❌

Reading view

↗️

A Decision-Theoretic Approach for Managing Misalignment

✇cs.AI, q-bio.NC updates on arXiv.org
By: Daniel A. Herrmann · Abinav Chari · Isabelle Qian · Sree Sharvesh · B. A. Levinstein
arXiv:2512.15584v1 Announce Type: new Abstract: When should we delegate decisions to AI systems? While the value alignment literature has developed techniques for shaping AI values, less attention has been paid to how to determine, under uncertainty, when imperfect alignment is good enough to justify delegation. We argue that rational delegation requires balancing an agent's value (mis)alignment with its epistemic accuracy and its reach (the acts it has available). This paper introduces a formal, decision-theoretic framework to analyze this tradeoff precisely accounting for a principal's uncertainty about these factors. Our analysis reveals a sharp distinction between two delegation scenarios. First, universal delegation (trusting an agent with any problem) demands near-perfect value alignment and total epistemic trust, conditions rarely met in practice. Second, we show that context-specific delegation can be optimal even with significant misalignment. An agent's superior accuracy or expanded reach may grant access to better overall decision problems, making delegation rational in expectation. We develop a novel scoring framework to quantify this ex ante decision. Ultimately, our work provides a principled method for determining when an AI is aligned enough for a given context, shifting the focus from achieving perfect alignment to managing the risks and rewards of delegation under uncertainty.
  •  
  • ↗️
↗️

DrugRAG: Enhancing Pharmacy LLM Performance Through A Novel Retrieval-Augmented Generation Pipeline

✇cs.AI, q-bio.NC updates on arXiv.org
By: Houman Kazemzadeh · Kiarash Mokhtari Dizaji · Seyed Reza Tavakoli · Farbod Davoodi · MohammadReza KarimiNejad · Parham Abed Azad · Ali Sabzi · Armin Khosravi · Siavash Ahmadi · Mohammad Hossein Rohban · Glolamali Aminian · Tahereh Javaheri
arXiv:2512.14896v1 Announce Type: cross Abstract: Objectives: To evaluate large language model (LLM) performance on pharmacy licensure-style question-answering (QA) tasks and develop an external knowledge integration method to improve their accuracy. Methods: We benchmarked eleven existing LLMs with varying parameter sizes (8 billion to 70+ billion) using a 141-question pharmacy dataset. We measured baseline accuracy for each model without modification. We then developed a three-step retrieval-augmented generation (RAG) pipeline, DrugRAG, that retrieves structured drug knowledge from validated sources and augments model prompts with evidence-based context. This pipeline operates externally to the models, requiring no changes to model architecture or parameters. Results: Baseline accuracy ranged from 46% to 92%, with GPT-5 (92%) and o3 (89%) achieving the highest scores. Models with fewer than 8 billion parameters scored below 50%. DrugRAG improved accuracy across all tested models, with gains ranging from 7 to 21 percentage points (e.g., Gemma 3 27B: 61% to 71%, Llama 3.1 8B: 46% to 67%) on the 141-item benchmark. Conclusion: We demonstrate that external structured drug knowledge integration through DrugRAG measurably improves LLM accuracy on pharmacy tasks without modifying the underlying models. This approach provides a practical pipeline for enhancing pharmacy-focused AI applications with evidence-based information.
  •  
  • ↗️
↗️

MedChat: A Multi-Agent Framework for Multimodal Diagnosis with Large Language Models

