InnoNurse | Health tech news

Health technology news and insights. Got HealthTech? Absolutely!

[innonurse]

Report: US digital health funding surges to $7.4B as AI reshapes the sector

- By InnoNurse Staff -

Based on Rock Health's H1 2026 Digital Health Market Report, the first half of 2026 marked a strong recovery in digital health investment, driven largely by AI, while funding became increasingly concentrated among a small number of well-capitalized companies.

According to the report, U.S. digital health startups raised $7.4 billion across 244 deals, up from $6.4 billion in H1 2025. The median deal size increased to $14 million, the highest since 2022. Funding concentration also intensified, with 20 mega-rounds ($100M+) accounting for 45% of all invested capital, meaning just 8% of deals captured nearly half of the funding.

Rock Health notes that AI has become standard rather than a competitive differentiator. As software becomes easier to build, investors are placing greater value on companies with durable advantages such as deep healthcare expertise, founder-market fit, ownership of larger clinical workflows, high-touch customer implementation, and strong partnership ecosystems. Success increasingly depends on execution, trust, and measurable outcomes rather than AI capabilities alone.

The report highlights that mental health remained the highest-funded clinical indication for the seventh consecutive year, supported by continued investment in AI-assisted care platforms with strong clinical oversight. Weight management and obesity ranked second, fueled by the expanding GLP-1 ecosystem, consumer health platforms, and emerging peptide therapies. Both segments rely heavily on direct-to-consumer business models, reflecting sustained consumer demand for accessible digital healthcare.

Strategically, Rock Health observes that companies are evolving from single-purpose applications into broader healthcare operating platforms. AI is enabling greater workflow automation, encouraging platform expansion instead of point solutions, increasing demand for customized implementations through forward-deployed engineering teams, and strengthening the importance of partnerships with trusted healthcare organizations.

The report also notes that the public market remains relatively quiet despite strong private valuations. Oura is preparing for an IPO following its record $900 million funding round, while Whoop also raised significant capital ahead of a future listing. Public companies such as Hinge Health and Tempus posted strong growth driven by AI, although weaker companies continued to face delistings and bankruptcies.

Mergers and acquisitions remained active, with 115 digital health acquisitions completed during H1 2026. Consolidation was strongest in revenue cycle management, while major healthcare companies and private equity firms continued acquiring strategic assets to expand their digital health capabilities.

Overall, Rock Health concludes that the digital health sector has entered a more mature phase where AI is an expected capability rather than a differentiator. Investment is increasingly flowing toward companies that combine AI with healthcare expertise, scalable platforms, trusted partnerships, and demonstrable business outcomes, signaling a healthier and more resilient market than in the immediate post-pandemic period.

Read more at Rock Health

///

Other recent news and insights

Pearl Health banks $110M in fresh funding to build out tech and AI for Medicare providers

Feds push back HIPAA security rule overhaul to July 2027

Medical data of minorities could be more vulnerable to cyberattack

Why rural healthcare fund’s $50B focus on tech upgrades may not help vulnerable hospitals and providers

AI framework helps identify new targets for CAR T cell therapy

Investment in Israeli health tech startups plunged 40% in 2025 – report

Precision fMRI maps prefrontal cortex in individuals, revealing fine-scale structures

Handspring Health lands $19M series B to expand behavioral health model for youth

AI unlocks previously invisible cortical lesions in MS using legacy MRI scans

Commure launches AI-powered referral and patient intake platform

Surgeons use teleoperated humanoid robots to perform live surgery—a world first

Leo Cancer Care secures $65M for upright radiotherapy

Patient-specific 3D-printed contact lenses in just 20 minutes

Real-time biosensor measures pH, glucose and lactate in preserved donor livers

Evidenced closes $24M fund I to back health tech startups riding regulatory tailwinds

Doctors want wearable data but healthcare isn't ready for it, AMA survey finds

[innonurse]

Prosus invests €400 million in French health-tech company Alan

- By InnoNurse Staff -

Tech investor Prosus announced a €400 million (US$460 million) investment in Alan, valuing the French health-tech company at €5.5 billion (US$6.3 billion).

The funding round includes both primary and secondary equity and is backed by existing investors Teachers' Venture Growth and Index Ventures, as well as new investor Dara Holdings.

