Headlines

A machine learning-derived composite biomarker built from twelve routine clinical measurements predicted waitlist mortality and transplantation in liver transplant candidates more accurately than MELD scores measured at initial evaluation.

A seven-year study of over 14,000 Chinese adults finds that combining online and offline social activity is associated with slower cognitive decline and frailty progression in middle-aged and older people.

A new review argues that AI-driven antiviral discovery must be benchmarked against respiratory RNA virus polymerases under virologically realistic conditions rather than relying on docking scores and general protein-ligand datasets.

Machine learning is accelerating target identification, toxicity prediction and drug repurposing, but researchers insist algorithmic discoveries still demand rigorous experimental validation.

Frontiers in Neuroscience has retracted a 2018 study on microRNA Let-7f-5p and stem cell survival in Alzheimer's disease models after image integrity concerns and the authors' failure to provide complete raw data.

A new review shows that hybrid deep learning and swarm-optimized AI models are achieving near-perfect prediction of lithium-ion battery charge, health, and remaining lifespan for electric vehicles and grid storage.

Researchers used a vector-free CRISPR activation system to boost dopamine biosynthesis genes in cellular models of Parkinsonian neurotoxicity, finding modest DDC upregulation, no GCH1 activation, and no added toxicity.

The journal Pathology & Oncology Research has retracted a 2018 osteosarcoma study after the authors reported improper image cropping, with an investigation confirming the findings are no longer supported by the data.

Researchers have built and validated a GMP-compliant, semi-automated manufacturing platform that produces clinical-grade extracellular vesicles from Wharton's jelly-derived mesenchymal stromal cells with consistent quality and one-year frozen stability.

A large two-center study finds that a low donor PaO2/FiO2 ratio raises the risk of early primary graft dysfunction after lung transplantation without affecting two-year survival.

A University of Toronto-led team will use a $2.3 million Google.org award to build OpenPep, an open-access database and AI platform for predicting peptide-protein interactions in drug discovery.

The largest cohort study to date finds that kidney transplants from donors with thin basement membrane disease perform as well as those from donors with normal biopsies, offering new reassurance for living donation programs worldwide.

After a Chrome clone was spotted on the Play Store earlier this week, a user on Reddit says they're seeing a fake version of WhatsApp in a "sponsored" slot above the real version. As reported earlier by Android Authority, the WhatsApp dupe has racked up over 500,000 downloads and, at the time of writing, its Play Store listing is still live. Google did not immediately respond to a request for comment. The WhatsApp clone is from a developer going by "Agment Assets sallu" who doesn't currently have anything else listed on the Play Store. Their listing for the WhatsApp impostor has the same icon, screenshots, and text as the real version, but … Read the full story at The Verge.

In a study of fingertip regeneration, a team led by JAX senior research scientist James Godwin found that some immune cells in mice can attack and kill cells needed to rebuild bone, while a subset of those same immune cells can protect the bone-making cells and allow regeneration to continue.

A new GeroScience study shows that the dietary polyphenol chlorogenic acid protects aging intestinal stem cells from oxidative stress in fruit flies through a mechanism requiring CncC, the fly counterpart of the mammalian NRF2 pathway.

Actor and angel investor Mark Wahlberg says he's OK with using AI in filmmaking, provided it's used "to enhance what you're doing."

New research illustrates the rapid rise of clinicians’ use of AI to answer medical questions despite little evidence of how it affects decision-making or patient outcomes. Researchers studied usage data from OpenEvidence, an AI search ...

The virus used to deliver healing DNA in gene therapy is under renewed scrutiny.

→ More former Genentech leaders are finding a new home. Lélia Delamarre posted on LinkedIn that she’s leading R&D at ARCH-backed cancer biotech Medici Therapeutics, one of this year’s

Older age, baseline oxygen use, and lower respiratory tract symptoms signal worse outcomes among hospitalized lung transplant recipients with respiratory syncytial virus infection.

Hungarian researchers have built a human induced pluripotent stem cell platform that reproduces the sensitivity of healthy dividing tissues to chemotherapy, revealing that stem cells and neural progenitors are far more vulnerable to anticancer drugs than conventional cancer cell lines suggest.

