Skip to content

Let's Data ScienceLEARN • BUILD • STAY AHEAD

News
Blog
Code Problems
Pricing
Contact

© 2026 Let's Data Science

Advertise|Terms|Privacy||Image Rights

Home
News
Healthcare AI

Topic20 articles · Updated June 23, 2026

Healthcare AI News

AI across healthcare: clinical diagnostic models, AI-driven drug discovery, hospital deployments, FDA activity, and the regulatory and payer changes that determine what actually reaches patients.

medicare
7.1
Jun 23, 2026
Medicare Pilots AI Prior Authorization, Causes Delays
Federal Medicare began a six-state pilot, the Wasteful and Inappropriate Service Reduction Model (WISeR), that uses AI-assisted reviews for prior authorization of certain procedures, according to KFF Health News and other reporting. WISeR launched in mid-January and applies to services including epidural steroid injections, knee procedures, and nerve stimulators in Arizona, New Jersey, Ohio, Oklahoma, Texas, and Washington, per KFF. The pilot affects roughly 6.4 million traditional Medicare beneficiaries, Yahoo reporting noted. Early reporting and a study cited by Senator Maria Cantwell found approvals that previously took about two weeks are now taking four to eight weeks, producing rescheduling, longer travel for patients, and reported administrative confusion, according to Healthcare Dive, FierceHealthcare, and KFF. CMS contracts with private firms to run "enhanced technologies, including artificial intelligence," for reviews, Yahoo Finance reports.
Quick Read+6
healthcare-ai
5.8
Jun 22, 2026
LLMs integrate clinical knowledge for LNM prediction
Large language models and machine learning predictions are combined in a newly developed knowledge-augmented framework to predict lymph node metastasis (LNM) in lung cancer. The framework integrates clinical knowledge with model outputs using LLMs to produce LNM predictions intended to support initial treatment decision-making.
Quick Read
pandemic-detection
6.2
Jun 21, 2026
Clalit Joins European AI Pandemic Research Consortium
Clalit Health Services, Israel's largest healthcare provider, has been selected as a partner in PANDAI (European Pandemics AI Observatory), an EU-funded initiative to build the first AI platform for early pandemic detection, prediction, and management. The project, backed by Horizon Europe with approximately EUR 8 million in funding, links Clalit with the World Health Organization (WHO), the University of Oxford, and research partners from the UK, Spain, Denmark, Luxembourg, and Bangladesh, according to the Jerusalem Post and The Jewish Chronicle. Prof. Eytan Wirtheim, CEO of Clalit, said the achievement 'demonstrates that the knowledge, research, and innovation developed by Clalit and the Israeli healthcare system continue to play an important role in the global effort to shape the future of medicine.' For practitioners, the consortium raises questions around cross-border data governance, federated learning, and privacy-preserving analytics at scale.
Quick Read
joint-commission
6.8
Jun 18, 2026
Joint Commission launches AI responsibility certification
The Joint Commission launched the Responsible Use of AI in Healthcare (RUAIH) certification in early June, a voluntary program that recognizes organizations that can demonstrate governance, effective data management, risk and bias reduction, monitoring and safety evaluation, and education and training, according to the Joint Commission website and a June 2 Fierce Healthcare report. The certification does not validate individual AI products, Fierce Healthcare reports, and any healthcare organization may apply without already holding Joint Commission accreditation, per Fierce. The Coalition for Health AI (CHAI) published implementation playbooks; Fierce Healthcare and the RUAIH guidance document describe those playbooks as mapping to the certification requirements.
Quick Read+3
kff-poll
5.8
Jun 18, 2026
Americans Increase Use of AI For Health Advice
According to a KFF tracking poll published June 17 (May 7-31 survey, n=2,480), 29% of U.S. adults now use AI tools or chatbots such as ChatGPT, Google Gemini, and Claude for health information or advice on a monthly basis, nearly doubling from roughly one in six in June 2024. About 31% use social media for health information at least once per month, and roughly one in six report doing so "every day," according to KFF. The majority of adults say they "never" or only "occasionally" use AI tools (71%) or social media (69%) for medical information. KFF also finds that among those using social media for health advice, fewer than four in ten follow up with a doctor (36%), consult another online source such as WebMD (35%), or check health agency websites like the CDC (21%). KFF notes that affordability and access barriers are a significant driver for younger and lower-income adults turning to AI and social media for health information.
Quick Read+2
healthcare-ai
7.4