Alan, which combines health insurance, care, and prevention services through an AI-enabled platform, reported more than €800 million in annual recurring revenue as of Q1 2026, up 53% year over year. The company serves more than 1.1 million members, operates in France, Spain, Belgium, and Canada, and is profitable in France, its largest market.

According to the companies, the investment will support Alan's international expansion and further development of its AI-powered products. Prosus plans to support growth through its global network and technology capabilities, while Alan is expected to contribute to Prosus' broader healthcare and consumer services strategy.

The transaction remains subject to regulatory approvals, including those from French financial authorities.

Header image: Jean-Charles Samuelian-Werve (left) and Charles Gorintin (right), the co-founders of Alan. Credit: Alan.

Read more at Prosus

///

Other recent news and insights

AI-assisted, real-time deep-brain stimulation therapy for walking impairments in Parkinson's disease

AI screens 6 million compounds to uncover two leads against drug-resistant gonorrhea

Modular nanorobot self-assembles, targets cancer cells and cuts viability

Mini robot simplifies dental treatment by preparing teeth for crowns

AI tool reveals hidden organ damage caused by high blood pressure

New framework renders AI more trustworthy for cancer subtyping

Prosper AI raises US$30m from Andreessen Horowitz to scale AI platform

Ovum raises $4M seed

Cadence raises $100M Series C led by Spark Capital to automate chronic care

AI and polygenic scores improve breast cancer risk assessment

AI system detects sudden cardiac death risk, identifying thousands more patients annually

Portuguese tech links health data of 70 million people in Europe and Brazil

Assort Health scores $120M series C to scale voice AI agent platform for healthcare

Healthcare AI platform Xsolis suffers data breach impacting 1.4M individuals

[innonurse]

Reviewing SweetDream Through a Privacy-First Lens

I've spent the last few weeks testing AI companion apps the way I'd test any tool I plan to trust with personal data, and the differentiator that kept pulling SweetDream ahead wasn't the flashiest demo reel. It was the quiet confidence of its privacy-first design. On sweetdream.ai, the entire experience is built around the idea that your conversations and your generated content belong to you, full stop. That framing changes how the whole product feels.

What surprised me on closer inspection is that discretion doesn't come at the cost of capability. You still get the deeply customizable character creation, where you shape your AI girlfriend's looks, voice, backstory and little quirks, and you still get chat that genuinely remembers context instead of resetting every session. The privacy layer just sits underneath all of it, unobtrusive but constant.

Plenty of platforms talk about discretion as an afterthought. SweetDream treats it as the foundation, and that is exactly why, in my notes, it reads as the standout AI companion option rather than just another entry on the list.

[innonurse]

Hetairos AI enables rapid brain tumor classification from standard tissue sections

- By InnoNurse Staff -

Researchers at the German Cancer Research Center and Heidelberg University Hospital have developed Hetairos, an AI system that can classify brain and spinal cord tumors from standard stained tissue samples without requiring complex molecular testing.

Trained on more than 11,000 digitized tissue sections from 9,606 patients across 11 medical centers on four continents, Hetairos can identify 102 molecular subtypes of central nervous system tumors. In cases where the AI was highly confident (50–70% of samples), it achieved 87–88% accuracy and helped narrow down diagnostic possibilities even when uncertainty remained.

In head-to-head testing against five experienced neuropathologists, Hetairos achieved 68% diagnostic accuracy compared with 30% for human specialists when relying only on tissue images. When considering the top three likely diagnoses, the AI reached 84% accuracy, versus about 50% for the experts.

A prospective clinical study found that while traditional molecular diagnostics took an average of 12 days, Hetairos produced results in about 12 minutes after slide digitization, with complete workflows potentially delivering findings within 24–48 hours.

The developers emphasize that Hetairos is intended to complement rather than replace molecular testing, helping pathologists prioritize further analyses, interpret difficult cases, and improve access to advanced diagnostics in regions lacking specialized laboratory infrastructure. The technology could also reduce costs by using routinely prepared tissue sections instead of expensive DNA methylation testing. The study was published in the journal Nature Cancer and highlights the growing potential of AI-powered digital pathology to make sophisticated cancer diagnostics faster, more accessible, and more affordable worldwide.

