Heidi’s Hannah Allen joins the Healthcare Today Podcast to explain what it takes for AI tools to earn clinicians’ trust and give them back time with patients.

Healthcare AI regulation is missing a critical component: physician accountability. STAT News commentators argue that doctor… #Artificialintelligence #Clinicalsafety #Healthcareregulation #Medicallicensure #Physicianaccountability

Mental health professionals warn: AI chatbots like ChatGPT are NOT therapy. While they may support information-seeking, they cannot diagnose… #Artificialintelligence #Clinicalsafety #DigitalHealth #MentalHealth #Psychotherapy

Documentation site for all parts of the RCPCH Digital Growth Charts Project - rcpch/digital-growth-charts-documentation

YouTube video by Faculty of Clinical Informatics

  Even though almost every aspect of modern medicine is supported by digital infrastructure, the healthcare sector finds itself at the epicentre of an escalating cybersecurity crisis at the same time. Cyberattacks have now evolved from being just a financial or corporate problem to a serious clinical concern, causing patients' safety to be directly put at risk as well as disrupting essential healthcare.  With the increasing use of interconnected systems in hospitals and diagnostic equipment, as well as cloud-based patient records, the attack surface on medical institutions is expanding, making them increasingly susceptible to ransomware and data breaches posed by the increasing use of interconnected systems.  The frequency and sophistication of such attacks have skyrocketed in recent years, and the number of attacks has almost doubled compared to 2023, when the number of ransomware attacks in the United States alone climbed by a staggering 128 per cent in the same year. As far as data loss and financial damage are concerned, the consequences of these breaches do not stop there.  There are estimates of healthcare organisations losing up to $900,000 per day because of operational outages linked to ransomware, which excludes the millions—or billions—that are spent on ransom payments. In IBM's 2024 Cost of a Data Breach Report, healthcare was ranked as the highest cost per incident in the world, with an average cost of $9.8 million. This was significantly more than the $6.1 million average cost per incident within the financial sector.  In spite of this fact, the most devastating toll of cyberattacks is not in currency, but rather in the lives of victims. Studies indicate that cyberattacks have resulted in delayed procedures, compromised care delivery, and, in some cases, increased mortality rates of patients. There has been a troubling increase from the previous year, since 71 per cent of healthcare organisations affected by cyber incidents reported negative patient outcomes due to service disruptions in 2023.  With the rapid growth of digital transformation in healthcare, the line between data security and clinical safety is fast disappearing - making cybersecurity an urgent issue of patient survival rather than mere IT resilience as digital transformation continues to redefine healthcare. With cyber threats growing more sophisticated, healthcare is experiencing a troubling convergence of digital vulnerability and human consequences that is becoming more and more troubling.  There was once a time in healthcare when cybersecurity was viewed solely as a matter of data protection; however, today, it has become an integral part of patient safety and wellbeing, which is why experts are predicting that the threat of cybersecurity attacks will escalate significantly by the year 2025, with hospitals and health systems facing increasing financial losses as well as the threat of escalating risks.  Recent reports have highlighted hospitals being incapacitated by ransomware attacks, which have compromised critical care, eroded public trust, and left healthcare staff unable to provide care. "Patient safety is inseparable from cyber safety," emphasised Ryan Witt, Proofpoint's healthcare leader, emphasising that when digital systems fail, life-saving care can be compromised. Statistics behind these incidents reveal a frightening reality.  A study found that nearly seventy-eight per cent of healthcare organisations experienced disruptions in patient care as a result of ransomware, email compromise, cloud infiltration, and supply chain attacks. More than half of these patients experienced extended stays in the hospital or medical complications, while almost a third saw a rise in death rates.  Financial figures often overshadow the human toll of a major attack: although the average cost has fallen to $3.9 million from $4.7 million, ransom payments have risen to $1.2 million from $4.7 million. It is important to remember that there are no monetary figures that can fully capture the true impacts of systems that go dark-missing diagnoses, delays in surgery, and the lives put at risk of clinicians, nurses, and technicians.  Considering that time and precision are synonymous with survival in the healthcare sector, it has become clear that the encroachment of cybercrime is more than merely a technology nuisance and has become a profound threat to the very concept of care itself. Health Information Sharing and Analysis Centre (Health-ISAC) continues to play an important role in strengthening the industry's defences amidst increasing global cyber threats targeting the healthcare sector.  It serves as an important nexus for collaboration, intelligence sharing, and real-time threat mitigation across healthcare networks worldwide. Health-ISAC is a non-profit organisation run by its members. A vital resource for safeguarding both digital and physical health infrastructures, Health-ISAC has disseminated actionable intelligence and strengthened organizational resilience through the distribution of actionable intelligence and strengthening of organisational resilience.  It has recently been reported that the organisation has identified several security threats, including critical vulnerabilities found within Citrix NetScaler ADC, NetScaler Gateway, and Cisco Adaptive Security Appliances (ASA) that could potentially be exploited. Immediately after the identification of these flaws, Health-ISAC issued over a hundred targeted alerts to member institutions in order to minimise the risk of exploitation.  These vulnerabilities have been exploited by threat actors since then, highlighting how the healthcare sector needs to be monitored continuously and provide rapid response mechanisms. As well as detecting threats, Health-ISAC has also been involved in regulatory alignment, particularly addressing FDA guidance regarding cybersecurity for medical devices that was recently updated.  Revisions to the quality system considerations and the content of premarket submissions, issued in June 2025, have replaced the earlier version, which was issued in 2023, and incorporate Section VII of the Federal Food, Drug, and Cosmetic Act (FD&C Act). In this section, manufacturers are outlined in detail about their specific compliance obligations, including the use of cybersecurity assurance procedures, Software Bills of Materials (SBOMs), and secure development methods.  It has also been emphasised by Health-ISAC that there are related regulatory frameworks that will affect AI-enabled medical devices, such as the FDA Quality Management System Regulation, the EU Cyber Resilience Act, and emerging standards such as AI-enabled data providers. In the organisation's latest analysis, the organisation explored how the geopolitical climate has been shifting in the Asia Pacific region, where growing tensions between the Philippines and China, particularly over the Scarborough Shoal, which has now been designated by China as a maritime wildlife refuge, are reshaping regional security.  The significant investment Australia has made in asymmetric warfare capabilities is a further indication of the interconnectedness between geopolitics and cybersecurity threats. Denise Anderson, President and CEO of Health-ISAC, commented on the organisation's 15-year milestone and stated that the accomplishments of the organisation demonstrate the importance of collective defence and shared responsibility. She added, "Our growth and success are a testament to the power of collaboration and to our members' passion to improve the welfare of patients," she expressed. "With the emergence of sophisticated threats, a unified defence has never been more needed." In the near future, Health-ISAC plans to strengthen the intelligence sharing capabilities of the organisation, expand its partnerships throughout the world, and continue promoting cybersecurity awareness - all of which will strengthen the organisation's commitment to making healthcare safer and more resilient throughout the world.  The healthcare landscape is becoming increasingly digitalised, and preserving it will require not only a proactive defence but a coordinated, unified approach as well. As technology and patient care have converged, cybersecurity has become a clinical imperative, one that will require the collaboration of policymakers, hospital administrators, medical device manufacturers, and cybersecurity specialists.  Various experts highlight that through investment in secure infrastructure, workforce training, and continuous monitoring and assessment of risks, there is no longer an option but instead a necessity to maintain the trust of patients and ensure the continuity of operations.  There is a significant reduction in vulnerabilities across complex healthcare ecosystems when zero-trust frameworks are implemented, timely software patches are made, and transparent data governance takes place. Moreover, fostering global intelligence-sharing alliances, such as the one promoted by Health-ISAC, can strengthen our collective resilience to emerging cyber threats. With the sector facing a number of emerging challenges in the future - from ransomware to artificial intelligence-enabled attacks - it is imperative that cyber safety is treated as an integral part of patient safety in order to survive. In addition to protecting data, healthcare delivery is also preserving its most vital mission: saving lives in a world where the next medical emergency could be just as easily caused by malicious code as it would be caused by the hospital.

