Connect with us

Life

Predicting Fetal Well-being with Deep Learning

AI fetal monitoring using deep learning models for CTG interpretation improves neonatal outcomes and reduces provider burden.

Published

on

AI fetal monitoring

TL;DR:

  • AI models can predict fetal well-being using cardiotocography (CTG) signals.
  • Deep learning methods improve CTG interpretation, reducing false-positive rates.
  • Combining fetal heart rate (FHR) and uterine contractions (UC) data enhances model performance.

The Future of Fetal Monitoring

Imagine a world where artificial intelligence (AI) can accurately predict the well-being of a fetus during pregnancy and labour. This is no longer a dream; it’s a reality. Recent advancements in AI and deep learning are revolutionising the field of fetal monitoring, offering the potential to improve neonatal outcomes and reduce the burden on healthcare providers. In this article, we delve into the groundbreaking work on developing and evaluating machine learning models for cardiotocography (CTG) interpretation.

Understanding Cardiotocography (CTG)

Cardiotocography (CTG) is a crucial technique used during pregnancy and labour to monitor fetal well-being. It involves recording the fetal heart rate (FHR) and uterine contractions (UC) using doppler ultrasound. CTG can be done continuously or intermittently, with sensors placed externally or internally.

Currently, healthcare providers interpret CTG recordings using guidelines from organisations like the National Institute of Child Health and Human Development (NICHD) or the International Federation of Gynecologists and Obstetricians (FIGO). These guidelines define patterns in CTG and FHR traces that may indicate fetal distress.

The Role of AI in Improving CTG Interpretation

Despite its widespread use, CTG interpretation is complex and subjective, leading to high false-positive rates and intra- and inter-observer variability. This is particularly challenging in low-resource settings where access to skilled interpreters is limited.

Enter AI. Recent research has focused on using deep learning methods to improve CTG interpretation. These methods use physiological time series data as input, offering a more comprehensive analysis compared to traditional feature extraction techniques.

Advertisement

Developing and Evaluating Deep Learning Models for CTG

In a recent paper titled “Development and evaluation of deep learning models for cardiotocography interpretation,” researchers developed end-to-end neural network-based models to predict measures of fetal well-being. These models were trained on an open-source CTG dataset, the CTU-UHB Intrapartum Cardiotocography Database, which includes 552 FHR and UC CTG signal pairs.

Model Architecture

The researchers began with the CTG-net network architecture, which convolves the paired FHR and UC input signals temporally before conducting a depthwise convolution to learn the relationship between them. They added several methodological configurations, including architecture and hyperparameter optimization, single input variation, and the addition of clinical metadata.

Pre-processing and Pre-training

To improve data quality, the researchers created a pre-processing pipeline that included inputting missing measurements, random cropping, and additive multiscale noise for data augmentation and downsampling. This generated a large dataset for pre-training and training the models.

Intermittent versus Continuous CTG Use Cases

CTG use comes in two primary formats: intermittent and continuous. In high-resource settings, continuous CTGs are used to monitor fetal heart rate throughout labour. In low-resource settings, intermittent CTGs are often used, covering only about 30 minutes at any point during labour.

The researchers simulated intermittent settings by splitting the 90-minute signals in the dataset into 30-minute signals and training and evaluating the model at different time points. This helped understand how training and evaluating on intermittent time points impacts model performance.

Advertisement

Predicting Objective and Subjective Ground Truth Labels

The researchers used three outcome labels from the dataset:

  • Arterial umbilical cord blood pH: An objective measurement that tracks fetal acidosis, an indication of fetal distress.
  • Apgar score: A subjective measure recorded by a clinician after delivery that reflects the general health of the newborn.
  • Abnormal label: If either Apgar or pH results were abnormal.

Evaluating Model Prediction Robustness

The researchers performed several comparisons to evaluate model performance, including:

  • Performance on the dataset versus the state-of-the-art CTG-net model.
  • Apgar versus pH classification tasks.
  • FHR-only versus FHR+UC.
  • Base model using the last 30 minutes of labour (continuous case) versus intermittent measurements.
  • Base model of FHR+UC versus FHR+UC+Metadata.
  • Subgroup performance of the base model (FHR+UC) with subgroups determined by binarizing clinical metadata.