✇cs.AI, q-bio.NC updates on arXiv.org
By: Philip R. Liu · Sparsh Bansal · Jimmy Dinh · Aditya Pawar · Ramani Satishkumar · Shail Desai · Neeraj Gupta · Xin Wang · Shu Hu
arXiv:2506.07400v3 Announce Type: replace-cross Abstract: The integration of deep learning-based glaucoma detection with large language models (LLMs) presents an automated strategy to mitigate ophthalmologist shortages and improve clinical reporting efficiency. However, applying general LLMs to medical imaging remains challenging due to hallucinations, limited interpretability, and insufficient domain-specific medical knowledge, which can potentially reduce clinical accuracy. Although recent approaches combining imaging models with LLM reasoning have improved reporting, they typically rely on a single generalist agent, restricting their capacity to emulate the diverse and complex reasoning found in multidisciplinary medical teams. To address these limitations, we propose MedChat, a multi-agent diagnostic framework and platform that combines specialized vision models with multiple role-specific LLM agents, all coordinated by a director agent. This design enhances reliability, reduces hallucination risk, and enables interactive diagnostic reporting through an interface tailored for clinical review and educational use. Code available at https://github.com/Purdue-M2/MedChat.
  •  
  • ↗️
↗️

Multimodal Foundation Models for Early Disease Detection

✇cs.AI, q-bio.NC updates on arXiv.org
By: Md Talha Mohsin · Ismail Abdulrashid
arXiv:2510.01899v2 Announce Type: replace-cross Abstract: Healthcare data now span EHRs, medical imaging, genomics, and wearable sensors, but most diagnostic models still process these modalities in isolation. This limits their ability to capture early, cross-modal disease signatures. This paper introduces a multimodal foundation model built on a transformer architecture that integrates heterogeneous clinical data through modality-specific encoders and cross-modal attention. Each modality is mapped into a shared latent space and fused using multi-head attention with residual normalization. We implement the framework using a multimodal dataset that simulates early-stage disease patterns across EHR sequences, imaging patches, genomic profiles, and wearable signals, including missing-modality scenarios and label noise. The model is trained using supervised classification together with self-supervised reconstruction and contrastive alignment to improve robustness. Experimental evaluation demonstrates strong performance in early-detection settings, with stable classification metrics, reliable uncertainty estimates, and interpretable attention patterns. The approach moves toward a flexible, pretrain-and-fine-tune foundation model that supports precision diagnostics, handles incomplete inputs, and improves early disease detection across oncology, cardiology, and neurology applications.
  •  
  • ↗️
↗️

A novel statistical feature selection framework for biomarker discovery and cancer classification via multiomics integration

✇(Multiomics OR Omics) AND (Lung OR gastric OR Hepatocellular)
By: Moshira S Ghaleb · Maryam N Al-Berry · Hala M Ebied · Mohamed F Tolba

BMC Med Res Methodol. 2025 Dec 17. doi: 10.1186/s12874-025-02713-z. Online ahead of print.

ABSTRACT

BACKGROUND: Early cancer diagnosis is essential for improving prognosis and guiding treatment. However, the high dimensionality and complexity of omics data present major challenges. Computational approaches that extract stable biomarkers and enable reliable classification across cancer types and stages are needed.

METHODS: A novel feature selection method, sDCFE (synergistic Discriminative Cluster-based Feature Extraction), was developed by extending Fisher-like variance analysis with a median absolute deviation (MAD) regularization term and a cluster separation component to enhance robustness and interpretability. Features selected by sDCFE were compared with those obtained from XGBoost, and the intersected set of 82 genes was evaluated through functional enrichment (KEGG, Reactome, GO BP), survival analysis (Kaplan-Meier, Cox regression), and biomarker novelty assessment against six external resources. Hybrid classification models integrating XGBoost, sDCFE, and deep learning were applied to pancancer classification, and the framework was further extended to lung squamous cell carcinoma (LUSC) staging using RNA-seq and methylation data.

RESULTS: The overlap between sDCFE and XGBoost yielded 82 candidate biomarkers enriched in cancer-related pathways, including cell cycle regulation, immune signalling, and DNA repair. Novelty assessment stratified these genes into established, emerging, and novel categories. Six genes-HFE2, LOC339674, SERINC2, SFTA3, SOX2OT, and ACPP-emerged as the most promising candidates, supported by enrichment and survival associations across multiple cancers. The hybrid model achieved near-perfect pancancer classification on TCGA (accuracy = 99.3%, MCC = 0.992, AUC = 1.0) and demonstrated strong generalizability on PCAWG (accuracy = 94%, MCC = 0.929, AUC = 0.997). In the LUSC staging task, multiomics integration improved classification performance: the CNN-based model reached 84% accuracy, while logistic regression applied to sDCFE-ranked features achieved 88.5% accuracy with superior calibration, highlighting the robustness of the selected features.

