Companies in the early stages of AI are developing applications to assist clinicians and staff in a variety of work activities, from automated administrative tasks that reduce documentation time and free up clinician time, to predictive analytics and real-time computer vision that improve operating room efficiency. To reap the full value of these exciting developments, you need endpoint hardware that can provide a reliable and secure workspace.
Medical staff and clinicians move between different locations within the health system and between floors within a hospital. They have already experienced the frustration of not being able to access data in critical moments due to system downtime or cross-desktop compatibility issues. Consistent performance is essential in any care setting.
Whether it’s an AI-driven application or a legacy application, the endpoint is the moment of truth. As the use of AI increases, having hardware and software that supports high-performance experiences for all staff will improve operational efficiency, patient outcomes, and revenue.
Key issues to consider are hardware selection, AI readiness, security and compliance, operational interoperability, and total cost of ownership (TCO).
Hardware options for AI environments
PCs and thin clients are two common choices for endpoint device combinations used by health systems. As systems continue to face tight budgets, thin clients are an economical option that requires less IT staff time and money than PCs. Hospitals operating in a variety of computing environments can use thin clients to support VDI, cloud, hybrid, and on-premises environments, increasing the ROI on their investments.
Modern thin clients are housed in small form factors that are advantageous for busy hospital and clinic environments, and are fanless and solid-state, improving reliability, durability, and energy efficiency. It requires less maintenance and can be managed centrally, reducing administration time.
With the advent of AI, AI PCs join this mix. These require neural processing power and high-end CPU/GPU computing power, which can further drive up energy costs. As a result, prices are high and can be prohibitive for budget-conscious health systems. Traditional PCs tend to be moderately priced, but they typically need to be updated every three to five years, frequently draining your budget.
Thin clients are read-only devices and do not store any local data. When evaluating thin client options, consider the needs of your staff and clinicians to determine the best choice in terms of processing power, screen quality, number of ports, and other factors. Interoperability with major VDI vendors such as Citrix, Omnissa, and Microsoft is also a must.
Long refresh cycles are an advantage of thin clients and help lower costs and protect TCO. However, the content of standard contracts varies by thin client vendor. Features that have additional costs include healthcare-specific subscriptions based on enterprise models, upgrade fees, and support subscriptions throughout the endpoint lifecycle.
These large additional costs can reach a point where replacement seems like a better option, requiring new purchases and undermining sustainability goals by increasing waste.
As AI plays a deeper role in how clinicians and staff serve patients, health systems are considering the best hardware devices to support AI workflows. Large amounts of AI data for large language models are primarily processed in the cloud or on servers. AI applications that run locally are still limited by the cost of AI PCs. Thin clients built to support VDI and the cloud can provide a low-power, low-cost gateway to access AI workload resources. Traditional PCs still offload AI workloads to the cloud, which comes at a higher cost.
Endpoint protection
Health system executives now face the triple threat of criminals using AI to perpetrate breaches, infiltration of PHI data into the AI’s large-scale language models, and significant financial losses from successful cyberattacks. Add HIPAA violations to the mix and the losses are even worse.
Cybercriminals love healthcare, and attacks are steadily increasing. From 2019 to 2024, healthcare data breaches increased from 397 to 536 per year, surpassed only by the financial sector. According to DemandSage, healthcare breaches remain the longest to identify and contain, taking an average of 279 days. The HIPAA Journal points out that this is a major reason why the field is so popular and the value of medical data is so high. Stolen data lasts longer than a stolen credit card, which can be quickly canceled.
To combat emerging threats, we need to place greater emphasis on guard rails that limit the data that can be input into large-scale language models. On endpoints, locked-down, read-only thin clients do not store data locally, which helps reduce the attack surface. Another option, a zero client, has no local OS and connects to a remote desktop. These devices reduce risk for busy clinicians and staff as they travel between locations and access medical records and other PHI data remotely.
To increase data security, healthcare IT departments will also need to integrate thin clients (or PCs) with software that provides another layer of protection. Interoperability with Imprivata Enterprise Access Management with single sign-on enables advanced identity authentication and supports HIPAA-compliant reporting. For clinicians, it provides passwordless authentication, allowing them to authenticate once and instantly re-access the application as they move between workstations.
A secure AI future
In the coming years, AI-powered computing will become the foundation of healthcare systems, often starting at the endpoint. There is no doubt that AI PCs, like traditional PCs, will eventually become commoditized. However, compute and power costs continue and update cycles become more frequent. The lower TCO and attack surface alternatives will continue to be thin and zero clients. Healthcare executives envisioning an AI future will look to the combination of thin clients and AI PCs to deliver superior patient care and sustainable business outcomes.
Photo: Weiquan Lin, Getty Images
Kevin Greenway joined 10ZiG in 2012 and became CTO in 2015. He leads technology and product strategy across the company and works with global teams to ensure continued innovation in a fast-paced and disruptive market. Under his leadership, 10ZiG delivers modern, managed and secure endpoints through an integrated hardware and software approach.
A computer science graduate with numerous IT certifications, Kevin has over 25 years of experience in IT including remote connectivity, terminal emulation, VoIP, unified communications, and VDI remote protocols. Since joining 10ZiG, he has focused on VDI and End User Computing (EUC), overseeing strategic technology alliances with key partners such as Citrix, Microsoft, and Omnissa. Outside of work, Kevin is a devoted family man who enjoys spending time with his wife, two children, and his dog. He enjoys running, cycling and watching sports such as motorsport and football/soccer, especially his son’s team and Leicester City FC.
This post is Medcity influencers program. Anyone can share their perspectives on business and innovation in healthcare on MedCity News through MedCity Influencers. Click here to learn how.
