The epicenter of AI and health at MIT | Now on BlueSky: bsky.app/profile/aihealthmit… | Follow us on LinkedIn and Instagram.
Cambridge, MA
- Tweets 2,458
- Following 387
- Followers 7,483
- Likes 3,925
ALT We are deeply saddened by the untimely passing of Octavian-Eugen Ganea — a brilliant scholar, dear friend and colleague, generous mentor, avid mountaineer, and a humble soul. Octavian made significant research contributions to MIT Jameel Clinic during his time as a postdoc, driving the superior quality of research coming from Jameel Clinic. His dedication and passion for impactful work left a lasting mark on the MIT community and beyond and he will be dearly missed by us all.
13
43
224
EquiBind, a geometric deep learning model developed by #JameelClinic researchers @octavianEganea, @HannesStaerk, @BarzilayRegina, Tommi Jaakkola, & @MITChemE's @lucky_pattanaik, is 1200x faster than the fastest computational molecular docking model: bit.ly/3AM6AmI (1/3)
2
38
136
We are excited to announce that our inaugural 2022 Molecular ML conference site is now live! 😃 Registrations & poster submissions are now open: moml22.jclinic.mit.edu/ (Students register for free! 🌟 See More Info+ @ registration)
33
117
"You get a lot of rejections whatever you do...it is very important to stay optimistic and not let any rejection break you down...just continue and get good work done" — Octavian-Eugen Ganea #RememberOctavian #NeverGiveUp
10
112
We are excited to announce the AI for Healthcare Equity Conference, which will take place online on April 12th.
This conference will present machine learning techniques that support fairness, personalization and inclusivity.
Learn more & register here: bit.ly/3pQeuDm
1
59
98
To #RememberOctavian, we will share a series of posts starting today written by Octavian's friends & colleagues. They highlight the transformative research Octavian conducted & include inspiring quotes from an interview he gave on May 5th reflecting on his career. #NeverGiveUp
1
11
96
"#RememberOctavian by Benson Chen: Optimal Transport Graph Neural Networks (OTGNN) was one of the 1st projects that Octavian had worked on when he joined @mit. Inspired by his past research using hyperbolic geometry, Octavian sought new ways of representation & generation...(1/5)
1
20
95
The authors of MolScribe, which uses AI for molecular structure recognition, now present RxnScribe, a model which goes a step further by extracting reaction diagrams & predicting reactions in new diagrams.
Paper: arxiv.org/abs/2305.11845
Code: github.com/thomas0809/RxnScr…
ALT Top row (starting from left): Headshots of Yujie Qian, Jiang Guo, Zhengkai Tu Bottom row (starting from left): Headshots of Connor Coley and Regina Barzilay Bottom image depicts a diagram with various structural formulas showing how RxnScribe feeds into MolScribe and/or OCR.
2
18
97
9,482
Congratulations to #JameelClinic PI @cwcoley for winning the @OpenEyeSoftware Outstanding Junior Faculty Award in Computational Chemistry! 👏 acscomp.org/awards/the-comp-…
ALT Portrait photo of person staring directly into the camera.
3
8
85
We are a group of MIT faculty and students who are collaborating on developing machine learning methods for finding promising antiviral molecules for COVID-19.
Read more about it here: bit.ly/3b8earK
3
66
82
Jameel Clinic Faculty Lead in #AI @BarzilayRegina and @MIT PhD student @YalaTweets are transforming the future of mammograms using artificial intelligence.
Read more in the latest from the @washingtonpost
washingtonpost.com/technolog…
2
23
81
Severely ill patients with COVID-19 can succumb to bacterial pneumonia as a secondary infection.
We are announcing an open task for predicting activity against the bacteria Pseudomonas aeruginosa, which causes nasty lung infections.