As major health systems open their own programs, longevity medicine is gaining credibility and becoming more accessible.

arXiv:2608.02712v2 Announce Type: replace Abstract: Kernel generation for hardware accelerators such as GPUs and NPUs has become a proving ground for large language models (LLMs), and state-of-the-art systems raise correctness through pipelines that couple LLMs with agentic reinforcement learning and evolutionary search. Such pipelines generate, compile, and execute large numbers of candidate kernels, discarding most of them and forgoing the opportunity to distill failures into reusable knowledge. Many discarded candidates are near-miss operators that compile and run but fail numerical validation; each embodies genuine domain knowledge and a nontrivial investment in LLM inference, cross-compilation, and hardware execution. We argue for a paradigm shift: rather than regenerate, debug. Debugging is far more constrained than generating from scratch: the search space is small and feedback is dense. We present a domain-specific debug agent that addresses three core challenges in autonomous

arXiv:2610.21402v1 Announce Type: new Abstract: Optical Wireless Communication (OWC) systems provide high-bandwidth, spatially secure, and RF-interference-free digital transmission. However, direct free-space optical links suffer from extreme sensitivity to line-of-sight disruptions, optical attenuation, high ambient light saturation, and transmitter-receiver misalignment. Traditional channel diagnostic frameworks rely on external computing nodes or cloud-based processing, which introduces severe diagnostic latency and reduces edge node autonomy. This paper presents an autonomous, self-diagnosing optical wireless receiver node that integrates physical-layer signal telemetry extraction with an on-device TinyML classification engine. Built around an ESP32-S3 microcontroller, the receiver recovers 1000 bit/s On-Off Keying (OOK) digital payloads using a 5-sample majority voting mechanism while concurrently extracting real-time channel features: Bit Error Rate (BER), received signal

Li Renke, a visionary cardiovascular scientist who applied engineering principles to the human heart and developed conductive biomaterials to repair tissue damaged by heart attacks, died while attending an academic conference in the Chinese city of Xian earlier this month. He was 65. His death on October 10 was announced by The Chinese University of Hong Kong, Shenzhen (CUHK-Shenzhen), where he was presidential chair professor and associate dean of research and translation at the school of...

A multicenter Chinese study shows that an MRI-based nomogram combining clinical signs with AI-derived tumor habitat features can noninvasively predict high-risk hepatocellular carcinoma before surgery and stratify patients' risk of recurrence.

A YOLOv8-based deep learning model identified and measured children's coronary arteries on echocardiograms with expert-level accuracy in under two seconds per measurement.

A Columbia Nursing survey of 1,185 U.S. adults found that 63 percent would not endorse AI-generated medical responses without clinician review, though most accepted fast AI replies for logistical tasks.

A new study reports that exosomes released by inflamed synovial stem cells carry the circular RNA circRNF20, which worsens rheumatoid arthritis by activating a TLR4, neutrophil trap, and interleukin-6 signaling pathway.

Человечество всегда пыталось передать себя через время. Но до сих пор мы передавали потомкам главным образом культурный след человека: его тексты, образы, поступки и память о нём. Теперь впервые появляется возможность передавать способность отвечать, выбирать и действовать. Это уже не архив. Это начало новой субъектности. Читать и обсуждать

An updated systematic review of seven randomised trials finds that exercise rehabilitation after haematopoietic stem cell transplantation is safe and improves walking distance and leg strength, with the greatest benefit seen in allogeneic recipients.

A rat study using fluorescent tracer dextrans shows that transplanted intestines preserve normal gut motility four weeks after surgery while developing markedly increased paracellular permeability and expanded villus surface area.

A prospective two-phase observer-performance study found that a confidence-aware AI system with a radiologist-editable interface improved radiologists' accuracy and agreement in distinguishing the two most common benign parotid tumors on ultrasound.

Karl Deisseroth, Peter Hegemann and Georg Nagel share the 2026 Nobel Prize in Physiology or Medicine for discoveries on light-gated ion channels that founded optogenetics.

