The notion that oncologists could boost immunotherapy responses simply by giving infusions in the morning, rather than late afternoon, is an attractive one. So when a clinical trial published in Nature Medicine this month showed that lung cancer patients treated in the morning had a massive reduction in the risk of progression compared to those treated in the afternoon, many scientists were intrigued, if skeptical.

Now that study is coming under fire, as multiple scientists and sleuths raise serious concerns about the data and point out inconsistencies in the trial.

These have called the study’s conclusions even further into question, which experts told STAT already lacked strong biological plausibility, and Nature Medicine appended a note on the study on Thursday that it is starting an investigation into the concerns.

Continue to STAT+ to read the full story…

© Jenny Kane/AP

arXiv:2602.17308v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly used for diagnostic tasks in medicine. In clinical practice, the correct diagnosis can rarely be immediately inferred from the initial patient presentation alone. Rather, reaching a diagnosis often involves systematic history taking, during which clinicians reason over multiple potential conditions through iterative questioning to resolve uncertainty. This process requires considering differential diagnoses and actively excluding emergencies that demand immediate intervention. Yet, the ability of medical LLMs to generate informative follow-up questions and thus reason over differential diagnoses remains underexplored. Here, we introduce MedClarify, an AI agent for information-seeking that can generate follow-up questions for iterative reasoning to support diagnostic decision-making. Specifically, MedClarify computes a list of candidate diagnoses analogous to a differential diagnosis, and then proactively generates follow-up questions aimed at reducing diagnostic uncertainty. By selecting the question with the highest expected information gain, MedClarify enables targeted, uncertainty-aware reasoning to improve diagnostic performance. In our experiments, we first demonstrate the limitations of current LLMs in medical reasoning, which often yield multiple, similarly likely diagnoses, especially when patient cases are incomplete or relevant information for diagnosis is missing. We then show that our information-theoretic reasoning approach can generate effective follow-up questioning and thereby reduces diagnostic errors by ~27 percentage points (p.p.) compared to a standard single-shot LLM baseline. Altogether, MedClarify offers a path to improve medical LLMs through agentic information-seeking and to thus promote effective dialogues with medical LLMs that reflect the iterative and uncertain nature of real-world clinical reasoning.

arXiv:2602.16729v1 Announce Type: cross Abstract: We systematically evaluate the quality of widely used AI safety datasets from two perspectives: in isolation and in practice. In isolation, we examine how well these datasets reflect real-world attacks based on three key properties: driven by ulterior intent, well-crafted, and out-of-distribution. We find that these datasets overrely on "triggering cues": words or phrases with overt negative/sensitive connotations that are intended to trigger safety mechanisms explicitly, which is unrealistic compared to real-world attacks. In practice, we evaluate whether these datasets genuinely measure safety risks or merely provoke refusals through triggering cues. To explore this, we introduce "intent laundering": a procedure that abstracts away triggering cues from attacks (data points) while strictly preserving their malicious intent and all relevant details. Our results indicate that current AI safety datasets fail to faithfully represent real-world attacks due to their overreliance on triggering cues. In fact, once these cues are removed, all previously evaluated "reasonably safe" models become unsafe, including Gemini 3 Pro and Claude Sonnet 3.7. Moreover, when intent laundering is adapted as a jailbreaking technique, it consistently achieves high attack success rates, ranging from 90% to over 98%, under fully black-box access. Overall, our findings expose a significant disconnect between how model safety is evaluated and how real-world adversaries behave.

arXiv:2602.17568v1 Announce Type: cross Abstract: Normalization and scaling are fundamental preprocessing steps in time series modeling, yet their role in Transformer-based models remains underexplored from a theoretical perspective. In this work, we present the first formal analysis of how different normalization strategies, specifically instance-based and global scaling, impact the expressivity of Transformer-based architectures for time series representation learning. We propose a novel expressivity framework tailored to time series, which quantifies a model's ability to distinguish between similar and dissimilar inputs in the representation space. Using this framework, we derive theoretical bounds for two widely used normalization methods: Standard and Min-Max scaling. Our analysis reveals that the choice of normalization strategy can significantly influence the model's representational capacity, depending on the task and data characteristics. We complement our theory with empirical validation on classification and forecasting benchmarks using multiple Transformer-based models. Our results show that no single normalization method consistently outperforms others, and in some cases, omitting normalization entirely leads to superior performance. These findings highlight the critical role of preprocessing in time series learning and motivate the need for more principled normalization strategies tailored to specific tasks and datasets.