Jun 17, 2026
AI Outperforms Doctors in Hospital Diagnostics
The Atlantic reports that a team of primarily Harvard and Stanford researchers announced results of a study published in April in the journal Science that compared o1 preview (OpenAI's step-by-step reasoning model) against hundreds of physicians in a diagnostic obstacle course using written medical cases and real-world patient information, and that the model outperformed the clinicians. Multiple outlets including Harvard Magazine and STAT News confirm the study tested OpenAI's o1 reasoning model on 76 real emergency room cases from a Boston hospital, finding it matched or exceeded expert physician performance across triage, initial assessment, and admission stages. The Atlantic quotes co-senior author Adam Rodman (Harvard Medical School, Beth Israel Deaconess) saying, "I get a little bit queasy about how some of these results might be used," and co-senior author Arjun Manrai emphasized the results do not mean AI replaces doctors, per Harvard Magazine. The article notes that at least one generative-AI product has received FDA approval.
Quick Read+2
medical-imaging
6.3
Jun 17, 2026
China Extends National Medical Imaging AI Contest to ASEAN
ANTARA/Xinhua-AsiaNet reports that the National Medical Insurance Imaging AI Recognition Competition, jointly hosted by China's National Healthcare Security Administration and the People's Government of Guangxi Zhuang Autonomous Region, held a promotional event in Hanoi on June 10, 2026 following a seminar in Malaysia in May, according to GlobeNewswire and ANTARA. ANTARA reports the contest is scheduled to take place in Guangxi from August to October 2026 and features eight professional tracks across diagnostic scenarios including lung, breast, glioma, kidney and thyroid cancers, using multimodal imaging (CT, MRI, CTA, X-ray, ultrasound). ANTARA and Vietnam.vn report the organisers will provide annotated datasets reviewed by medical experts, and that data have undergone standardized anonymization and privacy procedures. The promotional events targeted universities, hospitals, research teams and companies across ASEAN, according to ANTARA and Vietnam.vn.
Quick Read+3
healthcare-ai
4.8
Jun 14, 2026
KFSH Highlights Human-Centered AI for Better Care
According to a KFSH press release distributed via GlobeNewswire, King Faisal Specialist Hospital & Research Centre (KFSH) will sponsor and participate in HLTH Europe 2026 in Amsterdam, June 15-18, joining the workshop "AI, Trust & Human Impact: What Healthcare and Global Brands Can Learn From Each Other." The release states KFSH's Centre for Healthcare Intelligence, established in 2019, has supported the development of 20 locally powered AI applications. The press materials report the hospital's Patient Flow and Capacity Command Centre has executed more than 170,000 interventions and reduced average bed waiting times from 32 hours to 6 hours. The release also highlights KFSH's Smart Neuroscience Ward and cites rankings from Brand Finance and Newsweek. Editorial analysis: These claims present operational metrics that, if externally validated, illustrate the kinds of measurable throughput and patient-experience gains health systems cite when arguing for production AI deployments.
Quick Read+1
nvidia
6.8
Jun 11, 2026
Nvidia Develops Clinical Transcription Model With Abridge
Nvidia and startup Abridge are collaborating to develop an AI model tailored for doctor-patient conversations, the Wall Street Journal reported. The model will be trained on Nvidia's open models, called Nemotron, and will be deployed exclusively within Abridge's transcription app, according to the Wall Street Journal. Nvidia is an investor in Abridge, and Abridge was valued at $5.3 billion last year, the report said. Abridge's co-founder and CEO Dr. Shiv Rao told the Wall Street Journal that the company will use de-identified clinical data to further train and customize the Nemotron models. PYMNTS reported the model is expected to launch later this year, citing interviews with company executives.
Quick Read
healthcare-ai
5.6
Jun 9, 2026
Austrian Physicians Identify Barriers to AI Breast Diagnostics
Per the now-published study by Kritz, Holawatsch, Behrens, and Hyll in the Journal of Medical Internet Research (J Med Internet Res 2026;28:e80274), researchers ran a cross-sectional, embedded mixed-methods survey of 46 Austrian physicians on barriers, facilitators, and intention to use AI for breast cancer diagnosis. About half (50%) already used AI tools, and both users and non-users expressed strong future-adoption intentions; non-users more often cited limited access (81%), high costs, and lack of training. Quantitative items drew on the Technology Acceptance Model (TAM), open-ended responses were coded via content analysis, and ordinary least squares regressions linked perceived facilitators, AI-related skills, and supportive colleagues to stronger intention, while perceived barriers and older age (50+) predicted more negative attitudes. The authors frame adoption as a staged process needing different support at each stage.