Learn more here: aicures.mit.edu/tasks
1
70
79
AI4Science is a new & growing interdisciplinary field covering a range of scientific areas. "We hope this piece makes it easier to get into one of the subfields & gets even more researchers as excited about the area as we are." — @HannesStaerk
Learn more: arxiv.org/pdf/2307.08423.pdf
ALT A circular graphic depicting AI and its use cases in Quantum, Atomistic, and Continuum Systems.
1
22
82
16,847
Remembering the life of a brilliant researcher and beloved colleague: Octavian-Eugen Ganea bit.ly/3nQQH6U
1
17
73
Congratulations to #JameelClinic AI faculty lead @BarzilayRegina on being inducted into @theNAEng! This immense professional achievement is awarded to those who have distinguished themselves through extraordinary technical accomplishments & leadership. nae.edu/289843/NAENewClass20…
ALT Regina Barzilay sitting on a bench, linking arms with her son.
3
13
73
18,980
Molecules are commonly depicted as 2D images in literature w/ significant variation in drawing styles/conventions, making it difficult for machines to recognize their structures. #JameelClinic researchers @Yujie_Qian @jiangfeng1124 @ZhengkaiTu @zli11010 w/ PIs @cwcoley &...🧵 1/4
ALT Headshots of researchers (top row left to right): Yujie Qian, Jiang Guo, Zhengkai Tu, (bottom row left to right) Zhening Li, Connor Coley, Regina Barzilay
1
10
67
11,885
Mark your calendars because #MoML is back @MIT on Nov. 8! Submit 2-page short papers until Oct. 13 for a chance to win the @octavianEganea Prize for Best Paper here: moml.mit.edu/
🌟Register now! (free for accepted papers + a waitlist for free registrations)
ALT Graphic that reads "MoML Molecular Machine Learning Conference Wednesday, November 8, 2023 Koch Institute @ MIT"
1
23
70
18,357
For the third session of the PETML Workshop, Tommi Jaakkola of @MIT_CSAIL will discuss "Machine Learning Approaches for Drug Discovery" followed by a panel with Regina Barzilay, Rafael Gomez-Bombarelli, Connor Coley, Jim Collins, & Alex Shalek.
Register: petml.mit.edu
1
14
62
#RememberOctavian by @GabriCorso: GeoMol is the first of Octavian's transformative works in modeling the 3D structure of molecular systems. It focuses on molecular conformer generation, a fundamental task in computational chemistry that consists of predicting possible 3D... (1/4)
1
11
62
AICures team releases top SARS-CoV-2 antiviral molecules from the @broadinstitute Library. Read about our new adaptive invariance algorithm here: bit.ly/2Y29Irf
31
62
We are excited to announce the Perturbations, Therapeutics, and Machine Learning Workshop, happening on September 28th.
Co-organized by both Jameel Clinic and the Eric and Wendy Schmidt Center @broadinstitute.
Register and view the full agenda here: petml.mit.edu/
17
56
While ML tools like AlphaFold can now predict protein structures, the same can't be said for metabolite structures. @samgoldman19, @jeremyWohlwend, @mstrazar07, Guy Haroush, Ramnik Xavier, & #JameelClinic PI @cwcoley devised a new computational method for the annotation of... 1/3
ALT Headshots (starting from top left, clockwise) of Samuel Goldman, Jeremy Wohlwend, Martin Stražar, Guy Haroush, Ramnik J. Xavier, and Connor W. Coley
2
10
56
7,681
Assistant prof. Connor Coley (@cwcoley) is developing data-driven, computational tools that would be able to predict molecular behavior and learn from both successes and mistakes.
Read more about his work here: news.mit.edu/2021/using-comp…
12
49
Recent research identified 200+ drugs with efficacy against SARS-COV-2, in addition to 100+ others already in clinical trials. We analyzed multiple screening results, and found significant divergences. How can AI models utilize this data?