The results showed that combining FHR+UC achieved the highest model performance for both pH and Apgar classification. The pre-training step enabled the highest model performance, and adding clinical metadata slightly improved model performance for pH but less so for Apgar.

Subgroup Evaluations

The researchers found significant differences in baseline performance between subgroups with frequent and infrequent UC signals gaps for pH prediction and for subgroups with frequent and infrequent FHR signal gaps for Apgar prediction. With metadata, the performance disparities observed with pH prediction were mitigated. However, including metadata increased the AUROC performance disparities for demographic and clinical-related subgroups on this task.

Open CTG Model for Research Use Cases

The researchers are currently exploring open-sourcing their models, hoping that other researchers and stakeholders can build on this work with their own datasets to evaluate it for their clinical use cases.

Limitations and Future Work

The study had limitations that constrain the generalizability of the findings. Future investigations should involve a larger and more diverse dataset sourced from maternity centers worldwide, encompassing varied clinical contexts, demographics, and outcomes. Additionally, further work is needed to understand how such prediction algorithms can be optimally integrated into clinical workflows to improve neonatal outcomes.

The Promise of AI in Fetal Monitoring

The development and evaluation of deep learning models for CTG interpretation hold immense promise for improving fetal monitoring and neonatal outcomes. By leveraging AI, healthcare providers can gain objective interpretation assistance, reducing the burden and potentially improving fetal outcomes.

Advertisement

Comment and Share:

What are your thoughts on the future of AI in fetal monitoring? Do you think AI can revolutionise the field and improve neonatal outcomes? Share your experiences and thoughts in the comments below. Don’t forget to subscribe for updates on AI and AGI developments and join the conversation!

Author


Discover more from AIinASIA

Subscribe to get the latest posts sent to your email.

Continue Reading
Advertisement

Life

10 AI Prompts to Turn Your Pet Photos Into Phone Wallpapers (Plus Freebies!)

Discover 10 amazing prompts for how to create AI-powered phone wallpapers with AI platform, Midjourney. Unleash your inner artist!

Published

on

AI pet phone wallpapers

TL;DR

  • Learn how to turn your pet photos into stunning AI-generated wallpapers for your phone.
  • Includes 10 creative AI prompts, from fantasy hero pets to minimalist silhouettes.
  • Tips on using tools like MidJourney or DALL-E, optimising images, and sharing creations.

AI Pet Phone Wallpapers

Whether you’re a proud pet parent or simply adore animals, turning your furry (or feathery) friend into stunning AI-generated art is an exciting way to personalise your phone. With AI tools like MidJourney, DALL-E, or Stable Diffusion, you can create incredible, artistic renditions of your pet that fit your style and aesthetic.

In this article, we’ll share 10 unique AI prompts tailored to transform your pet photos into captivating works of art for your phone. You’ll also find detailed tips for getting started and optimising your wallpapers for the perfect fit.

Handy hint: Like the look of one of these images? Tap and hold on your phone and choose ‘Set as Wallpaper’ (or a similar option)—AIinASIA Editor

Why Create AI Phone Wallpapers of Your Pet?

  • Celebrate Your Pet: Turn your beloved companion into a piece of art you can cherish every time you unlock your phone.
  • Personalisation: Make your phone truly unique with wallpapers that reflect your style and personality.
  • Explore Creativity: Experiment with different artistic styles—be it fantasy, minimalist, or futuristic—and see your pet like never before.

Getting Started with AI-Generated Pet Wallpapers

To create AI art that truly stands out, follow these simple steps:

  1. Choose Your AI Tool: Platforms like MidJourney, DALL-E, or Stable Diffusion are perfect for generating artistic images from prompts. Look for features like image-to-image transformations, which allow you to upload your pet’s photo as a reference.
  2. Pick a High-Quality Photo: AI tools perform better with clear, detailed images. Choose a well-lit photo where your pet’s features—eyes, fur, feathers—are easy to distinguish.
  3. Tailor the Aspect Ratio: Most phones use a 9:16 aspect ratio (e.g., 1080×1920 pixels). Many AI platforms let you specify image dimensions before generating your artwork.
  4. Experiment with Prompts: Use the creative prompts provided below, or modify them to suit your pet’s personality and your artistic preferences.
  5. Finalise and Optimise: After generating the artwork, use tools like Canva or Snapseed to crop, adjust colours, or apply finishing touches.

10 AI Prompts for Stunning Pet Wallpapers

1. Fantasy Realm Hero

Inspiration: Imagine your dog as a knight or your cat as a sorceress—ready to embark on epic adventures!

Prompt: “A majestic portrait of [your pet’s species] wearing intricate fantasy armour, glowing eyes, standing atop a mountain, ethereal lighting, detailed fur texture, high-resolution phone wallpaper.”

pet phone wallpapers

2. Watercolour Painting

Inspiration: A soft, elegant depiction that works beautifully as a lock screen.

Prompt: “A hand-painted watercolour portrait of [your pet’s breed], soft pastel tones, white background, minimalistic style, high-resolution for phone wallpaper.”

pet phone wallpapers

3. Neon Cyberpunk

Inspiration: Perfect for sci-fi fans who love bold, high-energy designs.

Prompt: “A cyberpunk-style portrait of [your pet’s name], glowing neon colours, city skyline background, futuristic armour, vibrant lighting effects, high-resolution wallpaper.”

pet phone wallpapers

4. Royal Portrait

Inspiration: Imagine your pet as king, queen, or duke—complete with a gilded frame.

Prompt: “A regal portrait of [your pet’s species] dressed in Victorian royal attire, soft studio lighting, intricate clothing details, realistic painting, designed for a phone wallpaper.”

pet phone wallpapers

5. Dreamy Galaxy Explorer

Inspiration: Great for cosmic dreamers who want their pet among the stars.

Prompt: “A whimsical depiction of [your pet’s breed] floating in a galaxy, surrounded by planets and stars, dreamy pastel colours, cosmic glow, phone wallpaper resolution.”

pet phone wallpapers

6. Comic Book Superhero

Inspiration: Give your pet a bold, action-packed look with superhero vibes.

Prompt: “A dynamic comic book illustration of [your pet’s species] in superhero attire, bold colours, action pose, halftone effects, phone wallpaper resolution.”

pet phone wallpapers

7. Pop Art Extravaganza

Inspiration: Bright, eye-catching art for fans of modern and retro styles.

Prompt: “A vibrant pop art portrait of [your pet’s species], bold colours, abstract shapes in the background, Andy Warhol-inspired design, phone wallpaper size.”

pet phone wallpapers

8. Winter Wonderland

Inspiration: A seasonal touch for those who love winter themes.

Prompt: “A serene portrait of [your pet’s name] sitting in a snowy forest, soft light, frost-covered trees, realistic fur details, high-resolution phone wallpaper.”

pet phone wallpapers

9. Minimalist Silhouette

Inspiration: Perfect for a sleek, understated home screen.

Prompt: “A minimalist black-and-white silhouette of [your pet’s breed], clean lines, solid colour background, modern art style, high-resolution for phone wallpaper.”

10. Anime-Inspired Portrait

Inspiration: A playful depiction of your pet in a whimsical anime world.

Prompt: “An anime-style illustration of [your pet’s species], big expressive eyes, pastel colours, soft background, whimsical vibe, high-resolution phone wallpaper.”

pet phone wallpapers

Pro Tips for AI Success With AI-Generated Art

  1. Fine-Tune Prompts: Be specific about the colours, textures, and settings to get the best results.
  2. Use Advanced Features: Many AI tools offer style transfer or reference image settings—use them for a more customised look.
  3. Generate Variations: Create multiple versions of the same prompt to explore different artistic outcomes.

Sharing Your Creations

Once you’ve created your masterpiece, share it! Post your custom wallpapers on Instagram, Pinterest, or your favourite social platforms. Tag AIinASIA to inspire fellow pet lovers and show off your AI creations.