A new commentary argues that for low-resource populations, healthcare access rather than prompt literacy is the primary inequality shaping whether conversational AI can safely support health self-management.

Two brands of HRT patches, including Estradot transdermal and FemSeven Conti patches, have been affected.

Exposure to common chemicals may alter how normal human breast tissue develops, according to a new study from researchers at Tufts University School of Medicine and Boston University

Jonathan the giant tortoise was born when Darwin was still a young man. Youthful gene activity may account for his exceptionally long life. The post The Oldest Living Land Animal Is 194. DNA May Now Reveal the Secret to His Longevity. appeared first on SingularityHub.

In a retrospective analysis of PET-CT records from 101,243 patients, researchers found that detectable brown fat was associated with lower prevalence of cardiometabolic and psychiatric conditions, while greater activity, particularly in the axillary region, was linked to more favourable metabolic outcomes. Medication screening identified losartan, which increased brown fat thermogenic activity in laboratory experiments and energy expenditure in mice, although its effects on human brown fat remain unproven.

Researchers have uncovered a previously overlooked immune cell that can recognize and kill small cell lung cancer, the most aggressive form of the disease globally. Published in Cancer Cell, the WEHI study found patients with higher levels of these cells lived longer, uncovering a hidden immune defense that could be exploited to attack these hard-to-treat tumors.

A systematic review of 88 studies shows radar and camera-based IoT systems dominate contactless vital-sign monitoring, while blood pressure, oxygen saturation, and temperature remain largely out of reach.

An opinion article argues that the activin-trap drug sotatercept, proven in pulmonary arterial hypertension and supported by the CADENCE trial in heart failure, should now be tested in staged trials across advanced heart failure, LVAD support, and heart transplantation.

Mount Sinai sent lab-grown heart and brain tissue models to the space station on SpaceX CRS-35 to study whether spaceflight speeds aging-like changes.

A proteomic study of the phase 1/2a delandistrogene moxeparvovec trial shows that high-dose AAV gene therapy triggers a sharp but self-limiting immune response even in boys without serious adverse events.

A biophysics-informed machine learning framework predicts how effectively antisense oligonucleotides knock down ALS-relevant genes using only a single experimental measurement per new target family.

More than 30 years after the direct connection was established between the molecular machinery controlling cell division and cancer, a new paper traces the evolution of cyclin-dependent kinase research and outlines the next generation of CDK-targeted therapies.

A systematic review of 91 studies shows AI-based medication recognition advancing rapidly through detection, OCR, and few-shot learning, while external validation and clinical testing lag behind.

A new analysis of more than 81,000 liver transplants found that female recipients aged 25 and older show higher excess mortality than male recipients in the first three months after transplant, regardless of donor sex. Researchers believe the poorer condition of female patients at the time of transplant may explain the early disadvantage, which reverses after the first year.

A first-of-its-kind mixed-methods study reveals how perioperative adverse events inflict lasting psychological and professional tolls on kidney transplant surgeons, shaping risk attitudes, decision-making and the availability of life-saving transplants.

INmune Bio and Quantum Computing Inc. partner to explore how AI and quantum computing can advance Alzheimer's drug research. The post INmune Bio Inc. and Quantum Computing Inc. Partner to Apply Quantum Optimization and AI to Advance Alzheimer’s Drug Research and Explore New Applications for XPro appeared first on TechRepublic.

THURSDAY, Oct. 15, 2026 — Kidneys donated by smokers are less likely to survive transplant, a new study says.Donated kidneys are 13% more likely to fail if they come from a person with any history of smoking, and the odds are even worse for organs...

I flew to San Francisco yesterday to take part in Rock Health’s annual summit. It’s been about half a year since I last visited, so the event served as a good gut check of where ...

Transplant patients in Poland who ate ice cream during high-dose chemo got painful mouth sores about half as often. The cold may be what matters.

A prospective South Korean cohort study found that plasma microbial cell-free DNA metagenomic sequencing substantially agreed with conventional microbiology when clinically relevant specimens were considered and detected additional bacterial and viral pathogens missed by routine testing in liver transplant recipients.