Quick Read
tempus-ai
5.4
Jun 9, 2026
Tempus AI Presents Asymmetric Upside From Data Moat
Seeking Alpha published a "Strong Buy" thesis on Tempus AI (NASDAQ: TEM), arguing that a proprietary clinical data moat and expanding pharma partnerships give the healthcare-AI company asymmetric upside. Per Seeking Alpha, Diagnostics revenue was $261M in Q1 2026 at a 62.3% gross margin, while Data and Applications grew about 41% year-over-year at a 73.1% margin, and the company holds more than 45M de-identified records and over 500 petabytes of data. The thesis cites FY2026 guidance of $1.59B-$1.6B revenue and roughly $65M adjusted EBITDA, and flags risks including continued unprofitability, about $1.3B of debt, and roughly 20% short interest. Tempus' official Q1 2026 results, reported May 5, corroborate the segment trend: total revenue of $348.1M (up 36.1%), Diagnostics of $261.1M, and Data and Applications of $87.0M.
Quick Read+1
healthcare-ai
4.8
Jun 8, 2026
Clinicians Build AI Prototypes in Two-Hour Hackathon
The San Diego chapter of the Society of Physician Entrepreneurs (SoPE) held its inaugural Physician AI Hackathon on May 30, 2026, at the University of California, San Diego, per a PRWeb release. According to the release, seven teams, each pairing at least one practicing clinician with engineers and students, built working clinical software prototypes in under two hours. The event was hosted by Royan Kamyar, M.D., MBA, who leads the chapter and is founder and CEO of sponsor Owaves, and was presented with UC San Diego's Engineering Innovators and Entrepreneurs Club in partnership with AIMed and UCSD's Office of Innovation and Commercialization. The release names Manifesto AI, a perioperative coordination tool built as a read-only layer on top of Epic, as the first-place winner, and GutGuide, a patient-education app, as second place. Event listings corroborate the half-day, in-person 'Idea to App in Two Hours' format.
Quick Read+2
healthcare-ai
6.8
Jun 5, 2026
Pennsylvania Sues Character.AI Over Medical Chatbot Claims
Pennsylvania has sued Character.AI, alleging one of its chatbots unlawfully posed as a licensed medical professional. According to the state's filing, reported by TechCrunch and NPR, a user-made character named "Emilie" presented itself as a licensed psychiatrist during testing by a state Professional Conduct Investigator, kept up the pretense while the investigator described depression, and fabricated a serial number for a Pennsylvania medical license when asked. Governor Josh Shapiro's administration says the conduct violates the state's Medical Practice Act, in what reporting describes as the first U.S. enforcement action focused specifically on chatbots impersonating medical professionals. Character.AI says user-created characters are fictional and that every chat carries disclaimers warning they are not real people. A companion analysis in The Conversation, by a Carnegie Mellon behavioral-science researcher, frames the case around how people calibrate trust in AI versus human experts.
Quick Read+2
radiology
6.0
Jun 4, 2026
Multi-pass LLM Improves Radiology Report Error Detection
A study now published in JMIR Medical Informatics reports a three-pass large language model (LLM) framework that raises precision for radiology-report proofreading while holding detection rates steady. Per the paper (also posted as arXiv:2506.20112), positive predictive value (PPV) rose from 0.063 (single-prompt baseline) to 0.159 (three-pass; 95% CI 0.090-0.252) on a 1,000-report test set drawn from MIMIC-III, while absolute true positive rate (aTPR) stayed near 0.012-0.014. The authors estimate per-1,000-report inference-plus-reviewer costs falling to USD 5.58 for the three-pass pipeline from USD 9.72 for the single-prompt baseline, with reports needing human review dropping from 192 to 88. External validation on CheXpert and Open-i supported higher PPV (CheXpert 0.133, Open-i 0.105) with stable aTPR. The authors frame the cascade as a practical way to cut false positives and reviewer workload in AI-assisted radiology quality assurance.
Quick Read+1
healthcare-ai
7.2
Jun 4, 2026
Mayo Clinic and Microsoft build healthcare frontier model
According to Microsoft and Mayo Clinic announcements on June 2, 2026, the two organizations will jointly develop and deploy a frontier AI model purpose-built for healthcare. The collaboration pairs Mayo Clinic's de-identified clinical data, longitudinal insights and clinical expertise with Microsoft's AI, cloud and engineering capabilities to support broad clinical reasoning and healthcare use cases, per Microsoft, Becker's Hospital Review and Fierce Healthcare. Crucially, Mayo Clinic will own the model, while Microsoft plans to distribute it through Azure Foundry APIs to healthcare organizations worldwide. Mayo Clinic president and CEO Gianrico Farrugia frames the effort as an extension of the Mayo Clinic Platform built on a patient-first, de-identified-data approach. Becker's reports the model will first be deployed inside Mayo Clinic's clinical environment for continuous testing and refinement. The announcements disclose no model architecture, training details or validation benchmarks.