Read more here: bit.ly/3dodKPm
2
32
54
12 awardees will be funded for their groundbreaking projects combining the power of AI and health! We hope this long-standing collaboration between @MIT and @MassGenBrigham will contribute to the betterment of patient health outcomes and AI-adoption by healthcare systems.
ALT Top row (left to right): Dimitris Bertismas, Bharti Khurana, Marzyeh Ghassemi, David Bates Middle row (left to right): Polina Golland, Sarah Friskan, Dina Katabi, Aleksandar Videnovic Bottom row (left to right): Collin Stultz, Elazer Edelman, Ashia Wilson, Mengyu Wang
26
15
58
15,170
Congrats to Regina Barzilay, Jameel Clinic AI Faculty Lead, on winning the The Wallace H. Coulter Lectureship Award.
This award recognizes an individual who has made an impact on education, practice, research in lab medicine or patient care.
Read more: aacc.org/community/merit-awa…
2
11
51
Where is the future of wearable technology headed?🔮 #JameelClinic faculty lead Jim J. Collins, H. Ceren Ateş, Peter Nguyen, Laura Gonzalez-Macia, @EdenMoralesNar1, Firat Güder, & @CanDincer83 wrote a review on wearable devices for @NatRevMater looking @ the current state...(1/2)
1
8
51
"In the end, nobody's gonna solve the mysteries of the universe. We are all pretty small, right? Like small seconds into the infinity of the universe...it makes a lot of sense to stay humble." — Octavian-Eugen Ganea #RememberOctavian #NeverGiveUp
7
50
#RememberOctavian by @HannesStaerk: With EquiBind, Octavian follows up on his novelty-laden paper, EquiDock, & adapts the method for predicting how drug-like molecules attach to proteins. Predicting the 3D structure in which a small molecule binds to a protein is a core... (1/5)
1
6
52
We are delighted to share the exciting news that Regina Barzilay has been elected to the prestigious @theNAMedicine in recognition of her groundbreaking contributions to the field of health and medicine through innovative machine learning applications.
bit.ly/MITNAM2023
1
9
51
5,004
We would like to welcome @MarzyehGhassemi as a new 2021 PI at Jameel Clinic.
Her team's project will focus on fairness in the lung transplant allocation system, developing computational solutions to identify and mitigate disparities in the current process.
7
46
Watch Jian Tang (@tangjianpku) from @Mila_Quebec discuss his work on deep generative models for molecular conformation generation at the AI Cures Drug Discovery Conference.
View the full presentation here: piped.video/MzO8NAMm8Ko
#AIHealth #AI #ML #DrugDiscovery
23
45
Congratulations to #JameelClinic's @HannesStaerk, @octavianEganea, @BarzilayRegina, Tommi Jaakkola, & @MITChemE's @lucky_pattanaik for the acceptance of their latest research paper to @icmlconf! The existing computational methods for drug discovery can be fast, but have... (1/2)
2
5
49
📣 We're thrilled to announce the date of our 2023 MIT-MGB AI Cures Conference: April 24th! 🌟 Mark your calendars for this exciting event where we'll be sharing cutting-edge clinical AI research & how to deploy these tools in hospital settings. REGISTER: mitmgb.ai/
ALT Poster for the 2023 MIT-MGB AI Cures Conference with a list of all the speakers and a short summary of what the AI Cures conference is about.
2
19
46
12,522
In drug design, molecular docking is typically treated as a regression problem. While deep learning models are typically faster than traditional search-based methods, there have been no significant improvements in accuracy. Now, a diffusion generative model... (1/4)
1
16
46
#RememberOctavian by Jason Yim: Protein interactions mediate nearly all biological processes — understanding their geometric mechanisms would advance our understanding of disease & aid in developing novel drugs and biotherapeutics. The computational task of modeling... (1/6)
1
9
45
With its ability to generate many different potential protein-to-ligand (drug molecule) poses at once, DiffDock is about to change the game for drug discovery thanks to #JameelClinic researchers: news.mit.edu/2023/speeding-d…
14
43
5,957
Published today in @CellCellPress, "A Deep Learning Approach to Antibiotic Discovery" highlights the discovery of a powerful antibiotic named Halicin. The senior authors of the study are J-Clinic faculty co-leads Profs. Regina Barzilay and James Collins.
bit.ly/39UEWDB
33
41
We are excited to announce the AI Cures Conference: Data-driven Clinical Solutions for COVID-19, which will take place online on September 29th.
