Pranav Rajpurkar
@pranavrajpurkar
Followers
25K
Following
664
Media
351
Statuses
2K
Co-Founder a2z Radiology AI https://t.co/mdi996SFZy | Assistant Professor Harvard Lab: https://t.co/Z6tBGoluEG
Joined March 2015
RT @jn_acosta: Our latest Perspective on how generative voice AI agents could transform healthcareβmaking medical expertise available 24/7,β¦.
0
8
0
Excited to share our work to be presented at CVPR this week β πΉπππ‘πΆβππππππ, an agentic approach to fixing measurement hallucinations in multimodal generative AI models. @aliceheimanxyz @XiaomanZhang99
3
5
28
Big launch today from our lab at Harvard: ππππππ βΒ a large-scale, high quality question answering benchmark for chest X-rays. This represents a challenge for existing large frontier models, but one we believe will serve as a good progress measure for next gen models.
5
30
124
π₯ If you're a US/Canada board certified radiologist interested in getting on an AI publication by participating in our validation study, reach out to us [link in thread] @a2zradiologyai.
1
2
2
New website. Let's get every radiologist an AI assistant.
7
17
82
This will be the a big year for code. The graph highlights my coding/writing days (brighter = more) since I started coding 10 months ago. My writing is also now AI-assisted and GitHub-tracked. Come the coding revolution.
0
0
12
After 14 years of writing code, I felt a phase shift today with Claude 4 + Cursor. English is now the programming language. Execution speed is defined by clarity of plan. Manual verification is becoming optional, even infeasible. Low-level skills will quickly deskill.
9
17
210
A pleasure to collaborate with Jung-Oh Lee, @HongYuZhou14, @tberzin, @DanielSodickson on rethinking the frontier of multimodal generative ai for complex medical data!.
Thanks to @pranavrajpurkar & team for leading this @npjDigitalMed exploration of video-text generative models, including a focus on unique opportunities in GI endoscopy, from workflow phase recogntion to integration of multimodal data. Open access:.
0
6
15
AI-generated draft sentences straight from the scan -> less time drafting, more time on complex decision-making. Like having an "AI fellow" by your side. See how we're building to streamline radiology workflows. Read the full announcement by me, @siddograMD and @SamirRajpurkar.
0
0
0
We're launching a new feature that goes beyond detection to synthesize findings into cohesive, sentence-level outputs. Imagine easily reporting:.βοΈ Umbilical hernias.βοΈ Post-surgical changes.βοΈ Coronary artery calcifications.βοΈ Nonobstructive kidney stones . and more.
1
0
0
Radiologists, what if your AI could draft prelims like a seasoned fellow β especially for all those incidental findings?Β π€― . Super excited to announce a2z Instant Prelims from @a2zradiologyai!.
2
3
24
This dataset is the largest in the current literature on AI for radiology that can be used to compare human-AI collaboration with AI alone or humans alone. Unique features include the large number of radiologists, disease labels and assessments.
0
2
12
We have just made the largest dataset for understanding radiologist-AI collaboration available to the world with 227 radiologists in our study.
2
26
136
A big day β we are making the largest dataset of chest X-rays available to the world. This is the first dataset to cross 100k patients.
13
76
379