CONCLUSION: sDCFE provides a principled extension of Fisher-like methods, enabling stable and interpretable biomarker selection. When combined with XGBoost and deep learning, the framework achieves highly accurate and biologically grounded cancer classification across both cancer types and stages. The identification of novel and prognostic biomarkers, including HFE2, LOC339674, SERINC2, SFTA3, SOX2OT, and ACPP, underscores its translational potential. These results position the framework as a promising precision oncology tool to support early diagnosis, risk stratification, and treatment decision-making.

PMID:41408184 | DOI:10.1186/s12874-025-02713-z

  •  
  • ↗️
↗️

Enhancing Transparency and Traceability in Healthcare AI: The AI Product Passport

✇cs.AI, q-bio.NC updates on arXiv.org
By: A. Anil Sinaci · Senan Postaci · Dogukan Cavdaroglu · Machteld J. Boonstra · Okan Mercan · Kerem Yilmaz · Gokce B. Laleci Erturkmen · Folkert W. Asselbergs · Karim Lekadir
arXiv:2512.13702v1 Announce Type: cross Abstract: Objective: To develop the AI Product Passport, a standards-based framework improving transparency, traceability, and compliance in healthcare AI via lifecycle-based documentation. Materials and Methods: The AI Product Passport was developed within the AI4HF project, focusing on heart failure AI tools. We analyzed regulatory frameworks (EU AI Act, FDA guidelines) and existing standards to design a relational data model capturing metadata across AI lifecycle phases: study definition, dataset preparation, model generation/evaluation, deployment/monitoring, and passport generation. MLOps/ModelOps concepts were integrated for operational relevance. Co-creation involved feedback from AI4HF consortium and a Lisbon workshop with 21 diverse stakeholders, evaluated via Mentimeter polls. The open-source platform was implemented with Python libraries for automated provenance tracking. Results: The AI Product Passport was designed based on existing standards and methods with well-defined lifecycle management and role-based access. Its implementation is a web-based platform with a relational data model supporting auditable documentation. It generates machine- and human-readable reports, customizable for stakeholders. It aligns with FUTURE-AI principles (Fairness, Universality, Traceability, Usability, Robustness, Explainability), ensuring fairness, traceability, and usability. Exported passports detail model purpose, data provenance, performance, and deployment context. GitHub-hosted backend/frontend codebases enhance accessibility. Discussion and Conclusion: The AI Product Passport addresses transparency gaps in healthcare AI, meeting regulatory and ethical demands. Its open-source nature and alignment with standards foster trust and adaptability. Future enhancements include FAIR data principles and FHIR integration for improved interoperability, promoting responsible AI deployment.
  •  
  • ↗️
↗️

Graph AI generates neurological hypotheses validated in molecular, organoid, and clinical systems

✇cs.AI, q-bio.NC updates on arXiv.org
By: Ayush Noori · Joaqu\'in Polonuer · Katharina Meyer · Bogdan Budnik · Shad Morton · Xinyuan Wang · Sumaiya Nazeen · Yingnan He · I\~naki Arango · Lucas Vittor · Matthew Woodworth · Richard C. Krolewski · Michelle M. Li · Ninning Liu · Tushar Kamath · Evan Macosko · Dylan Ritter · Jalwa Afroz · Alexander B. H. Henderson · Lorenz Studer · Samuel G. Rodriques · Andrew White · Noa Dagan · David A. Clifton · George M. Church · Sudeshna Das · Jenny M. Tam · Vikram Khurana · Marinka Zitnik
arXiv:2512.13724v1 Announce Type: cross Abstract: Neurological diseases are the leading global cause of disability, yet most lack disease-modifying treatments. We present PROTON, a heterogeneous graph transformer that generates testable hypotheses across molecular, organoid, and clinical systems. To evaluate PROTON, we apply it to Parkinson's disease (PD), bipolar disorder (BD), and Alzheimer's disease (AD). In PD, PROTON linked genetic risk loci to genes essential for dopaminergic neuron survival and predicted pesticides toxic to patient-derived neurons, including the insecticide endosulfan, which ranked within the top 1.29% of predictions. In silico screens performed by PROTON reproduced six genome-wide $\alpha$-synuclein experiments, including a split-ubiquitin yeast two-hybrid system (normalized enrichment score [NES] = 2.30, FDR-adjusted $p
  •  
  • ↗️
↗️