Read more at The German Cancer Research Center (DKFZ)

///

Other recent news and insights

Augmented reality system could make medical ultrasounds easier to interpret

Mapping brain network changes linked to bipolar disorder severity and treatment

[innonurse]

ResMed closes $340 million acquisition of Noctrix Health

- By InnoNurse Staff -

ResMed has completed its $340 million acquisition of Noctrix Health, adding a new wearable therapy for restless legs syndrome (RLS) to its sleep health portfolio.

The key asset is Nidra, an FDA-cleared wearable device that uses noninvasive nerve stimulation in the legs to reduce RLS symptoms and improve sleep quality. It is intended for adults with moderate-to-severe RLS who have not found sufficient relief from medication.

The acquisition expands ResMed beyond its traditional focus on sleep apnea, CPAP machines, masks, ventilators, and respiratory care. The company plans to integrate Nidra into its connected-device and digital health ecosystem as part of its broader strategy to become a more comprehensive sleep health company.

The deal aligns with ResMed's long-term push into home-based, connected healthcare and wearable therapeutics, following earlier acquisitions such as Propeller Health. It also gives ResMed access to a large sleep-disorder market, as RLS is considered the third most common sleep disorder after insomnia and obstructive sleep apnea.

The main challenge ahead will be scaling Nidra's adoption by leveraging ResMed's established relationships with sleep physicians, home-care providers, and digital health infrastructure.

Read more at Fierce Biotech

///

Other recent news and insights

AI-powered medical imaging startup Subtle Medical picks up $33M and taps new CEO

Wearable device can continuously monitor blood pressure without the pesky cuffs

New 3D map of the heart's electrical wiring can help patients with congenital heart disease

Nanoparticles boost delivery of lung cancer drugs 30-fold

Montréal research hospital opens new healthtech innovation hub

A look at wearable adoption trends and who's using 'smart' devices: Rock Health

New hand sensors turn post-stroke rehab into an on-screen game

AI helps accelerate breast cancer diagnosis for high-risk women

Smartphone unlock can measure heart rate, potentially bringing health monitoring to billions worldwide

Eight startups selected to advance heart and brain health technologies

AI can reveal brain tumor risks without costly genetic testing

Elation Health acquires Aster as it ramps up agentic AI for primary care

Fullspan Health debuts Healthline AI companion to connect consumers to medical content and providers

[innonurse]

Enzo Health launches AI-native EHR for home healthcare agencies

- By InnoNurse Staff -

Enzo Health has launched Enzo EHR, what it claims is the first AI-native electronic health record (EHR) system built specifically for home healthcare agencies.

Unlike traditional EHRs that require multiple add-on tools, Enzo's platform uses AI agents to automate the entire patient care workflow—from referrals and scheduling to documentation, compliance, and billing.

Key claimed benefits include:

Intake processing: Reduced from about 70 minutes to 5 minutes by automatically reviewing referral packages and extracting relevant information.

Scheduling: Reduced from 15 minutes to 30 seconds by automatically matching clinicians with patients.

Clinical documentation: Reduced by roughly 75% through AI-generated visit notes based on conversations during patient visits.

Billing: AI prepares claims and identifies potential issues that could lead to payment denials.

CEO Zach Newman said home healthcare providers often rely on several disconnected software products layered on top of outdated EHR systems, creating inefficiencies and information silos. Enzo's goal is to consolidate those functions into a single platform.

An early adopter reportedly improved intake decision-making speed by 10 times, cutting admission decisions from 15–20 minutes to 1–2 minutes while also reducing clinician documentation time by about 75%.

Founded in 2024, Enzo Health recently raised $20 million in Series A funding, bringing total funding to $26 million. The company says its revenue has grown 40-fold in the past year and that its technology supports more than 500,000 patients annually across over 100 organizations.

The broader vision is to use AI to handle administrative tasks so healthcare professionals can spend more time focused on patient care.

Read more at Fierce Healthcare

[innonurse]

Teladoc expands virtual care partnership with Walmart

- By InnoNurse Staff -

Teladoc Health announced a partnership with Walmart to offer virtual healthcare services through Walmart’s Better Care Services platform.

Under the agreement, Walmart customers can access:

Virtual urgent care

Dermatology consultations

Nutrition services

The services are available through insurance coverage or direct payment. Virtual urgent care visits cost $89 for cash-pay users and cover common conditions such as sinus infections, UTIs, flu symptoms, and pink eye. Dermatology patients can upload photos for review by a board-certified dermatologist, who provides a diagnosis and treatment plan within 24 hours. Nutrition services connect users with registered dietitians for personalized guidance, including support for diabetes, weight management, and prenatal nutrition.

