Silicon Canals
TRENDING
  • Mobility
  • AI
  • FinTech
  • Software & SaaS
No Result
View All Result
  • Home
  • News
    • COVID-19
    • Startups
    • Scaleups
    • FinTech
    • AI and chatbots
    • Travel & Mobility
    • Software & SaaS
    • Health & Medtech
    • (Crowd)funding
    • Blockchain
    • Internet of things
    • Hardware
    • Accelerators
    • E-commerce
    • Cybersecurity
    • Gaming & Virtual Reality
    • Drones
  • Features
    • Promoted content
    • How-to
    • Knowledge & Insights
    • Guest Contributions
    • Scaling-up in Europe
  • Amsterdampartner
  • Events
  • Newsletter
  • Jobs
  • About
    • Partner with Silicon Canals
    • Team
    • Contact
  • Home
  • News
    • COVID-19
    • Startups
    • Scaleups
    • FinTech
    • AI and chatbots
    • Travel & Mobility
    • Software & SaaS
    • Health & Medtech
    • (Crowd)funding
    • Blockchain
    • Internet of things
    • Hardware
    • Accelerators
    • E-commerce
    • Cybersecurity
    • Gaming & Virtual Reality
    • Drones
  • Features
    • Promoted content
    • How-to
    • Knowledge & Insights
    • Guest Contributions
    • Scaling-up in Europe
  • Amsterdampartner
  • Events
  • Newsletter
  • Jobs
  • About
    • Partner with Silicon Canals
    • Team
    • Contact
No Result
View All Result
Silicon Canals
No Result
View All Result

Google’s AI team classifies chest X-rays with superior levels of accuracy

Editorial team by Editorial team
December 6, 2019
in News, AI and chatbots, Health & Medtech
0
Google’s AI team classifies chest X-rays with superior levels of accuracy
41
SHARES
LinkedInTwitterWhatsAppFacebook

While millions of diagnostic examinations are carried out annually, chest X-rays play a vital role in diagnosing several diseases. But the usefulness of the same can be limited due to the challenges in interpretation that need thorough and rapid evaluation of 2D image depicting complex, 3D organs, and disease processes. Sometimes, major details can be missed by chest X-rays resulting in adverse outcomes for patients.

Recent efforts have improved lung cancer detection in radiology, differential diagnosis in dermatology, and prostate cancer grading in pathology. And, obtaining accurate clinical labels for the deep learning models for X-ray interpretation.

 - Partner content -
Cost of software - The Software house
The cost of building software
Marek Gajda, CTO of The Software House, shares his insights on building high-quality...Show More
Marek Gajda, CTO of The Software House, shares his insights on building high-quality software. Show Less
Read more

Most efforts have applied rule-based natural language processing to radiology reports or based on image review by readers. Eventually, both might introduce inconsistencies, that can be problematic at the time of model evaluation.

Deep learning models to resolve challenges!

In an effort to resolve this, researchers at Google devised Artificial Intelligence models to spot four findings on human chest X-rays. Advances in machine learning present an opportunity to create new tools to help experts interpret medical images. In the journal Radiology, the deep learning models for chest radiograph were published.

The team developed deep learning models for four important clinical finds such as pneumothorax (collapsed lungs), nodules, and masses, airspace opacities (filling of the pulmonary tree with material), and fractures. These were chosen in consultation with clinical colleagues and radiologists to focus on conditions that are critical for patient care.

These deep learning models were evaluated using several thousands of held out images from the dataset for which the high-quality labels have been collected using a panel-based adjudication process among radialogists who are certified by the board. Later, the held-out images have been reviewed independently by separate radiologists to make sure these are accurate.

The team leveraged more than 600,000 images sourced from two de-identified datasets. The first one was developed along with co-authors at the Apollo Hospitals and has a diverse set of chest X-rays gathered over several years from the hospital network across locations. The second one has been released publicly by the National Institutes of Health and served as a vital resource for machine learning efforts. But the same has limitations related to accuracy and clinical interpretation of available labels.

High-quality reference standard labels

In order to generate high-quality reference standard labels for model evaluation, the team has used a panel-based adjudication process. In this process, three radiologists reviewed the final tune and test set images and addressed disagreements via discussion. It let difficult findings that were only detected by a single radiologist to be detected and documented. Later, the discussions took place anonymously via an adjudication or online discussion system.

Google notes that while the models achieved an overall expert-level accuracy, the performance varied based on the corpora. For instance, the sensitivity to detect penumothorax among radiologists was nearly 79% for ChestX-ray14 images and just 52% for the same radiologists in other datasets.

The team hopes to lay the groundwork for exceptional methods with a corpus of the adjudicated labels for the ChestX-ray14 dataset that they have made available in open source. This comprises 2,412 training and validation set images and 1,962 test set images or 4,372 images in total.

Stock photo from create jobs 51/Shutterstock

Stay tuned to Silicon Canals for more European technology news.

close

Want to stay on top of the latest rounds, coolest tech and hottest startups?

Subscribe to our free, daily news blast and stay up-to-date!

Check your inbox or spam folder to confirm your subscription.

 - Partner content -
Nebula CTO Andre Witte
Cloud spend management 😱
Tracking cloud spend can be tricky. Learn from Nebula's CTO Andre Witte how to keep the...Show More
Tracking cloud spend can be tricky. Learn from Nebula's CTO Andre Witte how to keep the costs at bay in 2021. Show Less
Read more
Tags: Artificial Intelligencehealthtech
Share3Tweet9SendShare19

Featured events

07apr10:0012:30FeaturedWorkshop: pr for startupsHow to grow your startup with the help of public relations

Partner content | Work with us

Blue Tulip Awards announce 5 finalists in the Food & Water and Education & Employment theme

Impact media foundation 5 Media offers grants up to €100,000 per year to drive a sustainable future

Here’s how cryptocurrency will take over the world

Blue Tulip Awards is driven by purpose: why innovation and equality go hand in hand

Future of coding: CTOs discuss low-code versus custom code software development for their business

Breaking news from Amsterdam | Partner

Here’s why Just Eat Takeaway is expanding its offices in Amsterdam

Amsterdam-based Veylinx uses Nobel Prize-winning approach to solve the problem of unreliable survey data; raises €1.6M

Amsterdam-based fintech unicorn Mollie enters UK market to helps merchants there grow into Europe post-Brexit

Amsterdam-based HR tech Recruitee takes a “shortcut” to a future unicorn status

  • About Silicon Canals
  • Partner with Silicon Canals
  • Contact us
  • Newsletter
  • Disclaimer
  • Privacy policy
  • Terms & Conditions Silicon Canals

Silicon Canals 2014-2020 | Website: Bright Idiots

No Result
View All Result
  • Home
  • News
    • COVID-19
    • Startups
    • Scaleups
    • FinTech
    • AI and chatbots
    • Travel & Mobility
    • Software & SaaS
    • Health & Medtech
    • (Crowd)funding
    • Blockchain
    • Internet of things
    • Hardware
    • Accelerators
    • E-commerce
    • Cybersecurity
    • Gaming & Virtual Reality
    • Drones
  • Features
    • Promoted content
    • How-to
    • Knowledge & Insights
    • Guest Contributions
    • Scaling-up in Europe
  • Amsterdam
  • Events
  • Newsletter
  • Jobs
  • About
    • Partner with Silicon Canals
    • Team
    • Contact

Silicon Canals 2014-2020 | Website: Bright Idiots

X

This website uses cookies to ensure the best possible experience. We may share this information with our advertising and analytics partners. By clicking 'Accept', you agree to our use of cookies and similar technologies. More info