The growing use of AI tools is changing the skills doctors need to practice medicine, experts said during a panel at an Endpoints News event on Wednesday. Doctors no longer need to memorize as many ...

INmune Bio Inc. (Nasdaq: INMB) and Quantum Computing Inc. today announced a collaboration to apply quantum computing, artificial intelligence (AI) and data analytics to identify new potential applications for XPro (XPro1595), INmune Bio's investigational treatment for Alzheimer's disease and other disease conditions.

Researchers here assess the present animal study evidence for first generation stem cell therapies to produce benefits in the most common neurodegenerative conditions, and also the consequences of stroke. The results are promising. While these therapies have been fairly widely deployed in patients via medical tourism, that data will never be available for analysis in any useful way. A few formal clinical trials have taken place, with unfortunately mixed results, which suggests that the details of manufacturing process, source of cells, and other logistical factors are important to outcomes. In general, these stem cell therapies appear to reliably modulate chronic inflammation, but all other desired outcomes are quite unreliable where they have been assessed rigorously in human patients. This is one of the reasons why […]

arXiv:2610.20504v1 Announce Type: new Abstract: Long-horizon precise manipulation produces mixed-quality data for post-training vision-language-action (VLA) models: even successful demonstrations and intervention-assisted rollouts may contain detours, hesitations, failed attempts, and imperfect recoveries. The key challenge is to identify which action chunks actually advance the task, rather than uniformly cloning or discarding imperfect trajectories. We propose Graph-Reweighted Behavior Cloning (GR-BC), an offline iterative post-training framework that builds a task-conditioned goal-reachability graph over demonstrations and iteratively collected rollouts, including both successful and failed episodes with human interventions. Reverse graph search computes minimal remaining steps to task success within the dataset support, and each action chunk is weighted by how much its execution reduces this potential. This consequence-based weighting reuses recoverable segments from failed

arXiv:2610.20203v1 Announce Type: new Abstract: This paper investigates the use of Behavior Cloning (BC) for autonomous driving from a bird's-eye view (BEV) perspective in simulated urban environments. BC uses supervised learning to mimic expert driving behaviors. Previous works have applied BC in the CARLA simulator but did not fully address the challenges of traffic light compliance. Our approach enhances BC by integrating a kernel density estimator to adjust training sample weights based on action density, thereby improving the learning of rare but critical actions such as stopping at red lights and accelerating at green lights, specially in scenarios of scarce number of expert demonstrations. Using BEV inputs, which provide an abstract top-down view of the driving environment, our method simplifies the policy learning process. The trained convolutional neural network (CNN) outputs steering and acceleration actions based on these BEV inputs and additional state variables.

arXiv:2610.19774v1 Announce Type: new Abstract: Zero-shot text-to-speech (TTS) typically conditions on a short speech prompt, which conveys timbre but provides only limited evidence of a speaker's prosodic patterns. We formulate Long-Prompt Zero-Shot TTS, a task that uses long speech prompts at inference time to reproduce speaker-specific prosody without target-speaker fine-tuning. We introduce LP-TTS-Bench, a bilingual benchmark of 37 speakers with 63.95 prompt hours and 2,728 held-out test utterances. Beyond intelligibility and timbre similarity, LP-TTS-Bench evaluates timing, prominence, intonation, and model-based prosody metrics, enabling multidimensional evaluation of prosody reproduction. As a reference system, we develop LPTTS, whose hierarchical Long-Prompt Encoder compresses long speech prompts into a fixed number of continuous prosody vectors. Experiments show improvements in several prosodic dimensions over a short-prompt baseline, with overall gains as prompt duration