Nature Medicine, Published online: 20 February 2026; doi:10.1038/d41591-026-00011-9

When used as intended, an AI stethoscope enhanced the detection of cardiovascular disease — but low uptake and workflow challenges hampered effectiveness in a large pragmatic trial.

Nature Medicine, Published online: 20 February 2026; doi:10.1038/s41591-026-04230-y

How to accelerate cures for genetic diseases using an artificial intelligence-enabled discovery engine to identify druggable nodes and develop clinical targets.

npj Digital Medicine, Published online: 20 February 2026; doi:10.1038/s41746-026-02452-5

Artificial intelligence–enhanced microsurgical training: a systematic review

npj Digital Medicine, Published online: 20 February 2026; doi:10.1038/s41746-026-02441-8

CancerLLM: a large language model in cancer domain

A blood test for detecting cancer early being developed by the diagnostics firm Grail failed to meet its main goal in a giant study being conducted with England’s National Health Service, the company said Thursday.

Grail’s test has been the standard bearer for new technologies that promise a blood test can be used to detect many different types of cancer early and eventually even to indicate to scientists where in the body to look for tumors. The company already sells its test, called Galleri, for a list price of $1,000, although it is not yet approved by the Food and Drug Administration. Grail said Thursday it sold 185,000 tests in 2025, generating $136.8 million. 

The company’s shares were down 47% in after-hours trading.

Continue to STAT+ to read the full story…

© Adobe

You’re reading the web edition of STAT’s Health Tech newsletter, our guide to how technology is transforming the life sciences. Sign up to get it delivered in your inbox every Tuesday and Thursday.

Good morning health tech readers!

Please join me in congratulating my colleagues as STAT, who have been honored with a fourth Polk Award for our coverage of the Trump administration’s impacts on the federal health department and American science. The award recognizes the whole newsroom and in particular the work of Lizzy Lawrence covering a dramatic year of changes at the Food and Drug Administration. 

Continue to STAT+ to read the full story…

© Adobe

Nature, Published online: 18 February 2026; doi:10.1038/d41586-026-00290-9

An artificial-intelligence system uses clinical data, genetic information and literature searches to suggest diagnoses and provides the underlying reasoning.

arXiv:2602.15019v1 Announce Type: new Abstract: Bio-pharmaceutical innovation has shifted: many new drug assets now originate outside the United States and are disclosed primarily via regional, non-English channels. Recent data suggests >85% of patent filings originate outside the U.S., with China accounting for nearly half of the global total; a growing share of scholarly output is also non-U.S. Industry estimates put China at ~30% of global drug development, spanning 1,200+ novel candidates. In this high-stakes environment, failing to surface "under-the-radar" assets creates multi-billion-dollar risk for investors and business development teams, making asset scouting a coverage-critical competition where speed and completeness drive value. Yet today's Deep Research AI agents still lag human experts in achieving high-recall discovery across heterogeneous, multilingual sources without hallucinations. We propose a benchmarking methodology for drug asset scouting and a tuned, tree-based self-learning Bioptic Agent aimed at complete, non-hallucinated scouting. We construct a challenging completeness benchmark using a multilingual multi-agent pipeline: complex user queries paired with ground-truth assets that are largely outside U.S.-centric radar. To reflect real deal complexity, we collected screening queries from expert investors, BD, and VC professionals and used them as priors to conditionally generate benchmark queries. For grading, we use LLM-as-judge evaluation calibrated to expert opinions. We compare Bioptic Agent against Claude Opus 4.6, OpenAI GPT-5.2 Pro, Perplexity Deep Research, Gemini 3 Pro + Deep Research, and Exa Websets. Bioptic Agent achieves 79.7% F1 versus 56.2% (Claude Opus 4.6), 50.6% (Gemini 3 Pro + Deep Research), 46.6% (GPT-5.2 Pro), 44.2% (Perplexity Deep Research), and 26.9% (Exa Websets). Performance improves steeply with additional compute, supporting the view that more compute yields better results.

