Medical Image Analysis

image

Product

TAIMedImg's radiology assistance system can accept CT, MRI, X-Ray, Ultrasonic, Pathological and automatically analyze them, assisting doctors in clinical interpretation.



Technology

TAIMedImg employs medical data, machine learning and deep learning tools to develop a system that can efficiently analyse data to human precision.

Product Introduction

Glioblastoma
system_brain1_image

AI Primary Brain Tumor Detection System

Automatically detect the primary brain tumor in MRI images, as well as label three kinds of area (tumor, necrosis, and edema), calculate the tumor size in real time, and provide the result to clinicians for assisting clinical diagnosis.


Turn On AI Prediction

Function

  • 1. AI Label
  • –Suspicious area
  • –Accuracy
  • –Interactive labeling
  • 2. Instance Segmentation & classification by Brain tumor,Edema,Necrosis
  • 3. Clinical report info (including Radiomic)
  • 4. Time: 20–30sec

Advantage

Assist doctors in identifying all possible metastasis, reducing the probability of failing to diagnose the disease.

Pain Point

pain point
Metastatic
system_brain1_image

DeepMets" AI Metastatic Tumor Detection System

DeepMets can be used in metastatic brain cancer and acoustic neuroma. AI systems learn from professional knowledge and experiences to develop professional capabilities for diagnosis support.


Turn On AI Prediction

Function

  • 1. AI Inference
  • –Suspicious area
  • –Accuracy
  • –Interactive labeling
  • 2.Instance Segmentation & classification by Brain tumor,Edema,Necrosis
  • 3. Clinical report info (including Radiomic)
  • 4. Time: 20-30sec

Advantage

Assist doctors in identifying all possible metastasis, reducing the probability of failing to diagnose the disease.

Pain Point

pain point
Metastatic
system_brain1_image

AI Metastatic Tumor Detection System

DeepMets can be used in metastatic brain cancer and acoustic neuroma. AI systems learn from professional knowledge and experiences to develop professional capabilities for diagnosis support.


Turn On AI Prediction

Function

  • 1. AI Label
  • –Suspicious area
  • –Accuracy
  • –Interactive labeling
  • 2.Instance Segmentation & classification by Brain tumor,Edema,Necrosis
  • 3. Clinical report info (including Radiomic)
  • 4. Time: 20-30sec

Advantage

Assist doctors in identifying all possible metastasis, reducing the probability of failing to diagnose the disease.

Pain Point

pain point
system_brain1_image

AI Pulmonary Nodule Detection System

Deep tagging and AI models help lung cancer categorization, diagnosis and prognostic prediction. Combine medical knowledge and AI technologies to automatically identify and tag suspicious nodules, increase efficiency, and minimize errors and omissions.


Turn On AI Prediction

Function

  • 1. AI Inference
  • –Suspicious area
  • –Accuracy
  • –Interactive labeling
  • 2.Instance Segmentation & classification by nodule
  • 3. Clinical report info (including Radiomic)
  • 4. Time: 30-40sec

Advantage

Automatically detect pulmonary nodules in CT images and reach conclusions faster

Pain Point

pain point
system_brain1_image

AI liver lesion detection system

In addition, we developed a semi-automatic labeling tool based on AI. This tool can not only calculate the maximum diameter of the tumor but also label across 10 slices in 3D view. It saves plenty of time of labeling and diagnosis.


Turn On AI Prediction

Function

  • 1. AI Inference
  • –Suspicious area
  • –Accuracy
  • –Interactive labeling
  • 2. Semantic Segmentation + Classification by Liver, Lesion
  • 3. Clinical report info (including Radiomic)
  • 4. Time: 20–30 sec

Advantage

Automatically segment liver and lesion areas in CT images and display the radiomics in seconds.

Pain Point

pain point
system_brain1_image

AI-based echocardiographic image classification and quality assessment

Different angles lead to various heart ultrasonic images and pulsations. To capture static and dynamic information, we build a image system to verify ultrasonic quality in real time.


Turn On AI Prediction

Function

  • 1. AI Inference
  • –Suspicious area
  • –Accuracy
  • 2. Semantic Segmentation + Classification by patient
  • 3. Info: Identifiy different views with quality score
  • 4. Time: 10sec

Advantage

Identify 9 different echocardiography perspectives, assess each one on their quality, and assist doctors in diagnosing, raising the quality of medicine provided.

Pain Point

pain point
system_malaria_image

Automatic Malaria Detection

Use AI models to identify malaria in blood smear in time for further inspections. Our system starts from deriving digitized blood smear images with automatic microscopy, utilizes the well-annotated blood smear images to establish malaria detection algorithm, and identifies parasite-infected blood cells automatically.

Function

  • 1. AI Label
  • –Suspicious area
  • 2. Detection + Classification by patient
  • 3. Info: Identify different types of infected red blood cells

Advantage

Identify all red blood cells that seem to have been infected, label and classify them, and assist professionals in diagnosing

Pain Point

pain point

Copyright ©2019 Taiwan AI Labs. All Rights Reserved.