Integrating large language models (LLMs) into healthcare can enhance workflow efficiency and patient care by automating tasks such as summarising consultations. However, the fidelity between LLM…

Japan Clinical Alarm Management Market growth at a 13.47% CAGR By top leading companies, global trends, advancements, Industry size, share and regions, analysis, forecast to 2035

Background: Diagnostic errors and administrative burdens, including medical coding, remain major challenges in healthcare. Large language models (LLMs) have the potential to alleviate these problems, but their adoption has been limited by concerns regarding reliability, transparency, and clinical safety. Objective: This study introduces and evaluates two LLM-based frameworks, implemented within the Rhazes Clinician platform, designed to address these challenges: Generation-Assisted Retrieval-Augmented Generation (GARAG) for automated evidence-based treatment planning and Generation-Assisted Vector Search (GAVS) for automated medical coding. Methods: GARAG was evaluated on 21 clinical test cases created by medically qualified authors. Each case was executed three times independently, and outputs were assessed using four criteria: correctness of references, absence of duplication, adherence to formatting, and clinical appropriateness of the generated management plan. GAVS was evaluated on 958 randomly selected admissions from the MIMIC-IV database, in which billed ICD-10 codes served as ground truth. Two approaches were compared: a direct GPT-4.1 baseline prompted to predict ICD-10 codes without constraints, and GAVS, in which GPT-4.1 generated diagnostic entities that were each mapped onto the top 10 matching ICD-10 codes through vector search. Results: Across the 63 outputs, 62 (98.4%) satisfied all evaluation criteria, with the only exception being a minor ordering inconsistency in one repetition of case 14. For GAVS, the 958 admissions contained 8,576 assigned ICD-10 subcategory codes (1,610 unique). The vanilla LLM produced 131,329 candidate codes, whereas GAVS produced 136,920. At the subcategory level, the vanilla LLM achieved 17.95% average recall (15.86% weighted), while GAVS achieved 20.63% (18.62% weighted), a statistically significant improvement (p < .001). At the category level, performance converged (32.60% vs 32.58% average weighted recall; p = 0.986). Conclusions: GARAG demonstrated a workflow that grounds management plans in diagnosis-specific, peer-reviewed guideline evidence, preserving fine-grained clinical detail during retrieval. GAVS significantly improved fine-grained diagnostic coding recall compared with a direct LLM baseline. Together, these frameworks illustrate how LLM-based methods can enhance clinical decision support and medical coding. Both were subsequently integrated into Rhazes Clinician, a clinician-facing web application that orchestrates LLM agents to call specialized tools, providing a single interface for physician use. Further independent validation and large-scale studies are required to confirm generalizability and assess their impact on patient outcomes.