Final Thoughts on AI Personalised Wallpapers

AI tools are opening up endless possibilities for creating personalised and unique art. With these prompts, you can transform your pet into a star on your phone screen—whether they’re ruling a fantasy kingdom, exploring space, or posing as a royal icon.

Get started today and let your imagination run wild. Don’t forget to drop a comment or tag us with your favourites—your pet might just become the next AI art sensation!

You may also like:

Advertisement

Author


Discover more from AIinASIA

Subscribe to get the latest posts sent to your email.

Continue Reading

Business

Everyday Hacks with Google and Microsoft AI Tools

Learn how AI productivity tools like Gemini and Copilot are transforming everyday workflows—and why a human touch still matters.

Published

on

AI productivity tools

TL;DR

  • Google’s Gemini and Microsoft’s Copilot are reshaping productivity with advanced AI features integrated into their suites.
  • Gemini offers features like email summarisation, smart replies, and data analysis through Google Workspace, starting at $20 per user/month.
  • Copilot, integrated into Microsoft 365, helps with document drafting, meeting automation, and data visualisation, starting at $30 per user/month.
  • Both tools are designed to streamline repetitive tasks, improve collaboration, and spark creativity.

Embracing AI Productivity Tools In Your Daily Workflows

If you’ve ever wished for a workday where you can skip the admin grind and focus on the good stuff—creative ideas, strategy, and, dare I say it, an actual lunch break—Google’s Gemini and Microsoft’s Copilot are here to make it happen. These AI productivity tools are redefining how we work, play, and (let’s be honest) survive in today’s fast-paced, tech-driven world. Let’s explore how they’re transforming workflows and what’s in it for you.


Gemini: Google’s AI Assistant for Productivity Bliss

Launching in January 2025, Google’s Gemini is set to become the ultimate assistant for users of Google Workspace (yes, that’s what used to be G Suite). With AI weaved into Gmail, Docs, Sheets, Slides, and even Meet, Gemini aims to make the everyday less tedious and a whole lot smarter.

Gemini’s Features You’ll Actually Use

  • In Gmail: Writing emails? Gemini can do it for you. Endless threads? Summarised. Replies? Smart and context-aware.
  • In Google Docs: It’s like having a co-writer on speed dial, helping draft, summarise, or even whip up images for your content.
  • In Google Sheets: Need data entered or trends spotted? Gemini has you covered with its enhanced Smart Fill feature.
  • In Google Slides: Presentation nerves? Gemini’s image creation and background removal tools ensure your slides look the part.
  • In Google Meet: From automatic note-taking to real-time translated captions, meetings just got a lot less exhausting.

Subscription Plans

  • Gemini Business: $20 per user/month with basic features and usage limits.
  • Gemini Enterprise: $30 per user/month, giving you premium features and more elbow room to explore Gemini’s capabilities.

Why Gemini Stands Out

As Upcurve Cloud says, Gemini doesn’t just help you “get things done”—it’s about doing them better. Whether it’s cutting through email chaos or polishing your next big presentation, Gemini’s AI helps you reclaim time for what matters. Bonus points for Google’s enterprise-grade security keeping your data safe.

Real-World Example: Gemini as a Productivity Tool

Creospark has many examples of using Copilot, such as Adore Me, a lingerie and sleepwear company, has leveraged Gemini to dramatically reduce the time spent on content creation—reducing over 35 hours of work creating product descriptions to just 30 minutes using Gemini, illustrating Gemini’s powerful capability to automate and accelerate content generation tasks, resulting in substantial time savings for e-commerce businesses.


Microsoft Copilot: The AI Powerhouse for Work Smarts

Not to be outdone, Microsoft has embedded Copilot into its Microsoft 365 ecosystem, turning apps like Word, Excel, PowerPoint, and Teams into supercharged productivity hubs. If you’ve ever dreamt of an assistant who “just gets it”, Copilot is the next best thing.

What Copilot Brings to the Table

  • In Word: Copilot drafts, edits, and summarises like a champ, cutting hours off document creation.
  • In Excel: Data insights? Charts? Done and done. Let Copilot turn rows of numbers into actionable insights.
  • In PowerPoint: Presentations are no longer a slog—Copilot suggests content, designs, and layouts tailored to your needs.
  • In Outlook: It sorts your inbox, drafts replies, and highlights what actually matters.
  • In Teams: Catch-up summaries, action items, and real-time transcription make meetings less painful.