arXiv:2602.13332v1 Announce Type: cross Abstract: Long-form clinical videos are central to visual evidence-based decision-making, with growing importance for applications such as surgical robotics and related settings. However, current multimodal large language models typically process videos with passive sampling or weakly grounded inspection, which limits their ability to iteratively locate, verify, and justify predictions with temporally targeted evidence. To close this gap, we propose MedScope, a tool-using clinical video reasoning model that performs coarse-to-fine evidence seeking over long-form procedures. By interleaving intermediate reasoning with targeted tool calls and verification on retrieved observations, MedScope produces more accurate and trustworthy predictions that are explicitly grounded in temporally localized visual evidence. To address the lack of high-fidelity supervision, we build ClinVideoSuite, an evidence-centric, fine-grained clinical video suite. We then optimize MedScope with Grounding-Aware Group Relative Policy Optimization (GA-GRPO), which directly reinforces tool use with grounding-aligned rewards and evidence-weighted advantages. On full and fine-grained video understanding benchmarks, MedScope achieves state-of-the-art performance in both in-domain and out-of-domain evaluations. Our approach illuminates a path toward medical AI agents that can genuinely "think with videos" through tool-integrated reasoning. We will release our code, models, and data.

npj Digital Medicine, Published online: 17 February 2026; doi:10.1038/s41746-026-02438-3

Empowering genetic discoveries and cardiovascular risk assessment by predicting electrocardiograms from genotype

npj Digital Medicine, Published online: 16 February 2026; doi:10.1038/s41746-026-02437-4

Randomized-controlled trial of skills-based vr vs. distraction vr vs. sham VR for chronic low back pain

An operational AI forecasting model developed by Hertfordshire University researchers aims to improve resource efficiency within healthcare.

Public sector organisations often hold large archives of historical data that do not inform forward-looking decisions. A partnership between the University of Hertfordshire and regional NHS health bodies addresses this issue by applying machine learning to operational planning. The project analyses healthcare demand to assist managers with decisions regarding staffing, patient care, and resources.

Most AI initiatives in healthcare focus on individual diagnostics or patient-level interventions. The project team notes that this tool targets system-wide operational management instead. This distinction matters for leaders evaluating where to deploy automated analysis within their own infrastructure.

The model uses five years of historical data to build its projections. It integrates metrics such as admissions, treatments, re-admissions, bed capacity, and infrastructure pressures. The system also accounts for workforce availability and local demographic factors including age, gender, ethnicity, and deprivation.

Iosif Mporas, Professor of Signal Processing and Machine Learning at the University of Hertfordshire, leads the project. The team includes two full-time postdoctoral researchers and will continue development through 2026.

“By working together with the NHS, we are creating tools that can forecast what will happen if no action is taken and quantify the impact of a changing regional demographic on NHS resources,” said Professor Mporas.

Using AI for forecasting in healthcare operations

The model produces forecasts showing how healthcare demand is likely to change. It models the impact of these changes in the short-, medium-, and long-term. This capability allows leadership to move beyond reactive management.

Charlotte Mullins, Strategic Programme Manager for NHS Herts and West Essex, commented: “The strategic modelling of demand can affect everything from patient outcomes including the increased number of patients living with chronic conditions.

“Used properly, this tool could enable NHS leaders to take more proactive decisions and enable delivery of the 10-year plan articulated within the Central East Integrated Care Board as our strategy document.” 

The University of Hertfordshire Integrated Care System partnership funds the work, which began last year. Testing of the AI model tailored for healthcare operations is currently underway in hospital settings. The project roadmap includes extending the model to community services and care homes.

This expansion aligns with structural changes in the region. The Hertfordshire and West Essex Integrated Care Board serves 1.6 million residents and is preparing to merge with two neighbouring boards. This merger will create the Central East Integrated Care Board. The next phase of development will incorporate data from this wider population to improve the predictive accuracy of the model.

The initiative demonstrates how legacy data can drive cost efficiencies and shows that predictive models can inform “do nothing” assessments and resource allocation in complex service environments like the NHS. The project highlights the necessity of integrating varied data sources – from workforce numbers to population health trends – to create a unified view for decision-making.

See also: Agentic AI in healthcare: How Life Sciences marketing could achieve $450B in value by 2028

Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is part of TechEx and is co-located with other leading technology events including the Cyber Security & Cloud Expo. Click here for more information.

AI News is powered by TechForge Media. Explore other upcoming enterprise technology events and webinars here.

The post AI forecasting model targets healthcare resource efficiency appeared first on AI News.