Features

AI HPC Platform enables you to perform Automated ML / AI

AI HPC System


Flowchart

Main Features


a

Super Fast AI

AI system for real-time gigapixel analysis

Intuitive

Real-time AI inference

3,800 times faster than existing AI systems

Rapid customized AI modeling

Fast with adaptive AI (evolving AI)

>99% accuracy

>99% accuracy in segmentation, detection, and classficiation

Unlimited scale up

Various scales of storage and computing packages to choose for AI HPC

Work for small or large dataset

AI models can be built based on limited data

Feature Description

Super fast AI system enables real-time analysis of gigapixel images and development of customized AI model in a short time based on a super fast and smart adaptive deep learning framework. AI models could be built to detect objects, defects, or even tissues of various types of cancers, syndromes, or diseases, and compute the quantitative information in an image. Moreover, AI models can be trained with limited data and produce accurate results. AI models could be easily improved and adapted by the smart-learning framework.

Benefits

Real-time analysis of gigapixel images, fast modeling, and rapid optimization of deep learning AI models

High-speed development of customized AI models

Customized AI modeling service is also available

AI models can be trained with limited data and produce accurate results

Super Fast AI Platform Utilities

Intuitive

Real-time AI Analysis

Real-time AI analysis on large or multi-dimensional data

Intuitive

Quantitative Report

Automatic AI analysis and quantitative report

Intuitive

AI Leaderboard

Automatic AI analysis and quantitative report

Intuitive

Multiple GPU Nodes

GPU resource manager.

Intuitive

Evolving AI

Train with adaptive learning (evolving AI).

Intuitive

One or Multiple Slides

Train or test for one or multiple images/data.

Wang CW* et al (2017) A Benchmark for Comparing Precision Medicine Methods in Thyroid Cancer Diagnosis using Tissue Microarrays, Bioinformatics (SCI, JCR 2016 5% (4/78) in MATHEMATICAL & COMPUTATIONAL BIOLOGY, IF=7.307))

Veta et al. (2015) Assessment of algorithms for mitosis detection in breast cancer histopathology images, Medical Image Analysis 20 237-248

Wang* and Yu (2013) Automated Morphological Classification of Lung Cancer Subtypes using H&E Tissue Images, Machine Vision and Applications, Volume 24, Issue 7, Page 1383-1391

Buckley et al., The ΔNp63 proteins are key allies of BRCA1 in the prevention of basal-like breast cancer (2011), Cancer Research, 71: 1933

Wang*, Robust Automated Tumour Segmentation on Histological and Immunohistochemical Tissue Images (2011), PLOS ONE, 6:2, e15818

a

Terapixel Image Platform

Super fast terapixel server

Intuitive

Full compatibility

We support all types of image format and image scanners

24x faster

24 times faster than Leica

Cross-platforms

Web-based platform for Windows, Linux, iOS, Mac, Android, etc

Feature Description

Super fast terapixel server allows multiple users to view and analyze gigapixel and even terapixel images with mobile devices, such as iPhones, iPads, Android devices, or laptops at the same time. Users could annotate and analyze the data using the built-in tools such as 3D effect, color deconvolution, color quantification, nuclei/object detection, length and area measurement, and quantitative reporting system.

Benefits

Real-time analysis of gigapixel images, fast modeling, and rapid optimization deep learning AI models

Analyze large scale microscopic images in real-time

Compatible with many scanners, such as Leica, Hamamatsu, 3DHISTECH, Motic

The terapixel platform in your pocket

Wireless & wired internet device

Allow multi-user to access at the same time

Compatible with various file formats, such as JPG, TIFF, SVS, DICOM, NDPI, MIREX, SCN, BMP, PNG

Terapixel Image Platform Utilities

Intuitive

Annotation Tools

Draw your annotation for your analysis with our most flexible annotation tools.

Intuitive

Annotation List

Review and modify your annotations. Add attributes, reshape, and so many more.