Subscription Options

  • Commercial Plans: $30 per user/month for enterprises and large teams.
  • Personal Plans: A modest $3/month price hike adds Copilot to Microsoft 365 Personal and Family subscriptions.

The Copilot Advantage

Copilot doesn’t just automate—it elevates. Imagine crafting a polished report in minutes or summarising a day’s worth of meetings at the click of a button. With its seamless integration across Microsoft 365 apps, Copilot ensures everything is right where you need it, when you need it.

Real-World Example: Copilot as a Productivity Too

Hargreaves Lansdown, a financial services company, has successfully used Copilot to streamline their workflow—financial advisers used Copilot to generate meeting summaries and annual review reports for clients, streamlining their workflow, demonstrating how Copilot can significantly improve efficiency in document creation and client communication in the financial sector.

Advertisement

The Real Impact of AI Productivity Tools

Both Gemini and Copilot are more than just fancy add-ons; they’re reshaping how work gets done. These AI tools are tackling the biggest pain points in the modern workplace:

  1. Saving Time: Whether summarising emails or automating reports, these tools free up hours in your day.
  2. Enhancing Creativity: With AI handling the grunt work, you can focus on big ideas and polished results.
  3. Improving Collaboration: From shared notes to instant insights, teams work better together.
  4. Boosting Decision-Making: Advanced data analysis gives you the insights you need to act fast.

A Word of Caution: Trust but Verify

AI tools like Gemini and Copilot can supercharge your productivity, but they’re not infallible. While they’re brilliant for handling the grunt work, they’re no replacement for human oversight and judgement.

Why Double-Checking Matters

  1. Accuracy: AI-generated outputs can occasionally misinterpret context or make errors.
  2. Appropriateness: AI doesn’t always understand nuances like brand voice, cultural sensitivity, or workplace-specific policies.
  3. Over-Reliance: Depending solely on AI may reduce your own engagement with the material, leading to missed opportunities or oversights.

Practical Tips for Responsible AI Use

  • Review Outputs Carefully: Always double-check AI-generated summaries, drafts, or data visualisations before sharing or implementing them.
  • Add the Human Touch: Make edits to ensure the content aligns with your goals, tone, and audience.
  • Use AI as a Starting Point: Let the AI handle the groundwork, but refine the output with your expertise.
  • Train AI Over Time: Where possible, provide feedback to improve the tool’s understanding of your preferences.

By staying engaged and bringing a human touch to AI-generated content, you can ensure your work remains accurate, professional, and impactful.


What’s Next? The AI Productivity Revolution

AI tools like Gemini and Copilot are undoubtedly paving the way for smarter, more efficient workplaces. But let’s remember that these tools are here to complement human skills, not replace them. By using AI responsibly—embracing its strengths while applying critical judgement—we can unlock its full potential without sacrificing quality or trust.


Your Turn

How do you strike the balance between relying on AI and adding your personal touch? Do you think we risk becoming too dependent on these tools? Share your thoughts in the comments below—we’d love to hear your perspective!

And don’t forget to subscribe for updates on AI and AGI developments here. Let’s build a community of tech enthusiasts and stay ahead of the curve together!

You may also like:

Advertisement

Author


Discover more from AIinASIA

Subscribe to get the latest posts sent to your email.

Continue Reading

Life

ChatGPT’s Thirst for Water

Explore the connection between AI water usage and wildfire management.

Published

on

AI water usage

TL;DR:

  • AI data centres consume millions of gallons of water annually for cooling purposes.
  • Reducing ChatGPT usage can help conserve water needed to combat wildfires in cities.
  • Companies are exploring alternative cooling methods and data centre locations to minimise water usage.

In the midst of the blazing wildfires in the greater Los Angeles area, residents are eager to contribute to firefighting efforts in any way possible. While donations and volunteering are top of mind, a surprising call to action has emerged on social media: refrain from using ChatGPT.

Wait! What does ChatGPT have to do with water conservation and wildfire management?