Criminal Liability in AI-Enabled Autonomous Vehicles: A Comparative Study

✇cs.AI, q-bio.NC updates on arXiv.org
By: Sahibpreet Singh · Manjit Singh
arXiv:2512.14330v1 Announce Type: cross Abstract: AI revolutionizes transportation through autonomous vehicles (AVs) but introduces complex criminal liability issues regarding infractions. This study employs a comparative legal analysis of primary statutes, real-world liability claims, and academic literature across the US, Germany, UK, China, and India; jurisdictions selected for their technological advancement and contrasting regulatory approaches. The research examines the attribution of human error, AI moral agency, and the identification of primary offenders in AV incidents. Findings reveal fragmented regulatory landscapes: India and the US rely on loose networks of state laws, whereas the UK enacted the pioneering Automated and Electric Vehicles Act 2018. Germany enforces strict safety standards, distinguishing liability based on the vehicle's operating mode, while China similarly aims for a stringent liability regime. The study concludes that globally harmonized legal standards are essential to foster technological innovation while ensuring minimum risk and clear liability attribution.
  •  
  • ↗️
↗️

A Multicenter Benchmark of Multiple Instance Learning Models for Lymphoma Subtyping from HE-stained Whole Slide Images

✇cs.AI, q-bio.NC updates on arXiv.org
By: Rao Muhammad Umer · Daniel Sens · Jonathan Noll · Christian Matek · Lukas Wolfseher · Rainer Spang · Ralf Huss · Johannes Raffler · Sarah Reinke · Wolfram Klapper · Katja Steiger · Kristina Schwamborn · Carsten Marr
arXiv:2512.14640v1 Announce Type: cross Abstract: Timely and accurate lymphoma diagnosis is essential for guiding cancer treatment. Standard diagnostic practice combines hematoxylin and eosin (HE)-stained whole slide images with immunohistochemistry, flow cytometry, and molecular genetic tests to determine lymphoma subtypes, a process requiring costly equipment, skilled personnel, and causing treatment delays. Deep learning methods could assist pathologists by extracting diagnostic information from routinely available HE-stained slides, yet comprehensive benchmarks for lymphoma subtyping on multicenter data are lacking. In this work, we present the first multicenter lymphoma benchmarking dataset covering four common lymphoma subtypes and healthy control tissue. We systematically evaluate five publicly available pathology foundation models (H-optimus-1, H0-mini, Virchow2, UNI2, Titan) combined with attention-based (AB-MIL) and transformer-based (TransMIL) multiple instance learning aggregators across three magnifications (10x, 20x, 40x). On in-distribution test sets, models achieve multiclass balanced accuracies exceeding 80% across all magnifications, with all foundation models performing similarly and both aggregation methods showing comparable results. The magnification study reveals that 40x resolution is sufficient, with no performance gains from higher resolutions or cross-magnification aggregation. However, on out-of-distribution test sets, performance drops substantially to around 60%, highlighting significant generalization challenges. To advance the field, larger multicenter studies covering additional rare lymphoma subtypes are needed. We provide an automated benchmarking pipeline to facilitate such future research.
  •  
  • ↗️
↗️

