01

Overcoming the limitations of existing
neuroendocrine tumor(NET) image reports

The limited number of pathologists produced per year, routinely performed diagnostic tasks, various types of diseases, and the
complexity of their causes have brought about limitations on accuracy in diagnosis results.
Xpathology can offer faster and more reliable diagnostic results by automated solutions using AI-based histopathological findings and deep learning-based digital image analysis technology.

02

Development and advancement of
digital pathology image analysis
technology based on transfer learning

Xpathology using transfer learning, a technology that converts deep neural networks learned from large amounts of general image data into deep neural networks for digital pathology image classification, can achieve high performance with only a small number of digital pathology image data.

Core Technology: CNN based Image Classification

This is one of the areas of computer vision research aiming at precisely distinguishing which class label an image belongs to when given a certain image.
The use of a convolutional neural network (CNN) model, which delivers high performance in image classification, provides more accurate identification of cancer tissue and cancer grade in pathological images.

03

Presenting objective basis for
pathological image analysis using
AI technology

AucureX utilizes 'eXplainable AI' technology, an explainable artificial intelligence that improves the reliability of pathologists' decision-making processes by providing objective data through localizing the basis of the results derived from the deep learning model into specific areas within the pathological image. This enhances the consistency of results when reading image data, allowing to expect more precise diagnosis.

Core Techology : eXplainable

This is a technology that enables users to accurately understand and interpret the movements of the artificial intelligence system and the final analysis so that they can explain the process of generating outcomes.