Sign in

JHU Computer Science

@jhucompsci.bsky.social
346 followers 146 following 827 posts

A diverse and collaborative community on the cutting edge of computing and technology within hopkinsengineer.bsky.social at the Johns Hopkins University. cs.jhu.edu • Baltimore, MD

PostsRepliesMedia
JHU Computer Science @jhucompsci.bsky.social · 09/10/2026
Hear from Suchi Saria on using #AI to catch sepsis early, which TIME recognized her for by including her on its 100/AI List. Learn more here: www.cs.jhu.edu/news/creator...
youtube.com
Suchi Saria on using AI to catch sepsis early
Subscribe to TIME’s YouTube channel ►► http://ti.me/subscribe-time ...
000
Reposted by JHU Computer Science
Johns Hopkins University @jhu.edu · 22/09/2026
Johns Hopkins University ranks No. 9 among national universities in the latest undergraduate rankings from U.S. News & World Report. The university also ranks No. 4 for undergraduate research/creative projects and No. 1 in biomedical engineering. Read more: bit.ly/4hjXl0W
A tall, white clock tower with a green, pointed roof is visible in the background, partially obscured by autumn foliage in vivid shades of orange and red. The sky is clear and bright.
022
JHU Computer Science @jhucompsci.bsky.social · 09/10/2026
Next week at #UbiCompISWC2026, @jhu.edu computer scientists will present “RF-HOI: Recognize Human-Object Interaction with Radio Frequency Signals,” the first framework that only uses radio frequency signals for HOI recognition. Learn more here: theeagleofthedesert.github.io/RF-HOI/
UbiComp ISWC 2026 logo.
000
JHU Computer Science @jhucompsci.bsky.social · 08/10/2026
Congratulations to Zongwei Zhou on his award!
cs.jhu.edu
Zhou among nine promising Johns Hopkins innovators awarded JHTV funding
Assistant Professor Zongwei Zhou has received a competitive translational funding award from Johns Hopkins Technology Ventures to develop AI infrastructure in support of early cancer detection.
011
JHU Computer Science @jhucompsci.bsky.social · 06/10/2026
“Autonomous robotic surgery is on the critical path to help solve the organ shortage,” says co-PI @mathias-unberath.bsky.social. “This work builds the foundation for its broader adoption in and around the operating room.”
cs.jhu.edu
Addressing the organ transplant shortage: SURPASS project tackles the “beyond possible”
The latest winning SURPASS projects unite experts from the Johns Hopkins University’s Whiting School of Engineering and Applied Physics Laboratory to solve world-changing problems.
000
JHU Computer Science @jhucompsci.bsky.social · 05/10/2026
Join us and the @jhu.edu Information Security Institute on November 4 for a joint seminar featuring @notredame.bsky.social’s Fanxin Kong! Learn more here: www.cs.jhu.edu/event/cs-isi...
Computer Science & ISI Seminar Series. Assured and Intelligent Cyber-Physical Systems. Novemeber 4, 2026, 11 a.m. 213 Hodson Hall. Fanxin Kong, University of Notre Dame.
000
Reposted by JHU Computer Science
Johns Hopkins University @jhu.edu · 27/08/2026
An AI-powered early warning system developed by Suchi Saria and her team at Johns Hopkins is now deployed in over 40 hospitals nationwide, detecting sepsis before clinicians may even suspect it. CNN's Jake Tapper featured the platform on the network's AI Friend or Foe series. cnn.it/4y5gD0A
112
JHU Computer Science @jhucompsci.bsky.social · 02/10/2026
Congratulations, @hmeld.bsky.social!
Congratulations! Harry Eldridge successfully defended his dissertation “New Techniques in Secret Sharing and Private Threshold Aggregation Protocols” under the guidance of advisors Matthew Green and Abhishek Jain. Harry plans to join Boston University as a postdoctoral researcher. We in the department are extremely proud of our students who have successfully completed their PhD. Congratulations on this achievement and best wishes as you begin an exciting new phase of life!