COMMA: A Communicative Multimodal Multi-Agent Benchmark

✇cs.AI, q-bio.NC updates on arXiv.org
By: Timothy Ossowski · Danyal Maqbool · Jixuan Chen · Zefan Cai · Tyler Bradshaw · Junjie Hu
arXiv:2410.07553v5 Announce Type: replace Abstract: The rapid advances of multimodal agents built on large foundation models have largely overlooked their potential for language-based communication between agents in collaborative tasks. This oversight presents a critical gap in understanding their effectiveness in real-world deployments, particularly when communicating with humans. Existing agentic benchmarks fail to address key aspects of inter-agent communication and collaboration, particularly in scenarios where agents have unequal access to information and must work together to achieve tasks beyond the scope of individual capabilities. To fill this gap, we introduce COMMA: a novel puzzle benchmark designed to evaluate the collaborative performance of multimodal multi-agent systems through language communication. Our benchmark features a variety of multimodal puzzles, providing a comprehensive evaluation across four key categories of agentic capability in a communicative collaboration setting. Our findings reveal surprising weaknesses in state-of-the-art models, including strong proprietary models like GPT-4o and reasoning models like o4-mini. Many chain of thought reasoning models such as R1-Onevision and LLaVA-CoT struggle to outperform even a random baseline in agent-agent collaboration, indicating a potential growth area in their communication abilities.
  •  
  • ↗️
↗️

A Knowledge Graph-based Retrieval-Augmented Generation Framework for Algorithm Selection in the Facility Layout Problem

✇cs.AI, q-bio.NC updates on arXiv.org
By: Nikhil N S (Indian Institute of Science · Bengaluru · India) · Bilal Muhammed (TCS Research and Innovation · Tata Consultancy Services Ltd) · Soban Babu Beemaraj (TCS Research and Innovation · Tata Consultancy Services Ltd) · Amol Dilip Joshi (TCS Research and Innovation · Tata Consultancy Services Ltd)
arXiv:2509.18054v2 Announce Type: replace-cross Abstract: Selecting a solution algorithm for the Facility Layout Problem (FLP), an NP-hard optimization problem with multiobjective trade-off, is a complex task that requires deep expert knowledge. The performance of a given algorithm depends on the specific characteristics of the problem, such as the number of facilities, objectives, and constraints. This creates a need for a data-driven recommendation method to guide algorithm selection in automated design systems. This paper introduces a new recommendation method to make this expertise accessible, based on a Knowledge Graph-Based Retrieval-Augmented Generation (KG-RAG) framework. In this framework, a domain-specific knowledge graph (KG) is constructed from the literature. The method then employs a multifaceted retrieval mechanism to gather relevant evidence from this KG using three distinct approaches: precise graph-based search, flexible vector-based search, and cluster-based high-level search. The retrieved evidence is utilized by a Large Language Model (LLM) to generate algorithm recommendations based on data-driven reasoning. This KG-RAG framework is tested on a use case consisting of six problems comprising of complex multi-objective and multi-constraint FLP case. The results are compared with the Gemini 1.5 Flash chatbot. The results show that KG-RAG achieves an average reasoning score of 4.7 out of 5 compared to 3.3 for the baseline chatbot.
  •  
  • ↗️
↗️

Beyond Task Completion: An Assessment Framework for Evaluating Agentic AI Systems