Learn more and register here, spots are limited: aicures.mit.edu/conference
32
42
Announcing the Jameel Clinic Seminar Series! Every week, starting September 22nd, we will have an expert in the field discuss a topic within AI and healthcare.
Online for everyone, in-person for MIT students only.
See the full schedule and lineup here: jclinic.mit.edu/seminar-seri…
14
42
Happening now: David Baker of @UWproteindesign joins us at #MoML2022 with a fascinating talk on using RoseTTAFold2 to assist with protein design!
ALT A crowd of people sits in an auditorium facing a PowerPoint presentation that has the speaker David Baker presenting via Zoom in the top right corner.
1
4
45
One of the most promising directions for next gen therapeutics is targeted protein degradation: hijacking the body's natural garbage disposal system to delete disease-causing proteins. But finding the RIGHT molecule to encourage the necessary protein interactions is like... (1/2)
ALT Entangled yellow, green, turquoise, blue, and orange protein helices with small orange particles floating around them.
1
4
43
Bowen Jing & @GabriCorso are @AmerChemSociety giving a presentation on Thurs @ 10:20am CT on their recent #JClinic-supported work on Torsional Diffusion, a new state-of-the-art molecular conformer generation method 🌟 Come say hi/reach out to them! arxiv.org/abs/2206.01729
1
10
39
Coarse-graining (CG) simulates the behavior of complex systems using a simplified representation; while CG of molecular simulations drastically accelerates simulation, it can suffer from info loss. #JameelClinic PI Rafael Gómez-Bombarelli, @WujieWang, @MinkaiX, Chen Cai,... (1/3)
2
9
41
🩺 L’IA peut-elle contribuer à guérir le cancer? Aziz Ayed, doctorant de la #JameelClinic, a discuté de notre réseau d’hôpitaux avec @superhenrytran, et de l’importance d’entraîner les modèles d’IA sur des populations variées: piped.video/S8INP61jxQQ?si=E-Fg…
ALT Photo of Aziz Ayed with Henry Tran (a.k.a. Henry Live) L'image d'Aziz Ayed avec Henry Tran (alias Henry Live)
4
41
19,974
AI has potential to transform cancer diagnosis and treatment, but still plays a minor role in oncology today.
Faculty from Jameel Clinic and the @kochinstitute aim to bridge the gap between the two communities—and joined together for a research meeting Monday, December 6.
5
35
Not all language models are created equal. But large language models like @OpenAI's GPT-3 have been shown to perform well in zero-shot scenarios. #JameelClinic PI @david_sontag & his collaborators found that LLMs perform well across a variety of tasks in clinical... (1/3)
2
8
40
We are excited to highlight @MarzyehGhassemi as a speaker at the AI for Healthcare Equity Conference! Here is a sneak peek of her discussing how ML can support inclusivity.
To hear her full presentation on April 12th, register: bit.ly/3dKNvpE
piped.video/xzZ__5pQTuQ
7
39
"Try to get as much feedback as possible from as many different people as possible w/ different backgrounds...and just try to keep whatever is useful for you...feedback should be taken as a way improve yourself." — Octavian-Eugen Ganea #RememberOctavian #NeverGiveUp
5
36
Watch Prof. Marinka Zitnik (@marinkazitnik) discuss her team's work on graph neural networks for COVID-19 drug repurposing at the AI Cures Drug Discovery Conference.