010
JHU Computer Science @jhucompsci.bsky.social · 02/10/2026
Anton Dahbura’s recent talk gave perspective on the privacy and security research happening at @jhu.edu. Learn more about the pre-USENIX Security ’26 workshop organized by @yaxingyao.bsky.social, @erye.bsky.social, and other JHU CS faculty:
cs.jhu.edu
Anton Dahbura delivers keynote at USENIX Security ’26 pre-workshop event
The pre-workshop was organized by Johns Hopkins Computer Science faculty in advance of the 35th USENIX Security Symposium.
001
JHU Computer Science @jhucompsci.bsky.social · 01/10/2026
Read the full paper here: ieeexplore.ieee.org/document/113... (9/9)
Results: Better cancer-detection AI, from words. +6.5% early-stage tumor detection (< 20 mm), +8.2% malignant vs. benign classification sensitivity, +3.1% segmentation boundary accuracy (NSD), 34,035 radiology reports learned from—only 141 paired with scans. Open source: github.com/MrGiovanni/TextoMorph. Xinran Li, Zongwei Zhou, Alan Yuille and contributors at NVIDIA, UCSF, Sun Yat-sen University and Hong Kong PolyU. Supported by the Lustgarten Foundation and NIH. Patent pending. TextoMorph • IEEE Transactions on Medical Imaging • Johns Hopkins University. 8/8
010
JHU Computer Science @jhucompsci.bsky.social · 01/10/2026
It also boosted the detection rate of large liver tumors to 87.7%, significantly outperforming previous methods, while generating synthetic tumors so realistic that radiologists frequently mistook them for real ones. (8/9)
Why it matters: Train AI on exactly what it keeps missing. Two groups of four CT scans and their zoomed in counterparts. The first group is labeled "Real tumors the AI missed," with four scans labeled "The tumor shows heterogeneous enhancement in its appearance.", "The lesion appears hypoattenuating with ill-defined borders.", "Cystic features with hypoattenuating appearance.", and "The image demonstrates a cyst." The second group is labeled "Synthetic look-alikes generated from the description" with four scans labeled "The lesion appears hypodense." twice, "Hypoattenuating lesion with ill-defined edges.", and "The image shows a cyst." Left: real tumors the detector missed, described in words. Right: synthetic look-alikes built from those words. Training on them lifted small pancreatic tumor detection from 66.7% to 87.5%. TextoMorph • IEEE Transactions on Medical Imaging • Johns Hopkins University. 7/8
100
JHU Computer Science @jhucompsci.bsky.social · 01/10/2026
The results are impressive: Training with these targeted cases increased the detection of hard-to-catch early-stage tumors by 6.5% and improved cancer classification sensitivity by 8.2%. (7/9)
Realism: Can you tell which one is real? Three paired images labeled "real" and "synthetic" in the following categories: cyst, PDAC, and PNET. Two radiologists reviewed 540 scans in a blinded test. They mistook TextoMorph tumors for real ones between 1 in 4 and more than 1 in 2 times—far more often than with previous synthesis methods. TextoMorph • IEEE Transactions on Medical Imaging • Johns Hopkins University. 6/8
100
JHU Computer Science @jhucompsci.bsky.social · 01/10/2026
Instead of just asking for “a tumor,” researchers can ask it to include specific characteristics, such as an “ill-defined, hypoattenuating liver lesion”—or in layperson’s terms, a faint, shadowy spot with blurry edges—and the AI creates exactly that. (6/9)
Not just one slice: The tumor exists in full 3D. CT images with various zoomed-in portions labeled "axial view," "sagittal view," and "coronal view." A single synthetic liver tumor seen in axial, sagittal and coronal planes. TextoMorph works on 3D CT volumes, so the tumor is consistent from every angle. TextoMorph • IEEE Transactions on Medical Imaging • Johns Hopkins University. 5/8
100
JHU Computer Science @jhucompsci.bsky.social · 01/10/2026
The system functions like a specialized artist. (5/9)