✇cs.AI, q-bio.NC updates on arXiv.org
By: Sreemaee Akshathala · Bassam Adnan · Mahisha Ramesh · Karthik Vaidhyanathan · Basil Muhammed · Kannan Parthasarathy
arXiv:2512.12791v2 Announce Type: replace-cross Abstract: Recent advances in agentic AI have shifted the focus from standalone Large Language Models (LLMs) to integrated systems that combine LLMs with tools, memory, and other agents to perform complex tasks. These multi-agent architectures enable coordinated reasoning, planning, and execution across diverse domains, allowing agents to collaboratively automate complex workflows. Despite these advances, evaluation and assessment of LLM agents and the multi-agent systems they constitute remain a fundamental challenge. Although various approaches have been proposed in the software engineering literature for evaluating conventional software components, existing methods for AI-based systems often overlook the non-deterministic nature of models. This non-determinism introduces behavioral uncertainty during execution, yet existing evaluations rely on binary task completion metrics that fail to capture it. Evaluating agentic systems therefore requires examining additional dimensions, including the agent ability to invoke tools, ingest and retrieve memory, collaborate with other agents, and interact effectively with its environment. These challenges emerged during our ongoing industry collaboration with MontyCloud Inc., when we deployed an agentic system in production. These limitations surfaced during deployment, highlighting practical gaps in the current evaluation methods and the need for a systematic assessment of agent behavior beyond task outcomes. Informed by these observations and established definitions of agentic systems, we propose an end-to-end Agent Assessment Framework with four evaluation pillars encompassing LLMs, Memory, Tools, and Environment. We validate the framework on a representative Autonomous CloudOps use case, where experiments reveal behavioral deviations overlooked by conventional metrics, demonstrating its effectiveness in capturing runtime uncertainties.
  •  
  • ↗️
↗️

Artificial Intelligence Platform Architecture for Hospital Systems: Systematic Review

✇Journal of Medical Internet Research
By: Musitapa Maimaitiaili · Yiershatijiang Jiamaliding · Guangle Dai · Hui Xiao · Warisijiang Kuerbanjiang · Yuexiong Yi
Background: The construction of artificial intelligence (AI) platforms in hospitals forms the basis of the modern healthcare revolution. While traditional hospital information systems have facilitated digitalization, they are still limited by data siloes, fragmented workflows and insufficient clinical intelligence that impede organizations from realizing the promise of data-led decision-making. Objective: This review aims to provide a strategic roadmap for hospitals to build comprehensive AI platforms, moving beyond siloed AI applications toward infrastructure at the system level that supports sustainable, scalable, and interoperable intelligent services across clinical, operational, and administrative domains. Methods: A systematic literature search was performed in Web of Science, EMBASE, PubMed, and Scopus. The review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Studies were screened and selected for full text review by two independent reviewers with reference to AI platform construction, hospital informatics integration, and institutional deployment strategies. Results: A total of 30 high-quality studies were included in the final analysis. Based on the synthesis of evidence, a five-layer hospital AI platform architecture is proposed, consisting of: (1) infrastructure layer, (2) data layer, (3) algorithm layer, (4) application layer, and (5) security and compliance layer. The review highlights key implementation strategies such as modular deployment, real-world scenario validation, and interdepartmental collaboration. It also identifies critical challenges, including legacy system integration, lack of data standardization, computing resource limitations, organizational resistance, regulatory uncertainty, and economic sustainability. Conclusions: The successful construction of hospital AI platforms requires not only advanced technologies but also institutional readiness, strategic planning, and cultural adaptation. Intelligent hospitals of the future must emphasize privacy-preserving computing, seamless AI integration into clinical workflows, and dynamic performance evaluation systems. Building organizational capacity and fostering cross-disciplinary collaboration will be essential to achieving long-term impact and scalability.
  •  
  • ↗️
↗️

Single-cell and spatial transcriptomic characterization of pulmonary pleomorphic carcinoma

✇Omics In Lung
By: Atsushi Matsuoka · Kazuhiko Shien · Shuta Tomida · Masayoshi Ohki · Kazuya Hisamatsu · Ryota Fujiwara · Kosei Ishimura · Ryunosuke Fujii · Tomoaki Higashihara · Naohiro Hayashi · Kazuhiro Okada · Ryo Yoshichika · Fumiaki Mukohara · Mao Yoshikawa · Yuma Fukumoto · Ken Suzawa · Yasuaki Tomioka · Shin Tanaka · Kentaroh Miyoshi · Mikio Okazaki · Seiichiro Sugimoto · Yusuke Otani · Atsushi Tanaka · Hirofumi Inoue · Yosuke Togashi · Hidetaka Yamamoto · Daisuke Ennishi · Shinichi Toyooka

Commun Biol. 2025 Dec 16;8(1):1773. doi: 10.1038/s42003-025-09162-w.