View the full presentation here: piped.video/wNXSkISMTw8
9
39
On Oct. 21, we hosted the first #MoML2022 Conference to honor our late postdoc @octavianEganea. Words alone can’t fully describe what a remarkable individual Octavian was, so we launched a website honoring the richness of his life & his work: octavianganea.com/
ALT Photo of Octavian-Eugen Ganea in a red weatherproof jacket standing with a large mountain peak in the background.
8
41
Parkinson's is notoriously difficult to diagnose. @MIT #JameelClinic PI Dina Katabi & her team developed an AI-powered, router-like device that can detect the severity & progression of someone's Parkinson's just from their breathing. MIT News: bit.ly/3Tc8xzB (1/3)
1
14
38
Happy International Women's Day! Join us in celebrating these brilliant scientists at the forefront of AI + healthcare!
10
37
Well-done to the Top 4 Best Paper nominees at #MoML2023 👏👏👏
1️⃣ AF2BIND: Artem Gazizov, @sokrypton et al.
2️⃣ PINDER: @AkdelMehmet, @NaefLuca, et al.
3️⃣ MiniFold: @jeremyWohlwend, Mateo Reveiz, et al.
4️⃣ EVEscape: @sarahgurev, et al.
ALT Photo of a group of people standing in front of a presentation showing photos of Octavian-Eugen Ganea. Sarah Gurev (3rd from left) is holding up the check for winning the Best Paper Prize. Left to right: Luca Naef, Jeremy Wohlwend, Sarah Gurev, Sergey Ovchinnikov
1
9
40
5,885
A new investigation by @statnews and MIT Jameel Clinic found that subtle shifts in data fed into popular health care algorithms can cause their accuracy to plummet over time, raising the prospect AI could do more harm than good in many hospitals.
statnews.com/2022/02/28/seps…
2
17
34
Prof. Bertsimas and his team are developing powerful analytics to confront COVID-19. Projects include predicting effective resource allocation, a disease progression calculator, and developing a new convenient test.
Learn more here: bit.ly/2JEt7pR
#SARSCoV2 #AI #ML
25
35
Jameel Clinic researchers identify synergistic antiviral blends for treating viruses like SARS-CoV-2 using machine learning.
Authors from Jameel Clinic include @WengongJin, @ItsJonStokes, Jim Collins, Tommi Jaakkola and @BarzilayRegina.
Read more here: bit.ly/3COAkx8
2
16
32
Exciting to see #JameelClinic AI faculty lead
@BarzilayRegina talking about the future of AI in healthcare on the 6th season of @cazadorcerebros in their latest episode "Cuántica + IA = Futuro" or "Quantum + AI = Future" on @rtve. 📺 Watch here: bit.ly/3xXxbdE
12
35
"You really always have to stand up and fight for whatever you believe in." — Octavian-Eugen Ganea #RememberOctavian #NeverGiveUp
4
36
#JameelClinic PI @marzyehghassemi & her collaborators investigated the potential biases in clinical notes in a fascinating paper to be shown at the upcoming @AIESConf in Aug. Clinical notes are becoming a popular source of data for ML tools in healthcare. If the... (1/2)
1
11
28
Can we train AI models to make more "human-like" judgments? 🔍 The devil is in the details i.e. labels. Read more from @AparnaBee, @david_madras, David H. Yang,@dhadfieldmenell,@ghadfield, & #JameelClinic PI @MarzyehGhassemi in @ScienceAdvances science.org/doi/10.1126/scia…
ALT Top row of headshots (left to right): Aparna Balagopalan, David Madras, David H. Yang Bottom row of headshots (left to right): Dylan Hadfield-Mennell, Gillian K. Hadfield, Marzyeh Ghassemi
10
33
6,036
ChatGPT's popularity has drawn attention to the potential of LLMs to transform healthcare. But researchers have found that GPT-4’s tendency to encode societal biases raises serious concerns for its use in clinical decision support.
