Natural language processing, or NLP, is a branch of artificial intelligence that allows machines to understand human language. In recent years, NLP has become a part of consumer products and operates with sometimes surprising levels of accuracy.
Virtual assistants use NLP to understand spoken questions like “What’s the weather like today?” They can contextualize the meaning of the question, then use data from online sources to respond with meaningful answers.
Medical NLP has long been a subject of research and development, as it holds considerable potential for uncovering meaningful insights from unstructured medical text. This presents a great opportunity for healthcare providers, clinical researchers, and payers to discover valuable information regarding disease progression, treatment efficacy, trends, and trends. population health trends and many other use cases that cannot be identified using manual data analysis and evaluation techniques.
However, progress has historically been slow due to the complexity inherent in medical language.
Healthcare IT News Tim O’Connell, a practicing radiologist and CEO of emtelligent, a medical NLP technology company, to talk about how NLP technology works with child prose how NLP is still poorly understood, and how successful medical NLP can only be if it caters to clinicians.
Q. Please break down NLP for our readers. How does NLP technology read, understand, and structure human prose (e.g. doctor’s documents)? Furthermore, what does NLP then allow clinicians, technicians, and researchers to do?
ONE. Natural language processing has existed since the earliest days of computers. NLP combines computer science, linguistics, and artificial intelligence to create software that can understand or even generate human language. Today we will only talk about NLP software that can understand human language, not the kind used to generate text.
NLP software uses a variety of methods to read text and “understand” some or all of the content it is provided with. A prime example in the medical field is why finding electronic health records without NLP can be so difficult.
Suppose a doctor wants to look up a patient’s histogram to determine if they have hepatitis. Using only keyword-based search can return dozens or hundreds of unrelated visits, such as a family history of hepatitis or whether the patient has been tested. But if clinicians can use NLP software to read and understand patient charts, they can search much more efficiently. It would be like asking a smart assistant, “Does this patient have hepatitis?”
NLP solves the problem of scale needing humans to read medical texts to understand it. When you can use computers to understand large amounts of medical text, you open up new use cases for things like automated adjudication in insurance, predictive analytics, bridging the gap in care, developing new AI models, better clinical applications for care providers, etc.
Q. NLP is a breakthrough in healthcare. But you assume the technology is still poorly understood, with confusion over how NLP and AI work together. Please explain.
ONE. Yes, there is still a lot of confusion about NLP and AI. Although NLP is a field under the broad umbrella of artificial intelligence, when people think of capital-A capital-I “Artificial Intelligence”, they often think of computers that: 1) Independently decision making to perform classification or prediction tasks, or 2) Natural interactions with humans.
NLP is often more than just that; rather, it is a required component of almost any AI “system”.
For example, to use the most “Hollywood” use cases, when a robot is talking to a person, NLP is used to generate the robot’s speech and to understand the person’s response. Applying those capabilities to our use cases, if your goal is to create a medical AI model that predicts a patient’s cancer recurrence, you’ll want to use NLP software. to manage the training data well, so the model has received the necessary information.
If you are training an X-ray model to look for pneumonia on a chest X-ray, you may need hundreds of thousands of images labeled pneumonia, so you will be running NLP software on newspapers. Report X-rays to find positive cases of pneumonia to include in your training kit. .
In short, NLP is both a type of AI and a tool used “behind the scenes” to create and improve AI software. NLP software will be used to extract training data from patient graphs to create new AI models that can predict and prevent medical errors and illnesses, improving outcomes for patients.
Q. You think NLP software can only be successful if it caters to the clinician, something that has not really been achieved. Please explain what you mean and what it will take NLP to get where you think it needs to go.
ONE. There’s a saying that “medical school is a four-year terminological course” and it’s true. I mention this because I truly believe we need good medical NLP software, created by clinicians and NLP experts working together, that really understand the content of the medical text. economic.
One problem I hear from clinicians is that older NLP systems don’t read text the way doctors or nurses do; they said it misinterpreted the word, didn’t absorb the negative word properly, or was confused by the lists in the report.
When people using really great NLP software can understand the original meaning of medical texts, there will be a whole new world of possibilities for improving healthcare systems and patient care. we. NLP can be used to create new applications such as automated patient summaries, as well as intelligent search and documentation tools that allow them to spend more time with patients and reduce screen time. image.
By applying NLP to analytics and data science, healthcare facilities, payers, and governments will be able to receive higher quality data about patients. Some of the inefficiencies of our healthcare system are due to lack of data because it is too expensive to pay people to extract that data from the graphs.
NLP solves this problem and can provide administrators and analysts with the data they need to make better, more informed, and faster decisions.
Follow Bill’s HIT coverage on LinkedIn: Bill Siwicki
Email the writer: [email protected]
Healthcare IT News is a publication of HIMSS Media.