Quick Read+10
sentri7
6.6
Jun 3, 2026
Sentri7 Missed Fentanyl Thefts at Tennessee Hospital
A nurse at Erlanger Baroness Hospital in Chattanooga, Tennessee, diverted fentanyl for months in 2025 before coworkers, not the hospital's AI monitoring, noticed he appeared impaired, according to a Tennessee Board of Nursing order reported by KFF Health News and CBS News. The hospital uses Sentri7, an AI drug-diversion monitoring product from Wolters Kluwer. KFF reports a hospital diversion specialist later found about five instances of missing fentanyl waste that Sentri7 did not flag, and that the software was in its initial learning phase at the time. KFF says Erlanger declined to comment on Sentri7, and a Wolters Kluwer spokesperson declined detailed answers while saying the company remains confident in its software. The report notes hospitals are not required to disclose use of such tools or report malfunctions. Editorial analysis: rare-event detection and data-quality gaps commonly limit anomaly-detection systems, underscoring the need for independent validation and human oversight.
Quick Read+2
healthcare-ai
5.4
May 27, 2026
Study Examines AI's Role in Patient-Centered Care
A qualitative interview study titled "Artificial Intelligence in Patient-Centered Care and Macro-, Meso-, and Micro-Level Determinants of Rehumanization and Dehumanization" examines interactions between Artificial Intelligence and patient-centered care, mapping macro-, meso-, and micro-level determinants of rehumanization and dehumanization. The study's background states the digital transformation of health systems has accelerated adoption of AI in care delivery.
Quick Read
iks-health
6.8
May 22, 2026
IKS Health Acquires ARAI Solutions to Expand AI Capabilities
Per a Business Wire press release (May 13, 2026), IKS Health has acquired ARAI Solutions, an AI management and technology company that builds biomedical knowledge graphs and clinical ontologies. Business Wire and CityBiz report that ARAI's ontology and reasoning engine will be integrated into IKS Health's four-layer healthcare AI stack to accelerate development in autonomous coding, clinical decisions, denial prevention, prior authorization reasoning, and precision medicine. CityBiz and HitConsultant note that financial terms were not disclosed and that the deal is intended to reduce reliance on third-party large language model infrastructure. Industry context: Domain-specific knowledge graphs and ontologies are increasingly used to improve explainability, traceability, and regulatory alignment in clinical AI applications.
Quick Read+9
physical-ai
6.9
May 22, 2026
Kawasaki Establishes Physical AI Center with Nvidia and Partners
Kawasaki Heavy Industries announced the establishment of the Kawasaki Physical AI Center San Jose, a Silicon Valley hub for "physical AI," per a company press release dated May 22, 2026. Kawasaki named NVIDIA, Analog Devices, Microsoft, and Fujitsu as partners for the center, which the press release says will accelerate Japan-U.S. collaboration in AI and semiconductors. The press release states the center will initially focus on healthcare and elder care, offering end-to-end hospital solutions that integrate robotics and AI. Reuters, citing the Nikkei, earlier reported plans for a joint development center in San Jose and said the collaboration will apply Nvidia simulation technology to Kawasaki robotics platforms such as Corleo.
Quick Read+9
revenue-cycle-management
6.8
May 21, 2026
Innovaccer Acquires CaduceusHealth to Expand RCM
Innovaccer announced an asset acquisition of CaduceusHealth, a revenue cycle management (RCM) services provider for ambulatory care, in a Business Wire press release. Per the release, the deal is Innovaccer's fifth acquisition and expands its Flow suite into full-stack RCM capabilities that combine scheduling, patient engagement, billing, claims, and denial resolution. The press release states CaduceusHealth supports nearly 4,000 providers and manages roughly $5 billion in annual gross patient charges. Financial terms were not disclosed, Citybiz reports. "We started Innovaccer with the belief that the people who went into healthcare didn't sign up for administrative work," said Abhinav Shashank, CEO and Co-Founder, in the Business Wire release. Editorial analysis: Industry observers treating similar deals note they accelerate operational automation while retaining domain expertise.
Quick Read+7

See the full news feed