arXiv:2610.19600v1 Announce Type: new Abstract: Recent vision-language-action (VLA) models have shown strong performance on local robot manipulation, yet their success does not readily translate to long-horizon mobile manipulation. In our experiments, a pretrained subtask VLA achieves 80% success rate on atomic manipulation tasks but only 19% when these skills are composed into mobile manipulation. We argue that this gap stems from failures across the execution process rather than from local manipulation capability alone. To uncover these bottlenecks, we systematically diagnose failures in long-horizon mobile manipulation and identify three dominant regimes: task planning, navigation and base positioning, and local manipulation. Based on this diagnosis, we develop failure-guided agentic intervention that augments a fixed subtask VLA with demonstration-based planning guidance, adaptive visual memory, accessibility-aware base adjustment, and local end-effector adjustment. On

arXiv:2610.19445v1 Announce Type: new Abstract: Neural instrument synthesizers can clone the timbre of a solo instrument, letting music producers make music with the sound they desire. However, an instrument of interest is often embedded in a fully produced, multi-instrument recording, beyond the reach of existing methods. Source separation before timbre cloning is an approach, but it may suffer from separation artifacts and is also likely a detour for the task. In this work, we propose Point-and-Play, a system that clones the timbre of a cued source directly from a polyphonic mixture and uses it to synthesize new pitches and velocities. To achieve this, we introduce a Cue-Gated Statistics Pooling (CGSP) module that extracts the timbre code of the cued instrument from adapted CLAP embeddings of the music mixture. This extracted timbre code is then used to condition a separately trained note synthesizer together with pitch and velocity inputs. Experiments show that our proposed system

A new framework combines a platform-free federated learning blockchain consensus mechanism with optimized neural networks to secure electronic health record storage, sharing, and disease diagnosis.

Researchers used a temperature-tolerant LbCas12a genome editing system to restore the reading frame of a loss-of-function SlGLK2 allele in tomato, recovering the green shoulder phenotype and increasing soluble solids and tocopherol content in ripe fruit.

A retrospective Chinese cohort study shows that clinically relevant contamination of organ preservation fluid, especially when matched to donor respiratory organisms, sharply raises the risk of early donor-derived infections after liver and kidney transplantation.

A cross-sectional study of 160 Turkish children found that higher adherence to the Mediterranean diet was independently associated with fewer cavities and healthier gums, while higher BMI was linked to poorer periodontal health.

A nationwide Taiwanese cohort study finds that older cancer survivors who return to work show significantly better long-term survival than those who do not, across all cancer stages.

Stem Cell Reports has been named the Official Meeting Journal of the ASSCR 2026 Annual Scientific Meeting, taking place 9-11 November 2026 in Melbourne, Australia.

Issues affecting ageing populations worldwide will form the core of discussions for policymakers and health industry leaders gathered at the Abu Dhabi Future Health Summit, with a particular emphasis on how practising prediction and prevention can help ensure life is lived to its fullest. The United Arab Emirates (UAE) has positioned itself as a global leader in transformative healthcare policy, and Dr Asma Al Mannaei, executive director of the health life science sector at the Department of...

A JAMA study suggests a lab-based urgency score could rank heart transplant candidates by how sick they are, better than today's rules. What it could mean.

A leakage-controlled multimodal AI model combining PET/CT radiomics, deep-learning features, metabolic parameters and blood-based indices achieved moderate preoperative prediction of axillary lymph-node metastasis in breast cancer, with no statistically proven added value of imaging over clinical variables.

A combined spectral CT and radiomics nomogram achieved over 0.9 area under the curve in distinguishing pancreatic neuroendocrine tumors from more aggressive carcinomas before surgery.

A new PLOS Digital Health study maps international and EU ethical and regulatory requirements for AI-based medical imaging diagnostics into five categories and warns that security gaps and variable compliance could undermine patient autonomy, safety, and justice.

A five-center study shows deep learning models, especially nnU-Net, predict aggressive meningioma grades from MRI more reliably across hospitals than handcrafted radiomics, which collapsed under external validation.

In a Columbia University School of Nursing study surveying 1,185 U.S.-based adults about AI use in secure medical portals, 63% of patients said they would not endorse AI-generated medical responses without a clinician's review. This suggests that while patients see value in using AI for certain health care tasks, they prefer a human review when medical advice is involved. The study, "Public Perceptions of Timeliness for AI Use and Human Oversight," was published in the JAMA Network Open on Oct. 14, 2026.

