AI is a short acronym that encompasses a long list of technologies.
As healthcare organizations evaluate and implement artificial intelligence, there is a lot of confusion about what exactly it involves. The attention surrounding large-scale language models such as ChatGPT and generative AI has overshadowed other types of AI. Some healthcare organizations have been using artificial intelligence for years, perhaps without realizing they were using it.
Given the rapid growth in AI spending in healthcare, it is important for health systems to understand the different AI technologies, how to use them, and which ones offer the best value and return on investment (ROI). Systems have already invested heavily in AI software. According to Gartner, AI spending in healthcare and life sciences is expected to grow from $11.6 billion in 2024 to $19 billion by 2027, at a five-year CAGR of 16.6%. That’s too much money to spend without knowing exactly what you’ll get in return.
AI is best understood as a portfolio of complementary technologies and capabilities, some of which go beyond simply automating manual and repetitive management tasks to providing in-depth analysis, prediction, and optimizing outcomes and value. Some provide a course of action to achieve this. Here is a guide to the different technologies that can be classified as AI.
Machine Learning – This is the most mature technology in the AI portfolio and the one most systems are familiar with. It uses data and algorithms to enable AI to mimic the way humans learn, improving accuracy over time. Machine learning algorithms are commonly used to make predictions or classifications based on patterns in data. It is commonly used for purposes such as stratifying patients based on risk, identifying gaps in care, and providing personalized medicine to improve patient outcomes, especially for high-risk patients. will be done. It is also used to automatically scan medical images, helping radiologists proactively identify patients at risk of stroke or heart attack for intervention long before an acute event occurs. Deep Learning – This is a subset of machine learning and is closer to human reasoning. Multilayer neural networks, called deep neural networks, are used to simulate human decision making. Unlike machine learning models, which require structured and labeled input data to be effective, deep learning models can create accurate outputs from raw, unstructured data. One of the most common uses of this in medicine is image analysis. Natural Language Processing (NLP) and Natural Language Generation (NLG) – This uses machine learning to enable computers to understand and communicate human language. Combining computational linguistics with statistical modeling, machine learning, and deep learning enables computers and digital devices to recognize, understand, and produce text and speech. In the medical field, to translate medical records into plain English, analyze medical records, and provide an overview of a patient’s chart to improve information for doctors and nurses at the bedside/point of care (POC). used for computer-aided coding. Productivity. Generative AI/Large-Scale Language Models (LLM) – Similar to NLP, this AI can create original content such as text, images, video, audio, and software code in response to user queries. It can perform tasks such as powering online chatbots to schedule appointments and analyzing patient sentiment from various sources. One of the most compelling use cases for Gen AI/LLM, with much evidence shown in HIMSS 24, is the seamless sharing of nurse and clinician notes via the mobile phone running the application. The ability to edit irrelevant content while capturing and converting it to text. You can make final edits to these notes before automatically entering them into Epic’s electronic health record (EHR).
There are other technologies that are actually AI, although they are not necessarily thought of as such. This includes medical robotics and its subfields.
Robotic Process Automation (RPA) – Also known as software robotics, it employs intelligent automation technology to perform repetitive tasks such as extracting data, filling out forms, and moving files, freeing up humans to perform other tasks. Allows you to focus on yourself. It can also be used to improve call center operations and enable customer and patient self-service across multiple channels. Machine Vision – This allows medical devices to “see” the task they are performing and make real-time decisions based on that input. It helps with everything from identifying injuries and interpreting medical images to medication management and diagnosis. Advances in this field are paving the way for virtual reality (VR) and augmented reality (AR), both of which have great potential for robot-assisted surgery. This field also includes what we commonly think of as medical robots. It is a semi-autonomous machine that administers medicines, assists in surgeries and rehabilitation, monitors patients, and even serves as a companion for those who benefit from it. Robot-assisted surgery – Medical robots currently deployed in surgeries are equipped with 3D cameras to record the surgery. The video is streamed to a computer screen somewhere to assist the surgeon in performing the surgery using a surgical robotic arm such as the da Vinci Surgical System. This allows for minimally invasive surgery and rapid patient recovery, resulting in superior patient outcomes while reducing length of stay (LOS).
Building an AI portfolio
Faced with such pressing needs and promising technologies, how can healthcare organizations decide which AI to invest in?
There is no single answer. It is determined individually depending on each organization’s resources, needs, and priorities. No one AI technology fits all or solves all problems, so systems must prioritize technologies that promise the most value and ROI.
Factors organizations should consider, including the technology’s cost, ease of implementation, potential resistance from providers and payers using it, disruption to existing workflows, compatibility with existing systems, and potential savings. There are many. . Organizations also need to consider whether to build or buy AI technology. Although building one increases operational transparency, it may require resources and expertise that the system does not have.
Carefully building a portfolio of the most useful and impactful AI technologies is the best way to ensure your organization reaps the maximum benefit from this amazing innovation.
Identity Data Management (IDM) for AI data fidelity and readiness
Of course, critical to the success of any AI or analytical data program is the quality of the patient/member/consumer identity data used, including identity data management (IDM). Defective, missing, or duplicate data can hinder AI performance and make it difficult for organizations to achieve desired ROI and deliver value from their AI initiatives. For organizations to have a meaningful impact on their AI investments, they need high-quality IDM processes and resources.
Organizations lacking confidence in their IDM capabilities should partner with experts who can assess, benchmark, and enhance their operations to maximize the benefits from AI technology.
Photo: Warchi, Getty Images
Andy Dé is Verato’s Chief Marketing Officer, leading the go-to-market strategy, planning, and execution of Verato’s market-leading hMDM platforms and solutions. Prior to joining Verato, Dé held innovation, go-to-market, and product management leadership roles at SAP Health Sciences, GE Healthcare, Tableau, Alteryx, and MedeAnalytics. Dé is passionate about medical innovation and writes the Health Sciences Strategy Blog, which has readers in 47 countries. He has been quoted and published in major healthcare publications and is a member of the Forbes Communications Council and Fast Company Executive Committee. Mr. De received scholarships from leading institutions in the United States, Canada, and Israel to earn master’s degrees in engineering and business. He completed business administration programs at Harvard Business School, MIT’s Sloan School of Management, and Kellogg School of Management.
This post is published through the 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.
