Loading video...

Video Failed to Load

Go Home

We’ve trained a multimodal AI model to turn routine pathology slides into spatial proteomics, with the potential to reduce time and cost while expanding access to cancer care.

satyanadella's profile picture

Satya Nadella

6,836,046 subscribers

2,802,714 views • 3 months ago •via X (Twitter)

Health & WellnessScience & Technology
LIVE

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

Helping to address barriers to cancer screenings and care, reducing disparities and expanding access are just some of the ways we work every day to shape the outcomes for people with cancer.
0:29

Helping to address barriers to cancer screenings and care, reducing disparities and expanding access are just some of the ways we work every day to shape the outcomes for people with cancer.

Merck

29,382,947 views • 1 year ago

What if a foundation model could align histology, spatial biology & clinical data to reveal latent biomedical insights? 🚀 Introducing Haiku — a tri-modal foundation model trained on 26.7M+ spatial proteomics patches with matched H&E and clinical text, aligned in one shared embedding space. 📄 🧵👇
1:10

What if a foundation model could align histology, spatial biology & clinical data to reveal latent biomedical insights? 🚀 Introducing Haiku — a tri-modal foundation model trained on 26.7M+ spatial proteomics patches with matched H&E and clinical text, aligned in one shared embedding space. 📄 🧵👇

Zhi Huang

13,236 views • 2 months ago

🚨First Lady Casey DeSantis says Florida invests in IVERMECTIN as a possible cure for cancer. “In Florida we have invested over a billion dollars in cancer care and research. We’ve also looked into innovative research like Ivermectin and its potential to try to combat cancer.”
0:33

🚨First Lady Casey DeSantis says Florida invests in IVERMECTIN as a possible cure for cancer. “In Florida we have invested over a billion dollars in cancer care and research. We’ve also looked into innovative research like Ivermectin and its potential to try to combat cancer.”

Chris Nelson 🏝️🇺🇸

60,034 views • 5 months ago

DISCO-MS allows molecular analysis via proteomics and AI after tissue clearing. Now we are in the game to discover the molecular underpinnings of diseases. To this end, we are expanding our proteomics team. Please apply at (distribute😊):
0:58

DISCO-MS allows molecular analysis via proteomics and AI after tissue clearing. Now we are in the game to discover the molecular underpinnings of diseases. To this end, we are expanding our proteomics team. Please apply at (distribute😊):

Ali Max Erturk

31,956 views • 2 years ago

1/n: 🔬We introduce NicheCompass (NC) a graph learning model to build spatial atlases across millions of cells, quantitatively characterize cell niches, infer cellular communication, and map new data into these atlases. NC can be used with various spatial technologies, ranging from cellular to subcellular resolution including even multimodal approaches. We have explored interesting biology by analyzing cell niches spanning from development to cancer. Led by Sebastian Birk in a fun colab with @cartalop: paper:
1:04

1/n: 🔬We introduce NicheCompass (NC) a graph learning model to build spatial atlases across millions of cells, quantitatively characterize cell niches, infer cellular communication, and map new data into these atlases. NC can be used with various spatial technologies, ranging from cellular to subcellular resolution including even multimodal approaches. We have explored interesting biology by analyzing cell niches spanning from development to cancer. Led by Sebastian Birk in a fun colab with @cartalop: paper:

Mo Lotfollahi

61,763 views • 2 years ago

Google presents Still-Moving Customized Video Generation without Customized Video Data Customizing text-to-image (T2I) models has seen tremendous progress recently, particularly in areas such as personalization, stylization, and conditional generation. However, expanding this progress to video generation is still in its infancy, primarily due to the lack of customized video data. In this work, we introduce Still-Moving, a novel generic framework for customizing a text-to-video (T2V) model, without requiring any customized video data. The framework applies to the prominent T2V design where the video model is built over a text-to-image (T2I) model (e.g., via inflation). We assume access to a customized version of the T2I model, trained only on still image data (e.g., using DreamBooth or StyleDrop). Naively plugging in the weights of the customized T2I model into the T2V model often leads to significant artifacts or insufficient adherence to the customization data. To overcome this issue, we train lightweight Spatial Adapters that adjust the features produced by the injected T2I layers. Importantly, our adapters are trained on "frozen videos" (i.e., repeated images), constructed from image samples generated by the customized T2I model. This training is facilitated by a novel Motion Adapter module, which allows us to train on such static videos while preserving the motion prior of the video model. At test time, we remove the Motion Adapter modules and leave in only the trained Spatial Adapters. This restores the motion prior of the T2V model while adhering to the spatial prior of the customized T2I model. We demonstrate the effectiveness of our approach on diverse tasks including personalized, stylized, and conditional generation. In all evaluated scenarios, our method seamlessly integrates the spatial prior of the customized T2I model with a motion prior supplied by the T2V model.
1:43