This provocative suggestion raises the question: what does ChatGPT have to do with water conservation and wildfire management? Let’s dive in and explore the hidden connection between AI and water usage.

The Water-Guzzling Reality of AI Data Centres

AI platforms like ChatGPT rely on massive data centres to function. These data centres are packed with high-performance computer chips that process user queries, generating immense heat. To prevent servers from overheating and crashing, sophisticated cooling systems are employed, many of which rely on water.

“Because the computers and chipsets that power servers are so densely packed, they generate an incredible amount of heat. Running complex AI applications like ChatGPT requires immense amounts of computing power, which generate lots of heat 24 hours a day.”
HP Newquist, Artificial Intelligence Historian and Author
Tweet

The sheer volume of water consumed by these data centres is staggering. Some large facilities use millions of gallons of water annually to absorb and dissipate heat through cooling towers or evaporative cooling methods. This water usage can have a significant impact on local water supplies, especially in drought-prone areas.

The Collective Impact of Individual AI Usage

While it may seem like an individual ChatGPT query has a negligible effect on water consumption, the collective impact of millions of users interacting with AI services adds up quickly. By making small changes, such as delaying non-urgent AI tasks, Angelenos can contribute to conserving water and energy for critical firefighting efforts.

“Data centres that host AI services like ChatGPT continuously consume power and indirectly rely on water for cooling. By making small changes like delaying non-urgent AI tasks, Angelenos can contribute to ensuring that water and energy remain available where they’re needed most during these high-risk periods.”
Daniel Kearney, Chief Technology Officer, Firmus Technologies
Tweet

Exploring Alternative Cooling Solutions

As AI continues to advance, finding sustainable solutions for data centre cooling is becoming increasingly important. Some companies are exploring alternative cooling methods and data centre locations to minimise water usage.

One promising approach is immersion cooling, which involves submerging hardware in a special cooling liquid. This method can significantly reduce water consumption compared to traditional cooling techniques. Additionally, some AI companies are opting to build data centres in colder regions, where lower ambient temperatures can help regulate server temperatures more efficiently.

Advertisement
“Many companies are building server centres near polar regions or in northern countries with low ambient temperatures.”
HP Newquist, Artificial Intelligence Historian and Author
Tweet

By investing in these alternative cooling solutions, AI companies can help mitigate their impact on water supplies and contribute to more sustainable technology practices.

The Fire Line: AI Water Usage and Wildfire Management

Current wildfires are exacerbated by the region’s ongoing drought, placing a strain on the state’s water supply. By reducing AI usage, residents can help free up water resources needed to combat these devastating fires.

“With California already experiencing an ongoing drought, the water necessary to fight these fires is further straining the state’s water supply. Water is a finite resource, and cutting back on AI use will have direct impacts on the state’s water availability and ability to respond to this climate disaster.”
Mia Montoya Hammersley, Assistant Professor, Environmental Law, Vermont Law and Graduate School
Tweet

As AI continues to permeate our daily lives, it is essential to recognise the hidden environmental costs associated with these technologies. By taking proactive steps to minimise our AI water footprint, we can help preserve valuable resources and support wildfire management efforts in major cities.

Wrapping Up: The Future of AI and Water Conservation

The connection between ChatGPT, water usage, and wildfire management serves as a stark reminder of the interconnectedness of technology and the environment. As AI continues to evolve, it is crucial for both companies and consumers to prioritise sustainable practices and invest in alternative cooling solutions.

By doing so, we can help minimise the environmental impact of AI and ensure that valuable water resources are allocated where they are needed most.

Comment and Share:

Have you ever considered the water footprint of your AI usage? How do you think we can encourage more sustainable technology practices? Share your thoughts and experiences in the comments below, and don’t forget to subscribe for updates on AI and AGI developments here. Let’s foster a community of mindful tech enthusiasts, working together to promote a more sustainable future!

You may also like:

Advertisement

Author


Discover more from AIinASIA

Subscribe to get the latest posts sent to your email.

Continue Reading

Trending

Discover more from AIinASIA

Subscribe now to keep reading and get access to the full archive.

Continue reading