Researchers at the University of Cincinnati College of Medicine have published the first proof-of-concept study showing the potential of an experimental gene therapy for hydrocephalus.

A new multimodal AI framework called MEGH-DDI combines chemical language models with graph neural networks to predict drug–drug interactions with over 98 percent accuracy.

Researchers at the University of Cincinnati College of Medicine have published the first proof-of-concept study showing the potential of an experimental gene therapy for hydrocephalus. The preclinical research from the lab of Agnes (Yu) Luo was recently published in the journal Fluids and Barriers of the CNS, which focuses on specialized cerebrospinal fluid (CSF) biology.

Researchers report that suppressing the lipid-metabolizing enzyme PLA2G7 in hippocampal neural stem cells restores inhibitory synapses and improves memory in aged mice.

A University of Cincinnati and UC Davis team showed that an AAV gene therapy targeting cerebrospinal fluid-producing cells reduced hydrocephalus-related brain swelling in animal models, pointing toward a minimally invasive alternative to shunt surgery.

For a long time, differences between females and males were mainly associated with the reproductive organs. Yet organs that have nothing to do with reproduction, such as the skin, can also be shaped by sex.

Every organ and tissue progressively loses function and integrity as it ages. In fact, researchers have identified "hallmarks of aging" that not only present specific processes impacted by age, but are also thought of as levers that, if successfully targeted, could help turn back the aging clock and result in tissue rejuvenation.

The on-demand availability of artificial intelligence chatbots means that most individuals have quick, easy access to a wealth of information. The ease and accessibility of these tools may make it tempting to turn to AI for advice about mental health. But it's important to note that these systems are prone to reinforcing a user's existing beliefs or, in some cases, providing inaccurate information.

His comments come after Merck officially opened its new 120,000-square-foot research facility in the Greater Philadelphia area.

The pancreas is a complex organ comprising multiple cell types that regulate blood glucose levels through hormones, such as insulin, and support digestion by producing digestive enzymes. Researchers at the University of Tsukuba previously developed an intraplacental transplantation method whereby pancreatic progenitor cells are injected into the placenta of a pregnant mouse and delivered to the developing fetus. However, whether transplanted cells could become incorporated into pancreatic tissue had not been demonstrated.

NASHVILLE, Tenn. — AI tools were less likely to flag emergencies and deliver accurate, comprehensive gastroenterology advice in the six most common U.S. languages after English, according to a presenter at ACG Annual Scientific Meeting.“AI is already part of everyday life, and people are turning to it with health questions, often before speaking with a physician,” Harika Dadigiri, MD, resident physician at New York Medical College at Saint Mary’s and Saint Clare’s Hospital in New Jersey, told Healio regarding her presentation. “For patients who struggle to find reliable medical information in

A new review in Cellular and Molecular Life Sciences proposes hybrid biological-digital tumor twins, coupling patient-derived organoids with computational models in closed feedback loops, as a blueprint for adaptive precision oncology in breast cancer.

OpenEvidence and Johns Hopkins Medicine partner to develop AI tools for physician training and clinical care, with a focus on medical reasoning and patient safety. The post OpenEvidence, Johns Hopkins Partner to Bring AI Into Medical Training and Patient Care appeared first on TechRepublic.

The American Society of Hematology (ASH) has released guidelines on the management of myelofibrosis, a rare and life-threatening blood cancer. The new guidelines recommend expanded molecular testing, early transplant referral, appropriate use of JAK inhibitors and aspirin therapy in select patients.

Researchers combined structure-guided protein design with cis-regulatory editing at the native HPPD locus to create elite herbicide-tolerance alleles in rice and wheat.

Researchers in Wuhan developed a two-stage deep learning system that detects and classifies postoperative coughs from audio recordings of lung surgery patients with over 99 percent accuracy on clean data, while flagging noise robustness as the key hurdle before clinical deployment.

A new Journal of Medical Systems commentary proposes a task-level oversight-margin framework to quantify when human supervision of cardiovascular artificial intelligence falls behind the pace of algorithmic output.