Google presents Still-Moving Customized Video Generation without Customized Video Data Customizing text-to-image (T2I) models has seen tremendous progress recently, particularly in areas such as personalization, stylization, and conditional generation. However, expanding this progress to video generation is still in its infancy, primarily due to the lack of customized video data. In this work, we introduce Still-Moving, a novel generic framework for customizing a text-to-video (T2V) model, without requiring any customized video data. The framework applies to the prominent T2V design where the video model is built over a text-to-image (T2I) model (e.g., via inflation). We assume access to a customized version of the T2I model, trained only on still image data (e.g., using DreamBooth or StyleDrop). Naively plugging in the weights of the customized T2I model into the T2V model often leads to significant artifacts or insufficient adherence to the customization data. To overcome this issue, we train lightweight Spatial Adapters that adjust the features produced by the injected T2I layers. Importantly, our adapters are trained on "frozen videos" (i.e., repeated images), constructed from image samples generated by the customized T2I model. This training is facilitated by a novel Motion Adapter module, which allows us to train on such static videos while preserving the motion prior of the video model. At test time, we remove the Motion Adapter modules and leave in only the trained Spatial Adapters. This restores the motion prior of the T2V model while adhering to the spatial prior of the customized T2I model. We demonstrate the effectiveness of our approach on diverse tasks including personalized, stylized, and conditional generation. In all evaluated scenarios, our method seamlessly integrates the spatial prior of the customized T2I model with a motion prior supplied by the T2V model.

AK

40,474 views • 2 years ago

For patients diagnosed with cancer, time matters. Discover how we're helping reduce the time from identification of lung cancer to intervention:
0:22

For patients diagnosed with cancer, time matters. Discover how we're helping reduce the time from identification of lung cancer to intervention:

Intuitive

12,032 views • 2 years ago

Cancer care across the UK is at breaking point, and people with cancer are facing significant challenges across all areas of their lives. The next UK government must turn the crisis around and develop a plan to revolutionise cancer care.
1:17

Cancer care across the UK is at breaking point, and people with cancer are facing significant challenges across all areas of their lives. The next UK government must turn the crisis around and develop a plan to revolutionise cancer care.

Macmillan Cancer Support

94,717 views • 2 years ago

Multimodal AI is here 🤯 GPT-4 can now turn your images into a text file in a snap with the new code interpreter model. Witness the OCR magic in action 🔥
0:29

Multimodal AI is here 🤯 GPT-4 can now turn your images into a text file in a snap with the new code interpreter model. Witness the OCR magic in action 🔥

Shubham Saboo

727,655 views • 3 years ago

Access to quality care shouldn’t depend on your zip code. We’re expanding rural health access through innovations in telemedicine and AI-powered diagnostics—bringing cutting-edge care to every corner of the country. It’s all part of our mission to Make America Healthy Again.
0:46

Access to quality care shouldn’t depend on your zip code. We’re expanding rural health access through innovations in telemedicine and AI-powered diagnostics—bringing cutting-edge care to every corner of the country. It’s all part of our mission to Make America Healthy Again.

Secretary Kennedy

91,053 views • 1 year ago

oh may.. 3D Gaussian Splat is so crazy now last time we need weeks to turn a room into a 3D model.. now with AI 3DGS, you do it in seconds… even zoom into the text and looks perfect give it like 6 months… you can turn a whole city into a full detailed 3D model
0:22

oh may.. 3D Gaussian Splat is so crazy now last time we need weeks to turn a room into a 3D model.. now with AI 3DGS, you do it in seconds… even zoom into the text and looks perfect give it like 6 months… you can turn a whole city into a full detailed 3D model

el.cine

142,328 views • 7 months ago

AI Czar David Sacks: "Technologies like AI—they will expand access to care, cut administrative costs, reduce fraud, and deliver better results for patients and doctors."
0:57

AI Czar David Sacks: "Technologies like AI—they will expand access to care, cut administrative costs, reduce fraud, and deliver better results for patients and doctors."

Trump War Room

33,210 views • 11 months ago

A decade of work comes to life today. Meet Phoenix! As the world’s first AI Care Specialist, Phoenix works alongside human clinicians to make world-class care available anytime, anywhere. Globally, billions of people lack access to high-quality care. That’s why we reinvented the way care is delivered. Now, with Phoenix, the full potential of the AI Care revolution is clear.
1:06

A decade of work comes to life today. Meet Phoenix! As the world’s first AI Care Specialist, Phoenix works alongside human clinicians to make world-class care available anytime, anywhere. Globally, billions of people lack access to high-quality care. That’s why we reinvented the way care is delivered. Now, with Phoenix, the full potential of the AI Care revolution is clear.

Sword Health

22,821 views • 2 years ago

The health of rural America is a top priority. We’re expanding access to care through more community health centers. New innovations like AI and emerging technologies can help ensure we reach even the most remote communities. Every American deserves access to quality care, no matter their zip code.
2:02

The health of rural America is a top priority. We’re expanding access to care through more community health centers. New innovations like AI and emerging technologies can help ensure we reach even the most remote communities. Every American deserves access to quality care, no matter their zip code.

Secretary Kennedy

55,559 views • 1 year ago

Today, we're introducing PlayHT2.0 - our new Conversational Text-to-Voice AI Model that's trained and built to generate humanlike conversations across languages with <1s latency. Sign up for beta access -
1:19

Today, we're introducing PlayHT2.0 - our new Conversational Text-to-Voice AI Model that's trained and built to generate humanlike conversations across languages with <1s latency. Sign up for beta access -

PlayAI

52,378 views • 2 years ago