Prescriptions can be sent to pharmacies, including Walmart pharmacies, which may offer same-day or one-hour delivery in some locations.

The expansion follows the addition of Teladoc’s BetterHelp mental health services to Walmart’s platform earlier this year.

Teladoc said the partnership is part of its broader effort to expand access points for virtual healthcare. The company also has partnerships with platforms including Amazon and Instacart.

Teladoc says its services are available to more than 100 million people in the U.S. through employers and health plans.

Read more at Fierce Healthcare

///

Other recent news and insights

Finnish smart ring maker Oura plans IPO at over €9 billion as wearable market heats up

AI-guided drug search flags folic acid for diabetic wound healing

Wearable ultrasound patch for high-risk pregnancies could improve care

AI uncovers why squeezed tumors grow slower under physical pressure

Whole organ 3D imaging reveals remaining insulin producing cells in type 1 diabetes

Oura unveils AI-enabled Ring 5 with predictive health features

AI and simulations cut advanced brain MRI time by up to 90%

Seven AI models vote out medical hallucinations in 10,000 chatbot tests

[innonurse]

AI model reads complex heart MRIs without manual labels

- By InnoNurse Staff -

Researchers from Carnegie Mellon University and Cleveland Clinic have developed CMR-CLIP, an AI system that can interpret cardiac MRI scans without relying on manually labeled training data.

Instead of using expert annotations, the model learns by linking heart MRI videos with the radiology reports that physicians already write during routine clinical care. Trained on more than 13,000 patient studies and over a million images, CMR-CLIP analyzes the heart as a moving organ rather than as a collection of static images.

In testing, the system outperformed general-purpose AI models by as much as 35%, identified heart conditions it had never been specifically trained on, and achieved accuracy rates of up to 99% for some diagnostic tasks. It also showed promise for finding similar patient cases through natural-language searches and maintained strong performance when tested on datasets from other hospitals.

The researchers say the technology could help reduce the workload of cardiac imaging specialists, improve access to advanced heart diagnostics, support clinical decision-making, and serve as a training tool for physicians. The study was published in Nature Communications, and the team plans to expand the model to additional cardiac imaging applications, including automated report generation and decision-support systems.

Read more at Carnegie Mellon University

[innonurse]

Lauxera Capital Partners closes €520M Healthtech growth fund II

- By InnoNurse Staff -

Lauxera Capital Partners has closed its second Healthtech-focused investment fund at €520 million, exceeding its €500 million target.

The fund will invest in 12 to 15 European Healthtech companies, with individual investments ranging from €20 million to €50 million, primarily supporting international expansion into the U.S. market.

The firm focuses on sectors including medical devices, digital health, healthcare software, diagnostics, and medtech services. According to Lauxera, the fund is intended to address a financing gap for European Healthtech companies seeking growth-stage capital and operational support.

Lauxera has already invested in Acandis and Antaros Medical through the new fund. The firm also noted the 2025 acquisition of OrganOx by Terumo for approximately $1.5 billion.

Founded in 2020, Lauxera operates from Paris and San Francisco and manages more than $1 billion in assets. The company says its investment strategy also includes measuring healthcare-related impact metrics tied to patient outcomes, provider productivity, and healthcare system costs.

Read more at Lauxera Capital Partners/Business Wire

[innonurse]

AI model predicts Alzheimer’s-related cognitive decline from a single MRI scan

- By InnoNurse Staff -

Researchers at University of California, San Francisco developed a new AI model that can predict cognitive decline and Alzheimer’s-related outcomes using only a single baseline MRI scan and basic demographic information.

Published in Nature Aging, the multitask deep learning system outperformed existing AI methods by accurately predicting Alzheimer’s diagnosis, tissue changes, and both current and future cognitive scores without requiring expensive PET scans, genetic testing, fluid biomarkers, or lengthy neuropsychological assessments.

The model’s key innovation is a brain image segmentation approach that classifies tissue into gray matter, white matter, and cerebrospinal fluid, helping the AI better understand structural brain patterns linked to cognitive decline. Researchers trained and validated the system using datasets from the Alzheimer's Disease Neuroimaging Initiative, the Human Connectome Project, and the Dallas Lifespan Brain study.