Across the body: Liver, pancreas, kidney—each tumor generated from its own sentence. CT images with red squares indicating where the zoomed-in image next to it comes from, labeled "liver," "pancreas," and "kidney," with phrases like "...heterogeneous enhancement in its appearance.", "...enhancing features with washout characteristics," "Cystic features... hypoattenuating appearance," "The image shows a cyst," "...ill-defined lesions and atrophy.", "Hypoattenuating lesions...", "Benign cystic formation...", "PNET can be seen.", "Scattered renal cysts...", "...heterogeneously enhancing exophytic mass.", "...a renal mass.", and "The scan shows a renal cyst." Synthetic tumors are placed into full CT volumes (red box) and match the clinical phrase they were generated from, from "benign cystic formation" to "heterogeneously enhancing exophytic mass." TextoMorph • IEEE Transactions on Medical Imaging • Johns Hopkins University. 4/8
100
JHU Computer Science @jhucompsci.bsky.social · 01/10/2026
That’s why the research team’s tool uses detailed medical language from radiology reports to “draw” realistic 3D tumors on CT scans. (4/9)
Controllability: Change the words, change the tumor. CT images of tumors labeled "Liver," "Pancreas," and "Kidney" with words "cysts," "enhancing," "fatty", "heterogeneous," "hypoattenuating," "atrophy," "ill-defined," "necrotic," and "stones". The same organ, the same mask—a different descriptor gives a different tumor. Text controls what earlier methods could not: texture, heterogeneity, boundaries and pathology. TextoMorph • IEEE Transactions on Medical Imaging • Johns Hopkins University. 3/8
100
JHU Computer Science @jhucompsci.bsky.social · 01/10/2026
But real-world training data is often scarce, and existing methods for creating “fake” tumor images usually produce simple blobs that don't look enough like the complex, subtle tumors that confuse both doctors and machines. (3/9)
How it works: From radiology report to synthetic tumor—the TextoMorph pipeline. Diagram showing the overview of the system pipeline. (I) An LLM distills each report into tumor descriptors. (II) A 3D diffusion model learns to synthesize tumors conditioned on text, mask and healthy CT. (III) The frozen model generates tumors on demand. (IV) Detection AI is trained on them. TexttoMorph • IEEE Transactions on Medical Imaging • Johns Hopkins University. 2/8
100
JHU Computer Science @jhucompsci.bsky.social · 01/10/2026
A critical failure of #AI for cancer imaging is the false negative, where the AI scans a patient, but fails to detect a tumor—this usually happens because the AI hasn’t seen enough examples of specific, difficult-to-detect tumor types during its training. (2/9)
The idea: Radiologists already describe tumors in precise words: heterogenous, enhancement, solid, hypodensity, hypoattenuating, well-defined, arterial enhancement, cystic, metastases, circumscribed, smooth, ill-defined, cirrhosis, hypointense, walled, calcifications, washout. Every day, radiology reports are filled with phrases like "hypodense," "cystic," "ill-defined," "calcifications." TextoMorph turns that language into a control knob for generating tumors in CT scans. Images of tumors in CT scans. TextoMorph • IEEE Transactions on Medical Imaging • Johns Hopkins University. 1/8
100
JHU Computer Science @jhucompsci.bsky.social · 01/10/2026
A team led by researchers from @jhu.edu has developed a new system called TextoMorph, introduced in @ieeexplore.ieee.org IEEE Transactions on Medical Imaging. 🧵 (1/9) feat. @canzhao.bsky.social, collaborators from @ucsanfrancisco.bsky.social, & more, with support from @lustgarten.org
121
JHU Computer Science @jhucompsci.bsky.social · 01/10/2026
Happy Work Anniversary! Laura Henneman celebrating 20 years of incredible leadership! Computer Science. ... and 30 YEARS at Johns Hopkins!
010
JHU Computer Science @jhucompsci.bsky.social · 30/09/2026
We are proud to share that Russell Taylor has been honored with a 2027 IEEE Technical Field Award, among IEEE’s highest honors. These awards recognize outstanding technical achievements and leadership advancing technology for humanity. Join us in celebrating Russ and the full 2027 class:
invt.io
Celebrate the 2027 IEEE Technical Field Award Honorees!
Learn More
031
JHU Computer Science @jhucompsci.bsky.social · 30/09/2026
Congratulations to @benlangmead.bsky.social and @nathanielbrown.bsky.social on receiving a Best Paper Award for their work, “Bounding the Average Move Structure Query for Faster and Smaller RLBWT Permutations,” at the 24th Symposium on Experimental Algorithms!
cs.jhu.edu
Hopkins computer scientists win SEA Best Paper Award
The Symposium on Experimental Algorithms explores the role of experimentation and engineering techniques in the design and evaluation of algorithms and data structures.
071
JHU Computer Science @jhucompsci.bsky.social · 29/09/2026
...propose an AI-powered bridging 🌉 system that augments the standard social media feed. (13/13)
programs.sigchi.org
Conference Programs
000
JHU Computer Science @jhucompsci.bsky.social · 29/09/2026
& in “CrossWeave: Bridging Perspectives Across Online Communities with a Dual-Pane Design,” @williamjurayj.bsky.social, @psingh54.bsky.social, @danielkhashabi.bsky.social, @andrewjperrin.bsky.social, @tiziano.bsky.social, @ziangxiao.bsky.social, @qcznlp.bsky.social, & more... (12/13)
100
JHU Computer Science @jhucompsci.bsky.social · 29/09/2026
“Content Creation with Generative AI: How Do Content Creators Responsibly Use Generative AI Tools?” by Stephanie Milani & @cmu.edu researchers interviewed social media 📱 content creators to examine their motivations, practices, & specific challenges related to responsible generative AI use: (11/13)
dl.acm.org
Content Creation with Generative AI: How Do Content Creators Responsibly Use Generative AI Tools? | Proceedings of the ACM on Human-Computer Interaction
The rise of Generative AI (GenAI) has demonstrated significant potential to improve productivity and foster creativity among content creators, social media influencers with large audiences on platform...
110
JHU Computer Science @jhucompsci.bsky.social · 29/09/2026
...“‘Hi Alex’ or ‘Dear Dr. Morgan’?: Exploring How AI-suggested Politeness Strategies Influence Email Writing and Social Perception Among Native and Non-Native Speakers”: (10/13)
dl.acm.org
"Hi Alex" or "Dear Dr. Morgan"?: Exploring How AI-suggested Politeness Strategies Influence Email Writing and Social Perception Among Native and Non-native Speakers | Proceedings of the ACM on Human-Computer Interaction
As AI writing assistants are increasingly used for interpersonal communication, they may have profound impacts on interpersonal relationships. Politeness is one important aspect of social communication that is grounded in people’s perceptions of ...
100
JHU Computer Science @jhucompsci.bsky.social · 29/09/2026
And at #CSCW2026, @ziangxiao.bsky.social, @qveraliao.bsky.social, @jeffjianzhao.bsky.social, & more investigate how politeness strategies in AI-generated suggestions affect people’s email writing 📨 and alter their perception of social situations in... (9/13)
100
JHU Computer Science @jhucompsci.bsky.social · 29/09/2026
In “SciTaRC: A Plan-Annotated Scientific Tabular QA Benchmark for Language Reasoning and Complex Computation,” @phikoehn.bsky.social and team introduce an expert-authored benchmark for question ❓ answering over scientific tables: (8/13)
arxiv.org
100
JHU Computer Science @jhucompsci.bsky.social · 29/09/2026
“BiomedSQL: Text-to-SQL for Scientific Reasoning on Biomedical Knowledge Bases” by @tanaynayak.bsky.social, @danielkhashabi.bsky.social, & more introduces the first benchmark to evaluate scientific reasoning in text-to-SQL generation over a real-world biomedical knowledge base: (7/13)
arxiv.org
BiomedSQL: Text-to-SQL for Scientific Reasoning on Biomedical Knowledge Bases
Biomedical researchers increasingly rely on large-scale structured databases for complex analytical tasks. However, current text-to-SQL systems often struggle to map qualitative scientific questions i...