Preprint: medrxiv.org/content/10.1101/…
ALT Top row (left to right): Travis Zack, Eric Lehman, Mirac Suzgun, Jorge A. Rodriguez Middle row (left to right): Leo Anthony Celi, Judy Gichoya, Dan Jurafsky, Peter Szolovits Bottom row (left to right): David Bates, Raja-Elie E. Abdulnour, Atul J. Butte, Emily Alsentzer
1
7
35
4,592
How will ML models perform when encountering new environments? A new work accepted to @TmlrOrg from @learn_ng, @NHulkund, @kchonyc & #JameelClinic PI @MarzyehGhassemi aims to predict how well ML models can generalize to ensure they can be safely deployed: openreview.net/pdf?id=jYkWdJ…
ALT Clockwise starting from top left: Nathan Ng, Neha Hulkund, Marzyeh Ghassemi, Kyunghyun Cho
5
33
4,646
Biased data can teach us a lot about what a society values & practices. @NEJM's AI in Medicine series published a review article by @KadijaFerryman @Maxi_Macki & #JameelClinic PI @MarzyehGhassemi encouraging an artifact-based approach to biased data: nejm.org/doi/full/10.1056/NE…
ALT Top row (left to right): Kadija Ferryman, Maxine Mackintosh Bottom: Marzyeh Ghassemi
10
35
20,590
J-Clinic PI @David_Sontag and PhD student @HsseinMzannar at @MIT_CSAIL have developed a healthcare ML system that understands when to step in. The system can look at chest X-rays to diagnose pneumonia — and know when to defer to a radiologist.
Read here: bit.ly/31fZtiZ
12
34
Organized by Irene Chen, the first ML+Health PhD student and postdoc gathering was held! On 9/14, a wide group of labs, years of PhD and others gathered together at the Stata Amphitheater.
3
33
Adam Yala (@YalaTweets) and his team at Jameel Clinic have published their new paper "Robust Mammography-based Models for Breast Cancer Risk" in @ScienceTM.
"Mirai," the mammography-based model, achieves consistent accuracy across diverse populations: piped.video/pCGnRDf0Fmo
15
31
To create fair, trustworthy AI tools, scientists should train and test algorithms on data from diverse populations. PhD Candidate Adam Yala @yalatweets validated his group's model #Mirai in Taiwan, Sweden, Brazil, Israel, NC, GA, and MA.
More: statnews.com/2021/11/16/brea… @StatNews
1
8
29
It was a pleasure to share more about the #JameelClinic mission to bring AI into real-world healthcare settings with our Jameel-Wellcome Hospital Network project to our audience of clinicians today visiting from Taichung Veterans General Hospital!
ALT Ignacio Fuentes gives a presentation by a podium and TV screen to a small audience sitting around a large conference table.
6
30
Congratulations to @BarzilayRegina & Dina Katabi of #JameelClinic for their election to @americanacad! 👏 Founded in 1780, the Academy is one of the most prestigious honorary societies & the leading center for multidisciplinary, nonpartisan research. bit.ly/38E61Q4
2
32
Interested in a comprehensive introduction to ML for healthcare?
Sign up for Jameel Clinic PIs David Sontag (@david_sontag) and Peter Szolovits' edX class on Machine Learning for Healthcare, starting March 15th.
Learn more here: edx.org/course/machine-learn…
2
8
29
Did you miss the AI Cures Drug Discovery Conference? The full livestream recording is now available to watch, we hope you enjoy it.
View the full conference here: piped.video/Lp_0AkDBaR8
7
28
It’s been a long time coming, but we’re excited to kick off our first AI & Health Summer High School Program tomorrow! 🤩
8
30
3,564
Halicin, the powerful antibiotic discovered by J-Clinic, now has its own webpage.
Visit en.wikipedia.org/wiki/Halici… to learn more about it.
