AI Tool Predicts Cancer Survival from Face Photos 📷

Welcome to another edition of Horizon AI,

A new study suggests that, using AI technology, something as simple as a selfie could help predict a person’s chances of surviving cancer.

Let’s jump into it!

Read Time: 4.5 min

Here's what's new today in the Horizon AI

• Scientists Use AI Facial Analysis to Predict Cancer Survival Outcomes

• OpenAI and the FDA Discuss Using AI in Drug Evaluation

• AI Tutorial: Create professional avatar videos with just ONE photo

• AI Tools to check out

• The Latest in AI and Tech 💡

• AI Findings/Resources

AI News

AI RESEARCH

Scientists Use AI Facial Analysis to Predict Cancer Survival Outcomes

Developed by researchers from Mass General Brigham, the deep learning algorithm FaceAge uses a photo of a person's face to predict biological age and cancer survival outcomes—often outperforming human clinicians in short-term survival predictions.

Details:

• Scientists trained FaceAge on 58,851 photos of presumed healthy individuals from public datasets, then tested it on 6,196 cancer patients using photos taken at the start of radiotherapy.

• According to researchers, “how old someone looks compared to their chronological age really matters.”

• They found that, among the cancer patients, the older the FaceAge compared to their actual age, the worse the survival outcome, even after adjusting for chronological age, sex, and cancer type.

• When paired with FaceAge, clinicians improved their accuracy in predicting six-month survival rates from 61% to 80%.

• There are still some limitations, as researchers acknowledge potential biases in the data and are working to refine the model for broader applications.

This tool has immense potential to enhance clinical decision-making, offering a simple, non-invasive way to assess health risks and guide treatment plans. The scientists believe "its potential goes far beyond cancer care or predicting age" and are now testing the technology on a wider range of patients, as well as assessing its ability to predict other diseases, general health status, and lifespan.

Source

OPENAI

OpenAI and the FDA Discuss Using AI in Drug Evaluation

OpenAI is reportedly in talks with the U.S. Food and Drug Administration to explore how AI could streamline drug evaluation processes. The discussions center around a project called cderGPT, likely named after the FDA's Center for Drug Evaluation and Research (CDER).

Details:

• The collaboration seeks to speed up the final stages of drug evaluations, a process that often takes over a decade to complete.

• A small team from OpenAI, along with representatives connected to Elon Musk's DOGE, have held several meetings with FDA officials, but no formal agreement has been reached so far.

• According to former FDA Commissioner Robert Califf, the agency has been experimenting with AI for several years, noting that “there has always been a quest to shorten review times and a broad consensus that AI could help.”

Concerns remain about the reliability and limitations of AI models in such critical applications. However, if successful, the approach could significantly reduce the time it takes to bring new medications to market, benefiting both patients and the healthcare system.

Source

AI Tutorial

Create professional avatar videos with just ONE photo

With HeyGen’s newest feature, Avatar IV, you can create short, lifelike videos in seconds using just one photo and a script or audio file.

1. Go to the HeyGen website and sign up.

2. On the homepage, look for the Photo to Video with Avatar IV button.

3. Upload your photo.

4. Then, either write your script or upload an audio file. If you’re using a script, choose your preferred voice.

Tip: You can also try uploading a song as your audio input—your avatar will sing it back with realistic lip-sync and expression.

5. Once everything’s ready, click Generate to create your video.

Note: This feature is available to all users. Free users can generate videos up to 10 seconds long, with a limit of 3 videos per month.

AI Tools to check out

⚡ RapidScan: Transform your document workflow with AI automation.

🔍 Aithor: AI-powered research assistant.

👉 Common Room: Know who to target, when to engage, and how to convert with the most informed GTM AI.

📽 LTX-Video: The first DiT-based video generation model that can generate high-quality videos in real-time.

🎨 Recraft: Generative AI design tool for digital illustrations, vector art, icons, and 3D graphics in a uniform brand style.

AI Findings/Resources

👉 Developing with Devin AI

💡 What is Hugging Face and how to use it

📧 Email from Fiverr CEO to his team about AI

👨‍⚖️ AI avatar of shooting victim addresses his killer in court

The latest in AI and Tech

Apple is looking to add AI search engines to Safari

During the U.S. Department of Justice's antitrust trial against Google, Apple executive Eddy Cue revealed that Apple is exploring AI-driven search alternatives, such as Perplexity, OpenAI, and Anthropic, to integrate into Safari.

This initiative comes as Safari's search volumes have declined, marking the first drop in 22 years.

Amazon is developing new AI coding tool 'Kiro'

Kiro is designed as a customizable web and desktop application with multimodal capabilities, allowing developers to input not only text but also visual diagrams and other contextual information to generate code in near real-time by connecting with AI agents.

Meta is reportedly planning to add facial recognition to its AI-powered glasses

Meta is developing new AI-powered smart glasses, codenamed Aperol and Bellini, featuring a 'super-sensing' vision system capable of facial recognition and real-time contextual awareness, allowing the glasses to identify people by name.

WisdomAI secures $23M to tackle AI hallucinations

The startup, founded by former Rubrik co-founder Soham Mazumdar, has raised $23 million in a seed funding round led by Coatue, with participation from Madrona, GTM Capital, and Anthology Fund.

Nvidia CEO Jensen Huang gets first pay raise in a decade

Huang's base salary rose by 49% to $1.5 million, with a 50% bonus hike and $38.8 million in stock awards, bringing his total 2025 compensation to $49.9 million

The raise reflects Nvidia’s explosive growth, driven by its dominance in AI hardware, and brings Huang’s pay more in line with peers.

That’s a wrap!

Thanks for sticking with us to the end! Let’s stay connected on LinkedIn and Twitter.

Not subscribed yet? Sign up here and send it to a colleague or friend!

See you in our next edition!

Gina 👩🏻‍💻