100
JHU Computer Science @jhucompsci.bsky.social · 29/09/2026
@tanaynayak.bsky.social, @danielkhashabi.bsky.social, & more present the first systematic study of gold 🥇 context size in long-context question answering in “Hidden in the Haystack: Smaller Needles Are More Difficult for LLMs to Find”: (6/13)
arxiv.org
Hidden in the Haystack: Smaller Needles are More Difficult for LLMs to Find
Large language models (LLMs) face significant challenges with needle-in-ahaystack tasks, where relevant information ("the needle") must be drawn from a large pool of irrelevant context ("the haystack"...
100
JHU Computer Science @jhucompsci.bsky.social · 29/09/2026
In “Can Coding Agents Reproduce Findings in Computational Materials Science?” @williamjurayj.bsky.social, @danielkhashabi.bsky.social, and more present a benchmark for evaluating LLM-based agents’ ability to reproduce claims from computational materials 💻🪨 science: (5/13)
arxiv.org
Can Coding Agents Reproduce Findings in Computational Materials Science?
Large language models are increasingly deployed as autonomous coding agents and have achieved remarkably strong performance on software engineering benchmarks. However, it is unclear whether such succ...
110
JHU Computer Science @jhucompsci.bsky.social · 29/09/2026
“From Papers to Panoramas: Building Hierarchies of Scientific Literature at Scale” by Jash Shah, Weiqi Wang, @danielkhashabi.bsky.social, and more develops an approach to organizing broad swaths of scientific literature 🧪📚 into a high-quality hierarchical structure: (4/13)
arxiv.org
From Papers to Panoramas: Building Hierarchies of Scientific Literature at Scale
Scientific knowledge is growing rapidly, making it difficult to track progress and high-level conceptual links across broad disciplines. While tools like citation networks and search engines help retr...
100
JHU Computer Science @jhucompsci.bsky.social · 29/09/2026
@enalisnick.bsky.social, @metodjazbec.bsky.social, @canaesseth.bsky.social, @stephanmandt.bsky.social, and more show how to increase diversity by using softened remasking heuristics in “A Tale of Two Temperatures: Simple, Efficient, and Diverse Sampling from Diffusion Language Models”: (3/13)
arxiv.org
A Tale of Two Temperatures: Simple, Efficient, and Diverse Sampling from Diffusion Language Models
Much work has been done on designing fast and accurate sampling for diffusion language models (dLLMs). However, these efforts have largely focused on the tradeoff between speed and quality of individu...
100
JHU Computer Science @jhucompsci.bsky.social · 29/09/2026
At #COLM2026, in “It’s How You Ask: Gendered Linguistic Bias in LLMs,” @ktvank.bsky.social & @anjalief.bsky.social show that when prompts contain linguistic features more commonly used by women 💁‍♀️, they elicit shorter, less sophisticated, and less formal responses—also featured here: (2/13)
hub.jhu.edu
AI might be making women sound bad at work
Office correspondence is weaker when requested with language commonly used by women
100
JHU Computer Science @jhucompsci.bsky.social · 29/09/2026
Learn about the research our computer scientists will be presenting at the upcoming @colmweb.org #COLM2026 and @acm-cscw.bsky.social #CSCW2026 conferences—a 🧵! (1/13)
CSCW 2026. Mountains.
110
JHU Computer Science @jhucompsci.bsky.social · 29/09/2026
Congratulations, Zili!