12
31
#JameelClinic researchers Bowen Jing & @HannesStaerk were invited by @wpwalters to the recent @AmerChemSociety Conference to present their work on Generative Models for Molecular Docking at the Herman Skolnik Award Symposium honoring @wpwalters!
1
3
28
3,275
What if we had gene editing technology capable of imagining new proteins to rectify cancer-causing mutations? @json_yim spoke w/ MIT News about how FrameDiff generates new proteins and how it could change the future of drug development news.mit.edu/2023/generative…
4
31
2,314
“One of the big challenges right now is we have a lot of information about molecules, structures, and sequencing…but there’s lots of information missing between the scales of major processes — how do we study the connection between all these things?” — @ZhongingAlong #MoML2023
4
30
4,004
Lots of excitement and celebration (and Samoyeds) for #JameelClinic AI lead @BarzilayRegina's induction to the National Academy of Engineering Class of 2023! 🎈💐👏🐺 @theNAEng
ALT Photo of Regina Barzilay and her students in her office with balloons floating on the ceiling, multiple bouquets, and multiple Samoyeds.
3
28
2,986
#JameelClinic AI faculty lead @BarzilayRegina was invited to speak @aimbe's August Congressional Lunch Briefing for legislative staffers: "AI to Improve Health Care". As Congress considers how to regulate AI, expert medical & biological engineering outreach is critical.
ALT Regina Barzilay standing in front of United States Capitol Building.
ALT Regina Barzilay standing in front of MA Senator Edward J. Markey's office.
6
30
2,971
How can we ensure that equity is centered as opportunities for AI/ML implementation multiply in gastroenterology & hepatology? #JameelClinic PI @marzyehghassemi & her collaborators @AdjoaGIMD @UcheAnyaMD @tberzin @drfolamay adapted a framework for ethical AI/ML (1/2)
2
9
27
Just like proteins, RNA can fold into 3D structures that determine function. However, much less is known about RNA structure. To close this gap, Felix Faltings @MIT_CSAIL presents to Prof. Phil Sharp @MITBiology about #MachineLearning models that predict how RNA folds.
8
27
We have an exceptional speaker lineup including David Baker, @BarzilayRegina, Tommi Jaakkola, Rafael Gomez-Bombarelli, @wpwalters, @ZhongingAlong: moml22.mit.edu/speakers
10
28
#JameelClinic faculty lead @BarzilayRegina & @yalatweets collected 129,000 mammograms from U.S., Israel, Sweden, Taiwan, & Brazil, creating a globally diverse dataset for Mirai, their mammography-based risk model that outperformed the existing breast cancer risk model by... (1/3)
1
6
29
Regina Barzilay, @octavianEganea, @xinyuan_huang_, Tommi Jaakkola and colleagues have developed a ML model, called Equidock, that rapidly predicts how two proteins will attach — potentially accelerating the development of new medicines.
Read more here: news.mit.edu/2022/ai-predict…
9
30
Important work, published in @NatureBiotech, from Jameel Clinic's life sciences faculty lead Jim Collins, @lrsoenksen and team—who are utilizing CRISPR sensor technology to enable face masks to detect COVID-19.
