Ways Doctors Can Use AI in Clinical Practice, and How to Get Started
- Nisha Mehta, MD
- May 6
- 6 min read
As the breadth of AI tools for healthcare grow daily and capabilities grow stronger literally by the minute, it is becoming increasingly obvious that AI will change the practice of medicine. While we see sensational claims that artificial intelligence will render most doctors obsolete, we don’t think AI can replace physicians. What we do believe is that AI will transform clinical medicine, and we are proponents of using AI to ease clinician workload, to facilitate better communication and outcomes for patients, and to improve workflow and physician compensation. Below, we’ll cover ways that doctors can use AI and discuss several categories of existing AI tools as well as ones in the pipeline. Fundamentally, instead of being scared of AI (which is coming whether we like it or not), we think physicians should embrace AI and help influence what tools make it to market with the goal of helping clinicians and patients.
Disclosure/Disclaimer: This page contains information about our sponsors and/or affiliate links, which support us monetarily at no cost to you, and often provide you with perks, so we hope it's win-win. These should be viewed as introductions rather than formal recommendations, Our content is for generalized educational purposes. While we try to ensure it is accurate and updated, we cannot guarantee it. We are not formal financial, legal, or tax professionals and do not provide individualized advice specific to your situation. You should consult these as appropriate and/or do your own due diligence before making decisions based on this page. To learn more, visit our disclaimers and disclosures.
What are the major categories of artificial intelligence (AI) tools being developed in healthcare in 2025?
A 2024 McKinsey report showed that of their respondents from healthcare organizations (whether they be the physicians, other clinicians, administrators, payers, or otherwise), >70% stated they had already implemented AI or were planning on doing so in some capacity.
Go to a health tech conference, and you’ll see that there are literally hundreds of exhibitors claiming to incorporate AI solutions for healthcare. While each product has its own emphasis, broadly speaking, these are the categories that we see being emphasized in 2025.
AI scribing
Revenue cycle management
Prior authorization
Medical knowledge
Clinical decision support
Clinical diagnosis support
Administrative support tools
Clinical calculations
Clinician inbox management
Patient communications, like appointment scheduling, appointment reminders, billing, notes for work or school excuses, discharge instructions, and visit summaries
Patient education
Clinical trials
Which ways are doctors using AI tools in their daily practice, and what are some examples?
AI scribes / ambient scribing technology
There are many companies now in this space, as this was one of the quickest adaptations amongst physicians for AI in their daily use. This technology captures the clinical conversations between the clinician and the patient by recording the conversation and generating a clinical note for documentation purposes in real time (or shortly after the encounter concludes). It is essential to taking back some of the time (including after hours time) that so many physicians spend charting. Many AI scribing products are adding on billing and coding features, customizable templates, and more.
Related PSG Resource:
Abridge, founded by one of our physician members, offers PSG members 50% off AI scribing services, bringing the price down to only $99 dollars per month. You can try this free for one month without inputting payment information through our affiliate link with code PSG50, after which you will receive the 50% off price.
Medical references
The amount of knowledge out there that physicians have to keep track of never decreases - only increases. In an era of medicine where efficiency is key, RVUs determine compensation, and everyone is overworked, gone are the days where hospital teams can sit around having journal club for long periods of time and where clinicians are easily able to read the latest journals cover to cover. The fact is that keeping up with all the literature and clinical research and advancements is really hard when clinicians are spending so much time after hours doing administrative work (although we hope the AI scribes above can help with that)! Fortunately, nowadays, there are many tools available to help physicians identify data relevant to a patient’s clinical situation in a instantaneous chat GPT type model, such as:
Related PSG Resource:
OpenEvidence is the leading AI-powered medical information platform for physicians. It is currently used in over 7000 hospitals and care centers, and over 150 countries, serving millions of conversations per week. It is free and unlimited for physicians, and built from the ground up to give trustable, verifiable information. Every response is grounded in the high quality medical literature, spanning everything from clinical trials to meta-reviews to guidelines. Sign up today through our affiliate link.
Prior authorization forms
Anybody practicing medicine these days knows the pain of prior authorizations. Fortunately, there are many tools being developed to assist with all steps in the prior authorization process, from identifying those orders that need prior authorization and gathering relevant supporting data to navigating each individual insurance company's requirements to even generating a prior authorization letter or request for you.
Related PSG Resource:
OpenEvidence is free and unlimited for physicians, and can generate prior authorization letters while citing supporting medical literature, spanning everything from clinical trials to meta-reviews to guidelines. Sign up today through our affiliate link.
Inbox management
Getting physicians out of their inboxes so that they can actually engage in face to face patient care is a workflow struggle across many organizations. Being able to adequately triage lab results, patient requests, and more could dramatically improve efficiency, and as such, there are several relevant tools being built. These AI tools can prescreen incoming messages or alerts and direct them towards the appropriate people or care algorithms, thus saving the clinician time and energy.
Patient scheduling and communications and other front office tasks
Many front office tasks are quite algorithmic and can be handled without a traditional front desk team. This includes scheduling, appointment reminders, forms that need to be filled out prior to the patient's visit, and insurance verifications. There are AI bots that can do many of these things, and even be specifically created for or tailored to your practice’s workflow.
Billing and coding and revenue cycle management AI tools
There are seemingly innumerable tools being created in this arena, ranging from generating or verifying codes based on the note documentation all the way to complete revenue cycle management and everything in between. For example, AI tools can automatically review denials and generate supporting documentation and more. These AI tools can keep track of where claims are in the revenue cycle and in general optimize revenue cycle management, which of course is critical to physicians being paid for what they do.
Patient education and other forms
Clinicians can use AI tools to give their patients information about disease states, what to expect with a procedure, what items they’ll need at home after surgery, and more. AI tools can also generate forms for school or work for those who need an excuse or exemption. They can generate videos or even simulations of experiences. The possibilities here are endless.
Related PSG Resource:
OpenEvidence is free and unlimited for physicians, and built from the ground up to give trustable, verifiable information that can be used to generate patient information sheets and guidelines. Every response is grounded in the high quality medical literature, spanning everything from clinical trials to meta-reviews to guidelines. Sign up today through our affiliate link.
Clinical, diagnosis, and decision support AI tools
Many existing AI tools in existence or in development aim to keep an eye on the patient, give reminders to patients, identify possible red flags or diagnoses, allow for early detection of sepsis, and more. AI tools can detect suspicious nodules or other findings in radiology, detect abnormal rhythms or wave patterns on diagnostic studies, identify cancer cells on pathology, and more. AI can predict risk for different disease or infectious states by tracking patient data in real time and analyzing it, identify clinical trials that a patient may qualify for, and identify drug interactions or alert the clinician to possible allergies or relevant clinical history. Many remote patient monitoring and medication compliance tools exist. AI can pair up with wearable device technology to identify trends in differences in heart rate, sleep, and more to provide insights that may be difficult to capture in a standalone interaction at a 15 minute appointment.
Conclusion
These are just some of many use cases for artificial intelligence (AI) in development or already being widely used amongst physicians. We have no doubt that this list will continue to grow exponentially over the coming years (and even weeks). We’ll continue to try and update it when we see valuable tools coming to market. We are strong advocates for doctors being a part of healthcare innovation in this space!
Additional AI resources for physicians
Explore: