HealthAI @ Google • Prev: Head of AI @ Need, ML for Healthcare @ Google Brain, PhD @ Princeton (ML+Comp Neuro), Google PhD Fellow
- Tweets 44
- Following 1,212
- Followers 188
- Likes 85
New perspective in @NatureMedicine from our AMIE team at @Google!
Prospective evidence for conversational medical AI is hard, but non-negotiable
🔗 nature.com/articles/s41591-0…
🔗 rdcu.be/Pgr1WwyjF9Ca
Wonderful working together with @HardyShakerman @apalepu13 @drethangoh @AdamRodmanMD @AmiParekh10 @AvinatanH48021 @ymatias and many on this piece!
Trust in medical AI hasn't been earned yet.
But it potentially can be with randomized trials conducted in real world medicine
@NatureMedicine @HardyShakerman @cameronphchen
nature.com/articles/s41591-0…
1
3
188
A clinical consultation begins long before words are spoken — from a patient’s gait to their facial cues.
Today, our team at @Google advances AMIE towards real-time video consultations, reaching expert-level performance for the first time.
1
5
11
980
A few links to learn more:
📄 Paper preprint: goo.gle/3TOx4zG
🔗 Blog: research.google/blog/advanci…
📺 Demo Video (2mins): piped.video/ISlPOC0qsrY
1
3
78
Incredible collaborations across @GoogleResearch @GoogleDeepMind @GoogleForHealth!
Matthew Thompson, @_cjpark, @timstro, @valentinlievin, @Romes13, @AkshayGoelMD, @teyaberg, @BediSuhana42170, @meetshahdev, Pavel Dubov, Liviu Panait, Toshiyuki Fukuzawa, @SRSchmidgall, Craig Schiff, Joseph Xu, @aliya_rysbek
, Yana Lunts, @JanFreyberg, @rhemenway, @sunnyvirmani, David Racz, Carey Radebaugh, @joelle_barral, Kavi Goel, @dr2w, Katherine Chou, @AvinatanH48021, @ymatias, James Manyika, Gregory Wayne, @taotu831, @yun_liu, @drethangoh, Christina Chen, @RyutaroTanno
1
6
241
New @Nature study from our team @Google
on AMIE: our conversational medical AI evolving from diagnostic dialogue to managing disease over time.
Congrats to the teams, especially @apalepu13 @valentinlievin @HardyShakerman
@GoogleResearch @GoogleDeepMind
Thrilled to share that our paper on AMIE—our conversational medical AI—is now published in @Nature. We wanted to tackle one of the most challenging areas in clinical AI: management reasoning over multiple patient visits.
nature.com/articles/s41586-0…
11
1,620
Cameron Po-Hsuan Chen retweeted
Does medical AI really work in the real-world? It needs to be assessed carefully and responsibly.
We will be launching a first-of-its-kind nationwide randomized study with Included Health to evaluate AI in real-world virtual care to better understand its capabilities & limitations. This study is informed by years of foundational research across Google, investigating the capabilities required for a helpful & safe medical AI.
25
140
719
137,673
Today, @Google is announcing a strategic partnership with Included Health on a nationwide randomized study of AI in virtual care!
Official announcement: research.google/blog/collabo…
This is all possible because of the wonderful partnership across Included Health @GoogleResearch @GoogleDeepMind @GoogleForHealth and Google Platforms & Devices!
1
1
6
907
AI holds so much potential in transforming how people manage health, navigate the system, and interact with providers.
This study is informed by years of foundational research, including AMIE (lnkd.in/eY5MaReQ), Personal Health Agent (lnkd.in/e9xn73N4), and Wayfinding AI (lnkd.in/eM8dUV-Q).
1
69
with exceptional partners:
@AmiParekh10 @nupursrivastava @owentripp
@VibinRoy @HardyShakerman @sunnyvirmani
@AkshayGoelMD @apalepu13 @timstro @SRSchmidgall
@taotu831 @rhemenway @joelle_barral @dr2w @AvinatanH48021 @mdhowellmd @ymatias and many many more!
51
September is #ProstateCancer awareness month so we decided to share more about the work we're doing at Google Health and how we're working toward the shared mission of catalyzing the adoption of human-centered #AI in health! Very fortunate to work with such a fantastic group!
We already see the potential benefits of AI-driven prostate cancer research. From informing treatment plans to improving the efficacy & confidence of diagnoses, researchers Dave and Cameron share updates on our exciting work in this vital area. Read more: goo.gle/3re9E52
1
After 4+ years in the making, super excited for this critical translational step of the program! Very much looking forward to working with @verilylifesci and @LUMEA_DDx colleagues to move the field forward!
📣 Announcing a new partnership with @LUMEA_DDx to advance #digitalpathology using precision #AI algorithms for #prostatecancer detection, with the goal of improving patient outcomes. Get the details: bit.ly/3wbJ2p7
ALT Verily + Lumea partnership
2
10
Cameron Po-Hsuan Chen retweeted
Prostate cancer grading is a clinical problem that has attracted substantial ML research because it involves visual pattern assessment, a task well-suited for computer vision models. Learn about the PANDA challenge and how it has accelerated such research↓goo.gle/3HNvByB
4
50
175
Cameron Po-Hsuan Chen retweeted
NEW Comment by Po-Hsuan Cameron Chen, Craig H Mermel and Yun Liu: 'Evaluation of #ArtificialIntelligence on a reference standard based on subjective interpretation'. @GoogleHealth
bit.ly/2Zgf8S9
7
8
Cameron Po-Hsuan Chen retweeted
In July 2020, @JAMAOnc, part of @JAMANetwork published "Development and Validation of a Deep Learning Algorithm for Gleason Grading in Prostate Cancer.” This work was co-authored with @DrMahul & Dr. @nielsolson, leading experts and fierce patient advocates goo.gle/3no96Jg
3
8
Cameron Po-Hsuan Chen retweeted
Identifying whether a chest X-ray exhibits abnormalities is critical for patient triage and care. Today we demonstrate a method that uses #ML to rapidly assess and flag potentially abnormal chest X-rays for radiologist review. Read all about it below. goo.gle/3yGZdIQ
2
60
228
New study published!
Accurate assessment of prognosis is critical to patient care. In this study, we showed that AI model is more accurate at predicting outcomes than grading done by pathologists, validated on a large-scale independent dataset.
Predicting prostate cancer specific-mortality with artificial intelligence-based Gleason grading: go.nature.com/3A6U1zG
Read this and other first articles published in @CommsMedicine.
Our latest work in predicting outcomes from colorectal cancer histopathology slides with identification of novel tumor features! Very excited about this direction of using AI to predict unknown signals while understanding the underlying features that are influencing the model.
Our latest research applying #DeepLearning in #pathology is now at @Nature_npj Digital Medicine, describing our approach to predict prognosis in colorectal cancer and identify potentially novel tumor features. Read more at goo.gle/3dthK3I and ↓ goo.gle/2P0MxeP
1
14
A very exciting milestone from the team!
Together with @NorthwesternMed, we're working on a clinical study to explore whether AI models can help to reduce the diagnosis time in the mammography process, which if successful, could improve breast cancer patients' experience. Learn more: goo.gle/2O4Qe2w
1