Congratulations! Zili Huang Successfully defended his dissertation “Self-Supervised Learning for Conversational Speech Processing” under the guidance of advisor Sanjeev Khudanpur. Zili plans to join Meta Reality Labs as a research scientist. We in the department are extremely proud of our students who have successfully completed their PhD. Congratulations on this achievement and best wishes as you begin an exciting new phase of life!
000
JHU Computer Science @jhucompsci.bsky.social · 28/09/2026
A @jhu.edu team led by @mathias-unberath.bsky.social will work with software company Kitware to create virtual simulations and validation testing on four @arpa-h.bsky.social-funded robotic solutions for stroke.
cs.jhu.edu
Johns Hopkins engineers partner with industry to develop AI-powered robotic stroke treatment
Newly funded teams will lead programs to develop autonomous robots to deliver curative stroke treatment.
010
JHU Computer Science @jhucompsci.bsky.social · 28/09/2026
Congrats to our WSE Staff Awards nominees! Headshots of Megan Dakwa, Laura Henneman, and Jaimie Patterson surrounded by fall leaves over an exterior shot of Malone Hall.
000
JHU Computer Science @jhucompsci.bsky.social · 24/09/2026
On May 29, 2026, @matthewdgreen.bsky.social found that the encrypted reasoning blocks returned by popular AI APIs can be replayed within a session, across sessions, and across separate user accounts. Neither OpenAI nor @anthropic.com issued architectural fixes.
techtimes.com
Single Shared Encryption Key Let Anyone Read AI Reasoning Buried in Published Logs
Encrypted AI reasoning vulnerability in Anthropic, OpenAI, and Google APIs let researchers decode 315,320 published session logs and recover 182 developer credentials -- because all three providers us...
000
JHU Computer Science @jhucompsci.bsky.social · 24/09/2026
Congratulations to Russell Taylor on being selected for the 2027 @ieeeras.bsky.social IEEE Robotics and Automation Award! 🏆 This award recognizes a select few extraordinary individuals’ contributions to and significant advancements in the field of robotics and automation.
cs.jhu.edu
Russell Taylor to receive 2027 IEEE Robotics and Automation Award
Bestowed by the Institute of Electrical and Electronics Engineers, this prestigious award recognizes Taylor’s contributions to, leadership in, and translation of surgical and medical robotics.
020
JHU Computer Science @jhucompsci.bsky.social · 23/09/2026
and Pedro R. A. S. Bassi, Wenxuan Li, @jienengchen.bsky.social, Xinze Zhou, Alan L. Yuille, Zongwei Zhou, Hanxue Gu, Zheren Zhu, Sezgin Er, Ibrahim Ethem Hamamci, Bjoern Menze, Gulhan E. Akan, Kang Wang, and Yang Yang will present “RT-Super: Learning Tumor Segmentation from Future Reports.” (8/8)
000
JHU Computer Science @jhucompsci.bsky.social · 23/09/2026
In “Merlin Plus: A Large-Scale, Multi-Cancer, Image-Mask-Report Dataset,” Pedro Bassi, Wenxuan Li, Xinze Zhou, Alan Yuille, Zongwei Zhou, @akshay-chaudhari.bsky.social, @curtlanglotz.bsky.social, & more present the 1st large-scale CT dataset w/ radiologist-created tumor masks over 9 organs: (7/8)
github.com
GitHub - MrGiovanni/MerlinPlus: [MICCAI 2026] Merlin Plus is the first large-scale public CT dataset with radiologist-created tumor masks across 9 organs (spleen, bladder, gallbladder, stomach, duoden...
[MICCAI 2026] Merlin Plus is the first large-scale public CT dataset with radiologist-created tumor masks across 9 organs (spleen, bladder, gallbladder, stomach, duodenum, prostate, adrenal glands,...
