image segmentation using scribble labels, recording medium and device for performing the method

Provided is a method of operating an apparatus for prostate cancer analysis operated by at least one processor, the method including: receiving digital slide images prepared from serial sections of a prostatectomy specimen, and a gross image of the serial sections; acquiring prostate cancer-related histological information of each received digital slide image using an artificial neural network model trained to infer histological information from the digital slide images; generating digital pathology images by displaying the prostate cancer-related histological information inferred from the artificial neural network model on each digital slide image; and providing a histological mapping image in which a tumor region extracted from each digital pathology image is mapped to a gross image of the corresponding section.

Apparatus and method for prostate cancer analysis

Growing community of inventors

Seoul, South Korea

Won-Ki Jeong