Intuitive

Microscope Slide Labels

Add your microscope slide labels to the terapixel image platform database.

Intuitive

Multiple GPU Nodes

GPU resource manager.

Intuitive

Zoom-in, Zoom-out, No Lag

Get the details of your terapixel images smoothly. Zooming through the web browser has never been this fast.

Intuitive

3D Analysis

Apply 3D analysis for your terapixel images.

Intuitive

Gradient Map Visualization

Give your terapixel images a gradient map visualization.

Intuitive

Quantitative Analysis and Stain Separation Tools

Conduct quantitative analysis and stain separation using our platform.

Intuitive

Tumor Detection

AI will assist pathologists to find tumors on WSIs.

Wang C.-W.*, Huang C., Hung C. (2015) VirtualMicroscopy: ultra-fast interactive microscopy of gigapixel / terapixel images over internet, Nature-Scientific Reports 5: 14069 (SCI, JCR 2015 (7/63) in MULTIDISCIPLINARY SCIENCES, IF=5.228)

Wang, C.-W., Hung, C, Gigapixel/Terapixel Interactive Real Time Visualization and Cloud System, USA Patent (9,047,318)

Wang, C.-W., Hung, C, Ultra-high-resolution interactive video display and cloud system and its management methods, Patent, Taiwan(I571762)

a

3D Image Platform and AI

Super fast 3D reconstruction and analysis server, now with an AI

Intuitive

AI-powered

We support all types of image format and image scanners

24x faster

24 times faster than Leica

Cross-platforms

Web-based platform for Windows, Linuxe, iOS, Mac, Android, etc

Feature Description

Super fast 3D reconstruction and analysis server provide the interactive real time 3D reconstruction platform for users to analyze 3D shapes of images more precisely and accurately. It provides real-time 3D reconstruction of data and various image processing tools. Now, our 3D images platform is equipped with super fast AI. Our award-winning platform can do volummetric segmentation with high accuracy. Automatic segmentation of 3D object/tumor/organ/syndromes from various scanner modalities can be done in only seconds.

Benefits

Real-time and interactive

3D reconstruction image analysis

Wireless & wired device

Allow multi-user at the same time

Provide fully automatic image registration and difference anlysis function, enabling fusion and comparison of multiple 3D or 2D datasets. For data with higher or lower dimension, customized models could be built

3D AI to perform segmentation of 3D object/organ/tumor

Examples

Intuitive

Automatic segmentation of thoracic organs-at-risk in CT images (heart, aorta, trachea, esophagus) in seconds by 3D AI

Intuitive

Automatic segmentation of liver from CT data in seconds by 3D AI

Intuitive

Automatic segmentation of liver from 3D MRI data in seconds by 3D AI

Intuitive

Interactive and real-time 3D reconstruction of serial 2D images

Various applications:

Intuitive

Breast Cancer

Intuitive

Lung Cancer

Intuitive

Tumor Segmentation

Intuitive

AI for Chest X-Ray

Intuitive

3D AI in CT

Intuitive

3D AI in MRI

Intuitive

3D AI Platform

Intuitive

Fetal Ultrasound

Intuitive

Diabetic Retinopathy

Intuitive

Dental X-Ray Difference

Intuitive

Landmark Detection

Intuitive

Dental X-Ray Image Analysis

Intuitive

Breast Cancer X-Ray

Intuitive

Automatic Alignment

Intuitive

Sleep Apnoea Diagnosis

Intuitive

PCB Inspection

Intuitive

IC Inspection

Intuitive

Real-time Defect Detection

Intuitive

Occluded Object Detection

Intuitive

LCD Panel Inspection

Intuitive

Oil Spill Analysis

Intuitive

Traffic Analysis

Intuitive

Crowdness Analysis

Intuitive

Illegal Parking Detection

Intuitive

Face Recognition

Intuitive

Action Recognition

Intuitive

Behaviour Analysis

Intuitive

Accident Detection

Intuitive

Field Health Analysis

Intuitive

Forest Fire Detection

Patents and Awards

Patents

Gigapixel/Terapixel Interactive Real Time Visualization and Cloud System, Taiwan Invention Patent, I571762 , Wang, C.-W., Hung, C