100
JHU Computer Science @jhucompsci.bsky.social · 23/09/2026
“MRI2Rep: Autoregressive Structured Report Generation for 3D Liver MRI” by Zongwei Zhou and teammates at @yale.edu proposes an autoregressive framework for liver MRI report generation: (6/8)
arxiv.org
MRI2Rep: Autoregressive Structured Report Generation for 3D Liver MRI
Manual reporting of 3D MRI studies is time-consuming, yet end-to-end structured report generation for 3D liver MRI remains underexplored due to volumetric complexity and scarce paired data. We propose...
100
JHU Computer Science @jhucompsci.bsky.social · 23/09/2026
Wenxuan Li, Pedro R. A. S. Bassi, Xinze Zhou, Qi Chen, Alan L. Yuille, Zongwei Zhou, & more present the first large-scale, open-source longitudinal and multimodal dataset for multicancer screening in “CancerVerse: A Fully Open Longitudinal and Multimodal Dataset for Multicancer Screening”: (5/8)
cs.jhu.edu
100
JHU Computer Science @jhucompsci.bsky.social · 23/09/2026
...propose a simulated degradation-to-enhancement method that learns to reverse realistic acquisition artifacts in low-quality energy-integrating CT by leveraging high-quality photon-counting CT as reference: (4/8)
arxiv.org
Distilling Photon-Counting CT into Routine Chest CT through Clinically Validated Degradation Modeling
Photon-counting CT (PCCT) provides superior image quality with higher spatial resolution and lower noise compared to conventional energy-integrating CT (EICT), but its limited clinical availability re...
100
JHU Computer Science @jhucompsci.bsky.social · 23/09/2026
At #MICCAI2026, in “Distilling Photon-Counting CT into Routine Chest CT through Clinically Validated Degradation Modeling” Junqi Liu, Xinze Zhou, Wenxuan Li, Kai Ding, Alan L. Yuille, Zongwei Zhou, & more... (3/8)
100
JHU Computer Science @jhucompsci.bsky.social · 23/09/2026
At #IROS2026, “Reeling It In: Flexible Needle Pick Up via Thread Manipulation for Autonomous Suturing” by Zih-Yun Chiu and team proposes an autonomous framework that uses a suture thread 🪡 as an assistive tool for indirect needle pickup: (2/8)
arxiv.org
Reeling It In: Flexible Needle Pick Up via Thread Manipulation for Autonomous Suturing
Suture-needle pickup is necessary for autonomous suturing, as a needle can be unexpectedly dropped or strategically released to adjust the grasping configuration. Current methods for autonomous needle...
100
JHU Computer Science @jhucompsci.bsky.social · 23/09/2026
Explore what our researchers will be presenting next week at #IROS2026 and @miccaisociety.bsky.social’s #MICCAI2026—a 🧵 (1/8)
IROS 2026 Pittsburgh. Bridge illustration made out of yellow and black robots.MICCAI 2026. Illustration of Strasbourg skyline.
100
JHU Computer Science @jhucompsci.bsky.social · 23/09/2026
JHU CS alum Iris Gupta is one of @baltbizonline.bsky.social’s Inno Under 25 honorees!
bizjournals.com
Meet Baltimore's 2026 Inno Under 25 honorees
These up-and-coming leaders are pushing innovation in fields including artificial intelligence, fetal surgery and energy.
000
JHU Computer Science @jhucompsci.bsky.social · 22/09/2026
Congratulations, Prof. Zhou! Learn more about how this funding will help develop AI for #cancer detection:
cs.jhu.edu
Zongwei Zhou awarded $2.4 million NIH grant
The four-year R01 grant will fund the development of AI algorithms capable of detecting three types of abdominal cancers on CT scans, enabling earlier diagnosis and treatment.
000
JHU Computer Science @jhucompsci.bsky.social · 22/09/2026
Read more here:
cs.jhu.edu
Human genome milestone opens door for personalized genomics
The ability to quickly and affordably survey a patient’s entire genome is expected to accelerate research, diagnostics, and precision medicine.
000