According to the team, the approach could provide faster and more accessible screening for cognitive impairment in clinical and community settings, especially where specialized neurocognitive testing is limited. The researchers also believe the framework could eventually help predict progression in other neurodegenerative diseases such as Parkinson's disease, Amyotrophic lateral sclerosis, and Huntington's disease.

Image: Overall model architecture. Credit: Nature Aging (2026). DOI: 10.1038/s43587-026-01121-2

Read more at University of California, San Francisco/Medical Xpress

///

Other recent news and insights

Vital Signals raises $15M to transform long-term blood pressure management

[innonurse]

AI trial shows faster radiotherapy planning could expand global cancer treatment access

- By InnoNurse Staff -

An international clinical trial led by researchers from University College London and London School of Hygiene & Tropical Medicine found that AI can effectively plan radiotherapy treatments for cervical and prostate cancer, potentially expanding access to life-saving care worldwide.

The ARCHERY trial involved more than 1,000 patients across hospitals in India, South Africa, Jordan, and Malaysia. Researchers tested AI software that automates the complex process of radiotherapy planning, a task that normally requires many hours of work from specialized oncologists and physicists.

Results presented at the European Society for Radiotherapy and Oncology Congress 2026 showed the AI produced high-quality treatment plans in more than 95% of cervical cancer cases and 85% of prostate cancer cases. The technology can reduce planning time from weeks to just over an hour.

Researchers say the system could help address major shortages of radiotherapy specialists, especially in low- and middle-income countries where access to treatment remains limited. Cervical cancer is particularly affected, with 94% of deaths occurring in lower-resource nations.

Experts say the study is one of the largest and most rigorous international trials evaluating AI in cancer treatment, demonstrating that AI could improve efficiency, reduce waiting times, and help more patients receive potentially curative radiotherapy worldwide.

Read more at University College London

[innonurse]

Report: US hospital EHR purchases decline in 2025 as Epic gains market share and Oracle Health loses ground

- By InnoNurse Staff -

Acute care hospitals sharply reduced electronic health record (EHR) purchases in 2025, with buying activity down 40% from 2024 and nearly 50% from 2023, according to KLAS Research.

Hospitals delayed decisions due to uncertainty around government policy and shifted spending toward AI and operational efficiency technologies.

Despite the slowdown, Epic Systems strengthened its dominance in the U.S. acute care EHR market, increasing its share to 43.7% and winning most major contracts, especially among smaller and midsize health systems. Epic gained 77 hospitals and nearly 18,700 beds in 2025, with many customers switching from Oracle Health. Hospitals cited Epic’s strong platform, interoperability, and partnerships as key reasons for choosing the vendor. Its Community Connect program also helped attract smaller standalone hospitals.

Oracle Health, which acquired Cerner in 2022, suffered its third straight year of major losses, dropping to 21.9% market share after losing 56 hospitals and more than 14,600 beds. KLAS reported declining customer satisfaction, ongoing restructurings, layoffs, and uncertainty around Oracle’s long-term strategy. Many customers are waiting to see whether Oracle’s new AI-powered, voice-enabled EHR platform can restore confidence in 2026.

Meditech maintained stronger customer retention, particularly among legacy clients upgrading to its Expanse platform, though it still lost seven hospitals. Smaller vendors like TruBridge and Altera Digital Health saw limited gains, mainly among budget-conscious or existing customers.

Read more at Fierce Healthcare

[innonurse]

CRB Health Tech closes €50 million digital health fund focused on European healthcare transformation 

- By InnoNurse Staff -

Madrid, Spain-based CRB Health Tech has closed its third digital health investment fund, CRB Digital Health Tech III, at €50 million, exceeding its original €30 million target.

The fund attracted a mix of industrial, financial, and public-sector investors, including healthcare companies, financial institutions, and Spanish public entities.

The firm says the oversubscription reflects growing interest in digital health technologies across Europe, particularly in areas such as AI-driven healthcare, hospital digitalization, diagnostics, remote monitoring, and mental health solutions. CRB Health Tech positions the fund as part of a broader effort to support the technological transformation of healthcare systems in Europe.

The fund focuses on companies that have already validated their technology and are entering commercialization stages, aiming to support growth and international expansion. It has already approved ten investments across several European countries, including Spain, Germany, France, Italy, the Netherlands, and the United Kingdom.