ABSTRACT

Pulmonary pleomorphic carcinoma (PPC) is a rare subtype of lung cancer that comprises both epithelial and sarcomatoid components. The molecular basis of PPC, including the cellular dynamics of its components, remains largely unknown. To elucidate potential therapeutic targets for PPC, we perform a multi-omics analysis incorporating digital spatial profiling and single-cell RNA sequencing (scRNA-seq). PPC exhibits diverse driver gene alterations, including MET exon 14 skipping mutation (METex14) and ALK fusion. In spatial transcriptomics, MET gene and protein are overexpressed exclusively within the epithelial component and not in the sarcomatoid component, even in patients harboring METex14. Epithelial-mesenchymal transition (EMT)-related transcriptional changes, along with extracellular matrix (ECM) remodeling between the epithelial and sarcomatoid components, are observed. scRNA-seq identifies cell populations within the epithelial component that contribute to the malignant transformation and differentiation of the sarcomatoid component. They are characterized by an intermediate EMT state with ECM remodeling signature, suggesting their potential as novel therapeutic targets for PPC.

PMID:41402584 | PMC:PMC12708732 | DOI:10.1038/s42003-025-09162-w

  •  
  • ↗️
↗️

STAT+: Key digital health and device leaders depart FDA

✇STAT
By: Mario Aguilar

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.

A tumultuous year at the Food and Drug Administration will be capped off at the agency’s devices center with the departure of two key leaders, just as regulators are sorting through challenges related to artificial intelligence and launching new initiatives on software as a medical device regulation. 

  • Sources tell us Jessica Paulsen, a 15-year veteran of FDA and acting deputy director of its Digital Health Center of Excellence is leaving the agency. She’s been leading the center since last summer when the last acting head, SonjaFulmer, left FDA for Mayo Clinic. Fulmer took over for Troy Tazbaz who left in January to return to Oracle. The center’s work includes communicating with industry and developing guidances relevant to digital health. (FDA did not respond to a request for comment.)
  • Neuralink, Elon Musk’s frothy brain-computer interface company, poached David McMullen, director of FDA’s office of neurological and physical medicine devices, which is in charge of regulating Neuralink. McMullen spent three years atop the office and previously worked at the National Institute for Mental Health.
  • Both Paulsen and McMullen were at the forefront of important conversations about the future of regulation. I grabbed the screenshot above of the two leaders from a video of last month’s Digital Health Advisory Committee meeting on generative AI-enabled mental health devices. Separately, McMullen’s office will have oversight of behavioral health devices under the FDA’s new TEMPO pilot.
  • New to me: As part of the funding package that reopened the government last month, lawmakers passed full-year 2026 funding for FDA. Buried within the Senate report accompanying the legislation, lawmakers direct FDA to, within 90 days, (February) report on its authorities to regulate AI medical devices, and within 180 days, (May) report on “the status of the  FDA’s efforts regarding engagement on AI in drug development.” 
  • The Government Accountability Office last week released a report on medical device recalls which found, among other things, that “insufficient staff limit FDA’s ability to conduct oversight activities.” In other words, the FDA already does not have enough staff to oversee medical devices and is losing key leadership at a time when new technology and initiatives may require additional horsepower.

The future of the mammogram

Applying AI to mammograms to help radiologists spot signs of breast cancer is increasingly common but researchers and AI companies want to apply new analyses to the routine screening tests to trigger more proactive care to prevent future cancers, heart attacks, and strokes. In one important breakthrough, the startup Clairity received FDA authorization for AI that offer a prediction of somone’s five-year breast cancer risk based on a mammogram alone.