Read the full publication here: nature.com/articles/s41587-0…
7
30
There's now a plethora of free AI tools online that can sharpen blurry photos 📸 but what about 3D images? @nalinimsingh spoke w/ MIT News about the beauty of combining deep learning + physics when it comes to motion correction in MRI scans. news.mit.edu/2023/mit-resear…
1
7
29
2,606
The gap between what has been achieved in AI/ML research & what is being used in clinical settings is growing. AI & ML have already demonstrated far more accuracy & efficiency in diagnosing diseases like cancer & Parkinson's, even speeding up the process of drug discovery... 1/3
ALT Three headlines that say "MIT device senses breathing patterns to spot Parkinson's disease", "Is artificial intelligence about to transform the mammogram? An MIT researcher who survived breast cancer has devised a technique that seems to predict many breast cancer cases", "Can artificial intelligence overcome the challenges of the health care system? MIT and Mass General Brigham researchers and physicians connect in person to bring AI into mainstream health care"
1
5
28
#JameelClinic PI @MarzyehGhassemi kicked off the 4th season of @XtelligentHC @HITAnalytics Healthcare Strategies podcast hosted by Shania Kennedy w/ a great interview on AI/ML in value-based care, & how to avoid introducing/perpetuating negative biases: soundcloud.com/xtelligentmed…
ALT Headshot of Marzyeh Ghassemi
1
3
25
What's the latest in AI bias research? We spoke w/ @hammaadadam1, a 3rd year PhD student, on the exciting work happening in the Healthy ML group, run by #JameelClinic PI @MarzyehGhassemi. 1/4
ALT Headshots of Hammaad Adam (left) and Marzyeh Ghassemi (right).
1
5
25
Lung cancer is the world's #1 deadliest cancer, resulting in 1.7 million deaths worldwide in 2020, killing more people than the next 3 deadliest cancers combined. Sybil is a deep learning model built to assess each patient's lung cancer risk. The tool was developed by... 1/3
ALT (A) Annotation of lung cancers from LDCT scans in Sybil training.
1
10
27
5,240
What if we could monitor #COVID19 patients from a safe distance? J-Clinic PI Dina Katabi and her research group have developed an AI algorithm that monitors patient breathing, movement and sleep using wireless signals in the environment.
Read more here: bit.ly/2VbUdv4
12
28
We are celebrating our wonderful female PIs today on the International Day of Women and Girls in Science!
Regina Barzilay, Bonnie Berger, Yonina Eldar, Dina Katabi, @RosalindPicard, and Caroline Uhler—thank you for pushing science forward within #AIHealth.
#WomenInScience
8
25
What do physicians think about AI in healthcare? A big thank you to our partners at @MassGenBrigham for giving our summer students the chance to listen to these insightful lectures in the historic 200-year-old Ether Dome.
ALT Collage of photos taken in the Mass General Brigham Ether Dome and one in an MIT classroom.
6
28
2,729
2022 was another transformative year at MIT. Wonderful to see the work of #JameelClinic AI faculty lead @BarzilayRegina & @YalaTweets featured in MIT News' 2022 review! news.mit.edu/2022/mit-media-…
8
25
2,683
We're excited to announce the launch of the 1st #JameelClinic AI & Health Summer Program! This rigorous 1-week program is tailored to rising 10th-12th graders w/ a strong interest in AI + medicine. The deadline for applications is June 5, 2023.
Apply now: jclinic.mit.edu/hssummerprog…
ALT A graphic that reads "MIT Jameel Clinic AI & Health Summer Program" with an image of the MIT Great Dome and the image of a Black female student on a laptop seated on a part of MIT campus.
20
8
28
6,028
Randomized controlled trials (RCTs) are "gold standard" for estimating causal effects, but often *exclude* important patient populations. Meanwhile, observational studies cover broader patient populations, but can be misled by confounding & other biases. 1/3
2
6
30
Can co-training improve the performance of prompt-based learning when using unlabeled data? Prompting has become a promising approach for few-shot & zero-shot learning, but it can be sensitive to trivial variations & oft. requires much larger models vs those used for... (1/2)
1
6
27
We chatted w/ #JameelClinic researchers @pgmikhael & @jeremyWohlwend who have been hard at work on a lung cancer risk prediction tool. Keep an eye out for the results of this impactful research in 2023!✨: piped.video/KRQMnA-Y_zk
3
9
25
3,200
Falling Walls seeks to answer the following: “Which are the next walls to fall in science and society?" Congratulations to @YalaTweets & @BarzilayRegina for being selected as Life Sciences finalists for this year's @Falling_Walls Science Breakthrough! 👏bit.ly/3B1OMUN
4
25