Portfolio companies include startups working in healthcare workforce management, AI-powered infant cry analysis, and digital psycho-oncology. CRB Health Tech says it plans to build a portfolio of around 25 companies, with investments ranging from €500,000 to €2 million, while reserving additional funding for follow-on rounds.

The firm also highlighted women’s health, mental health, AI diagnostics, and healthcare efficiency tools as key areas of interest for future investments. According to CRB executives, demographic trends, pressure on healthcare systems, and advances in AI are contributing to the sector’s long-term growth potential.

Read more at Capital-Riesgo.es

///

Other recent news and insights

AI pathology framework can enable a deeper understanding of cancer

Home health tech company Enzo Health raises $20M

[innonurse]

PS-OCT and AI could improve early detection of keratoconus

- By InnoNurse Staff -

Keratoconus is a progressive eye disease that weakens the cornea, but in its earliest stage it can look normal on standard exams—making early diagnosis difficult, especially before refractive surgery.

A recent study shows that combining polarization-sensitive OCT (PS-OCT) with AI can detect subtle, early changes by analyzing corneal collagen structure rather than just shape. Traditional tools (like Pentacam or MS-39) rely on curvature and thickness, which often miss these microscopic changes.

Using data from 359 eyes, researchers compared AI models based on different imaging methods. All performed similarly for clear cases, but PS-OCT was better at distinguishing true early keratoconus from normal corneas. It identified structural differences—like altered collagen organization and slight thinning in specific layers—even when the cornea’s shape looked normal.

This matters because current methods often label borderline cases as “suspect,” potentially limiting treatment options. PS-OCT could improve diagnostic confidence by separating genuinely at-risk eyes from naturally thin but healthy ones.

Overall, the study suggests that adding structural, collagen-level insight to existing imaging could enable earlier and more accurate detection, leading to better-informed clinical decisions.

Image: Tomography is central for measuring global corneal geometry, but its limitations create a diagnostic gap, especially for early detection. PS-OCT may address this issue by combining ultrahigh-resolution tomography with polarization-sensitive functional imaging that can help distinguish true early disease from naturally thin but healthy corneas by focusing on collagen organization rather than thickness alone. Left column: eyes classified as healthy; middle, subclinical; right, keratoconus. Credit: R. P. Patil et al.

Read more at SPIE

///

Other recent news and insights

Oura adds birth control support to its period tracker

Battery-free skin-conformal wearable system can measure electrocardiogram signals

[innonurse]

IKS Health to acquire TruBridge for $557 million

- By InnoNurse Staff -

Inventurus Knowledge Solutions (IKS) Health—a healthcare technology and services company that provides software and outsourced operational support to hospitals and physician groups—has agreed to acquire TruBridge for approximately $557 million.

The deal is expected to be financed through a roughly $600 million loan over five years, backed by Citibank, JPMorgan Chase, and Deutsche Bank.

The acquisition will combine IKS Health’s care enablement and administrative support capabilities with TruBridge’s revenue cycle management and electronic health records (EHR) solutions, with a focus on rural and community hospitals. The combined organization is projected to support more than 2,000 healthcare providers and over 150,000 clinicians across the United States.

According to company statements, the transaction is intended to strengthen operational efficiency and healthcare delivery in underserved areas while expanding service offerings and organizational scale.

Read more at Fierce Healthcare

///

Other recent news and insights

AI squeezes individual breast cells to learn how to spot cancer risk

Conversational AI uses trusted medical protocols to help people decide when to seek care

Tava Health raises $40M for tech-enabled mental health platform

AI model detects normally 'invisible' tissue changes of pancreatic cancer at stage 0

[innonurse]

AI models help uncover the brain mechanisms behind dyslexia

- By InnoNurse Staff -

Researchers at EPFL’s NeuroAI Lab used advanced Vision-Language Models to simulate how the brain processes reading and to better understand dyslexia, a condition affecting up to 20% of people.

By identifying and disabling parts of the AI model that mimic the brain’s “visual word form area,” they reproduced reading difficulties similar to dyslexia—while other abilities like understanding images and language remained intact. This suggests the model captures key mechanisms behind the disorder.

The team also tested different fonts and found the AI performed better with fonts designed for dyslexic readers, helping guide the creation of improved, optimized fonts.

Beyond dyslexia, the research introduces a new AI-based framework for studying brain disorders more broadly, offering a way to experimentally test hypotheses that can’t be explored directly in humans.