Gigapixel/Terapixel Interactive Real Time Visualization and Cloud System, USA Patent, 9,047,318 , Wang, C.-W., Hung, C

Automatic Cephalometric Landmark Detection and Registration in Lateral Cephalograms, Taiwan Invention Patent , I499985, Wang, C.-W., Li C.

Fetal Ultrasound Automatic Segmentation Method and System (IMAGE RECOGNITION METHOD AND IMAGE RECOGNITION SYSTEM), USA Invention Patent, 9,020,252, Wang, C.-W.

Fetal Ultrasound Automatic Segmentation Method and System (IMAGE RECOGNITION METHOD AND IMAGE RECOGNITION SYSTEM), Taiwan Invention Patent, I501754, Wang, C.-W.

Image registration method, USA Patent, 9,208,560, Wang, C.-W.

Patents in application

Wang C., Chen Y., Cross-staining and multi-biomarker method for assisting in cancer diagnosis, Taiwan, 107118500

Wang C., Chen Y., Cross-staining and multi-biomarker method for assisting in cancer diagnosis-USA, 1070013US0

Wang C., Chen Y., Cross-staining and multi-biomarker method for assisting in cancer diagnosis-China, 1070013CN0

Wang C., Chen Y., Cross-staining and multi-biomarker method for assisting in cancer diagnosis-Japan, 1070013JP0

Wang C., Ko H., SPLINE IMAGE REGISTRATION METHOD, Taiwan, 107121575

Wang C., Ko H., SPLINE IMAGE REGISTRATION METHOD, USA, 1070025US0

Wang C., Ko H., SPLINE IMAGE REGISTRATION METHOD, China, 1070025CN0

Wang C., Ko H., SPLINE IMAGE REGISTRATION METHOD, Japan, 1070025JP0

Wang C., MEDICAL IMAGE ANALYSIS METHOD APPLYING MACHINE LEARNING AND SYSTEM THEREOF, Taiwan, 107124061

Wang C., MEDICAL IMAGE ANALYSIS METHOD APPLYING MACHINE LEARNING AND SYSTEM THEREOF, USA, 1070021US0

Wang C., MEDICAL IMAGE ANALYSIS METHOD APPLYING MACHINE LEARNING AND SYSTEM THEREOF, German, 1070021DE0

Wang C., MEDICAL IMAGE ANALYSIS METHOD APPLYING MACHINE LEARNING AND SYSTEM THEREOF, Japan, 1070021JP0

Awards

Excellence in Research Award, National Taiwan University of Science and Technology, 2018.2-2020.1

The Outstanding Research and Creativity Award, National Taiwan University of Science and Technology, 2016.2-2018.1

Innovation Award, the 16th China International Industry Fair, Shangai, China, 2014.11.3-7

1st Prize, the 4th Annual Creative Entrepreneurship Competition ─ New Business Development Group, National Taiwan University of Science and Technology, 2013.5

Young Scholar Award, National Taiwan University of Science and Technology, 2013.1-2015.12

Excellence in Research Award, National Taiwan University of Science and Technology, 2013.2-2016.1

Second place, Right Ventricle Segmentation Challenge in 4D Cardiac MRI, 2012 Rouen, sponsored by Toshiba, PIE Medical Imaging and Medis

First Prize, Fetal Femur Challenge, Organized by Oxford University, 2012

Distinguished Young Scholar 3-Years Research Fund, by National Science Council of Taiwan

Set for Britain, Selected to present the research in the House of Commons, London, March 8th 2010


© AI Explore Team 2019