Continue to STAT+ to read the full story…

© Adobe

  •  
  • ↗️
↗️

World Models Unlock Optimal Foraging Strategies in Reinforcement Learning Agents

✇cs.AI, q-bio.NC updates on arXiv.org
By: Yesid Fonseca · Manuel S. R\'ios · Nicanor Quijano · Luis F. Giraldo
arXiv:2512.12548v1 Announce Type: new Abstract: Patch foraging involves the deliberate and planned process of determining the optimal time to depart from a resource-rich region and investigate potentially more beneficial alternatives. The Marginal Value Theorem (MVT) is frequently used to characterize this process, offering an optimality model for such foraging behaviors. Although this model has been widely used to make predictions in behavioral ecology, discovering the computational mechanisms that facilitate the emergence of optimal patch-foraging decisions in biological foragers remains under investigation. Here, we show that artificial foragers equipped with learned world models naturally converge to MVT-aligned strategies. Using a model-based reinforcement learning agent that acquires a parsimonious predictive representation of its environment, we demonstrate that anticipatory capabilities, rather than reward maximization alone, drive efficient patch-leaving behavior. Compared with standard model-free RL agents, these model-based agents exhibit decision patterns similar to many of their biological counterparts, suggesting that predictive world models can serve as a foundation for more explainable and biologically grounded decision-making in AI systems. Overall, our findings highlight the value of ecological optimality principles for advancing interpretable and adaptive AI.
  •  
  • ↗️
↗️

Value-Aware Multiagent Systems

✇cs.AI, q-bio.NC updates on arXiv.org
By: Nardine Osman
arXiv:2512.12652v1 Announce Type: new Abstract: This paper introduces the concept of value awareness in AI, which goes beyond the traditional value-alignment problem. Our definition of value awareness presents us with a concise and simplified roadmap for engineering value-aware AI. The roadmap is structured around three core pillars: (1) learning and representing human values using formal semantics, (2) ensuring the value alignment of both individual agents and multiagent systems, and (3) providing value-based explainability on behaviour. The paper presents a selection of our ongoing work on some of these topics, along with applications to real-life domains.
  •  
  • ↗️
↗️

Explainable AI as a Double-Edged Sword in Dermatology: The Impact on Clinicians versus The Public

✇cs.AI, q-bio.NC updates on arXiv.org
By: Xuhai Xu · Haoyu Hu · Haoran Zhang · Will Ke Wang · Reina Wang · Luis R. Soenksen · Omar Badri · Sheharbano Jafry · Elise Burger · Lotanna Nwandu · Apoorva Mehta · Erik P. Duhaime · Asif Qasim · Hause Lin · Janis Pereira · Jonathan Hershon · Paulius Mui · Alejandro A. Gru · No\'emie Elhadad · Lena Mamykina · Matthew Groh · Philipp Tschandl · Roxana Daneshjou · Marzyeh Ghassemi
arXiv:2512.12500v1 Announce Type: cross Abstract: Artificial intelligence (AI) is increasingly permeating healthcare, from physician assistants to consumer applications. Since AI algorithm's opacity challenges human interaction, explainable AI (XAI) addresses this by providing AI decision-making insight, but evidence suggests XAI can paradoxically induce over-reliance or bias. We present results from two large-scale experiments (623 lay people; 153 primary care physicians, PCPs) combining a fairness-based diagnosis AI model and different XAI explanations to examine how XAI assistance, particularly multimodal large language models (LLMs), influences diagnostic performance. AI assistance balanced across skin tones improved accuracy and reduced diagnostic disparities. However, LLM explanations yielded divergent effects: lay users showed higher automation bias - accuracy boosted when AI was correct, reduced when AI erred - while experienced PCPs remained resilient, benefiting irrespective of AI accuracy. Presenting AI suggestions first also led to worse outcomes when the AI was incorrect for both groups. These findings highlight XAI's varying impact based on expertise and timing, underscoring LLMs as a "double-edged sword" in medical AI and informing future human-AI collaborative system design.
  •  
  • ↗️
❌