Read more at Ecole Polytechnique Federale de Lausanne (EPFL)

///

Other recent news and insights

Doctolib, AP-HP, and Roche launch startup programme to bridge healthtech–clinical gap

Oak HC/FT backs Courier Health's $50M series B to build out AI for the biopharma patient experience

UK HealthTech startup Calibre emerges from stealth with €2.8 million to tackle “health guesswork”

[innonurse]

PerturbFate uses single-cell data to decode complex disease pathways

- By InnoNurse Staff -

Researchers have developed PerturbFate, a computationally driven platform designed to make sense of the massive complexity behind diseases like cancer and neurodegeneration, where hundreds of genetic mutations can lead to the same outcome.

At its core, PerturbFate combines large-scale genetic perturbation experiments with a powerful data analysis pipeline that integrates multiple layers of single-cell data. By tracking RNA dynamics, gene expression, and DNA accessibility in parallel, the system reconstructs gene regulatory networks over time, linking diverse genetic changes to their downstream effects.

This computational approach allows scientists to identify shared “regulatory nodes” — key control points where different mutation pathways converge. In a melanoma case study, the platform analyzed data from over 300,000 cells and showed that distinct mutations all led to drug resistance through common regulatory programs.

By targeting these shared nodes instead of individual mutations, researchers were able to reduce resistance, highlighting how computational modeling of cell behavior can guide more effective combination therapies.

Overall, PerturbFate demonstrates how integrating high-dimensional biological data with advanced computational analysis can uncover unifying mechanisms in complex diseases, shifting the focus from individual genes to shared system-level controls.

Header image: Cell state transition trajectories from genetically perturbed cells. Credit: Laboratory of Single-Cell Genomics and Population Dynamics at The Rockefeller University

Read more at Rockefeller University

///

Other recent news and insights

Singapore healthtech startup injewelme raises $1.2m to advance AI monitoring technology

AI scans 400,000 Reddit posts to flag overlooked GLP-1 side effects

AI outperforms doctors at summarizing complex cancer pathology reports

Mega rounds power Q1 2026 health tech fundraising

Novo Nordisk signs deal with OpenAI to develop new drugs

Helical closes $10M seed to turn bio foundation models into systems

AI can now understand health signals from 15 seconds of your voice, including type 2 diabetes, fatigue and stress

Your phone already sees the warning signs: Sleep, movement and mood data can spot depression early

[innonurse]

Op-ed: Hospitals fueling AI innovation raise questions about fair value sharing

- By InnoNurse Staff -

In their Industry Voices op-ed, “Hospitals are fueling AI innovation—should they own a piece of it?”, Carolyn Metnick and Timothy Rozier-Byrd argue that a fair partnership between hospitals and AI developers should allow health systems to share in the financial and strategic value created from their data and expertise, while preserving patient trust, regulatory compliance, and appropriate risk protections.

The authors note that hospitals are rapidly adopting and building AI tools—independently and with technology partners—to reduce burnout, improve care, and cut costs. In doing so, they generate vast amounts of clinical data and operational insight that increasingly drive the performance and market value of AI systems. This builds on longstanding data-sharing practices, including health information exchanges, partnerships with major tech firms, and the monetization of de-identified data, but is now evolving toward continuous, real-time data ecosystems and enterprise-wide AI integration.

As a result, hospitals face growing legal and business complexities, including tax considerations in joint ventures, privacy compliance, reputational risk, and questions around data ownership—especially for derivative datasets produced by AI models. The authors stress the need for clear agreements on data rights, rigorous de-identification standards, and strong governance frameworks to ensure clinical validity, reduce bias, and maintain human oversight.

Ultimately, they argue that equitable AI partnerships depend on transparent value-sharing, well-defined contractual protections, and strategic discipline—enabling hospitals to capture both economic and clinical benefits while minimizing legal and ethical risks.

Read more at Fierce Healthcare

///

Other recent news and insights

3D-printed ATLAS platform helps model cancer cell clusters behind metastasis

'Smart' bandage tracks wound status in real-time

Organ-on-a-chip technology replicates decades of human aging in just four days

Mental health provider platform Headway acquires team behind AI company Tezi

Digital twin hearts deliver 100% success in arrhythmia trial

New AI tool predicts whether aggressive small cell lung cancer will respond to treatment