Although there are many similarities between the use of artificial intelligence in adult care and pediatrics, there are also unique and important considerations for the latter, due to the complexity of health problems in childhood care and adolescents.
These include understanding the subtleties of relations with patients -practitioners – including relationships with parents and caregivers – as well as confidentiality and confidentiality considerations as children go to adult care.
Supplier organizations must recognize these distinctions and ensure that AI is used in an appropriate manner to improve the experiences of patients, family and providers in a pediatric environment.
AI applications fed by natural language treatment and large languages models modify the way suppliers engage with patients, explains Dr. Natalie Pageler, Director of Health IT at Stanford Medicine Children's Health.
We recently told him about the special considerations necessary for a safe and effective use of AI in pediatric contexts.
“Children change so much”
The complex relationships between doctors and their young patients as they grow require that supplier organizations “are committed to the right person at the right time,” said Pageler.
“Children change so much – both age and when, where and how they take control of their health care during their transition to adulthood. Suppliers must understand how clinical notes are written, what information is shared through patient portals and which can access it, as well as to engage in online conversations via these platforms.
“NLP is used to better understand who speaks through patients of patients-is it a parent, a caregiver, a brother or a sister or the patient himself?” She continued. “Technology is increasingly used to examine clinical notes, helping to identify shared information and what should remain private, depending on which access.”
LLMs are used to deal with large quantities of metadata to support patient experiences. They can evolve the speed at which reports and other results are examined, process this information and better understand where attention must be concentrated.
This allows suppliers to improve patient experiences and the quality of care, while drawing attention to the areas that can relieve administrative burden thanks to automated support, if necessary, she added.
“To move forward, AI and other significant technological progress in health care will continue to be integrated into pediatrics,” said Pageler. “If necessary, the AI will also have an impact on the level of personalized care that we can provide.
“For example, adapting precise care for pediatrics,” she added. “Historically, this was a challenge because the data sets are much smaller and, fortunately, the majority of children are generally healthy. We see early signs that AI can support more precise care and precision medicine, for example, in genomics. “”
Additional pediatric considerations
Pediatric use cases for AI are generally more difficult to implement and require an additional list of considerations before being deployed. One of the best examples is the use of AI and ambient scribe technologies to support the taking of notes during the patient-patient interactions.
“Here is the challenge: unlike typical common individual conversations in adult care, pediatric visits often involve many different people in the variable ages room,” noted Pageler. “Rendez-vous can include whole families, sick children who cry, several caregivers and providers who meet multidisciplinary needs.
“Consequently, appropriate tests are necessary to determine the effectiveness of the ambient AI in more dynamic contexts, ensuring that it supports the commitment of providers with the patient and the family,” she continued.
AMA AMBIANT for creation and notation release suppliers to focus on the patient and their families. In particular in pediatrics, many rich data can be captured at the care point – a large part of which is behavioral or non -verbal.
For example, important ideas can be found in interactions between parents and families, the subtle nuances of the behavior of a child, body language and emotions – and how it is perceived by those of the room.
Do suppliers lack clues?
“Provider providers who are busy typing notes may lack essential indices to provide precise and specialized care,” said Pageler. “”The AI ambient scribes allow providers to spend more time with patients, strengthen relationships and relieve the load of administrative tasks.
“Unlike adults, where there is fewer considerations of confidentiality and confidentiality, ambient AI and other automated notes require additional surveillance levels,” she continued. “Suppliers must be aware of the sensitive information that is transcribed and determine whether it is correct – in addition to knowing who can access this information.”
Pageler sees AI applications helping to report potential areas of concern and support suppliers to draw attention to the most important information.
When evaluating deployments and AI algorithms in pediatric populations, there are four key considerations – or obstacles to overcome – currently at the top of the agenda, she said.
Ask
“First, the industry must have started and educated parties at the table to ask the right questions during the development and deployment of technology,” she suggested. “We need pediatric providers, as well as patients and families, to make sure we ask the right questions of computer leaders.
“Second, understanding the data sets that we have and these applications are built,” she continued. “Today, many industry data sets are focused on adults. This makes technological development in pediatrics more difficult. In addition to having smaller pediatric data sets, pediatric data is subject to stricter protections – and rightly.”
It is important to understand where existing pediatric data is insufficient, so suppliers must consider the additional data necessary to effectively carry out these projects, she added.
“Third, even with good data, it can be ineffective based on the complex nature of pediatric populations and the diversity of patients and families looking for care,” said Pageler. “Models built on adult data can have useful pediatric applications, but they must be tested in depth.
“Children's physiology and care needs change considerably with age, newborns to adolescents,” she continued. “Thus, algorithms formed by adults must undergo rigorous tests in different age groups. Take smart watches, for example. These are mainly designed for adults, but our cardiologists see successful applications in older children and adolescents who are fairly elderly to reliably wear an intelligent watch for prolonged periods.”
Right intervention required
Finally, no AI application is significant without the right intervention, she said.
“Pediatricians must be involved in The development of AI applications to help contextualize information, understand data and identify the appropriate intervention, “said Pageler.” For example, AI algorithms can help predict when it is unlikely that a patient will come for an appointment on the clinic.
“Instead of surrening the clinic's calendar to ensure efficiency, pediatric organizations can implement interventions such as reminders, childcare and transport support to allow pediatric families to attend their critical clinical appointments,” she added.
The pediatric health industry is supported by the potential of AI to strengthen links between patients, families and providers to improve the results of care. Pageler said Stanford has long argued that technology should not be deployed for technology.
“The use of the ambient AI that we have explored have an unexploited opportunity and a potential to reinvent all the pediatric patient experience,” she noted. “The pandemic revealed the power to see patients and families in different environments such as home, where these interactions can present richer and more usable data for providers.
“The ambient AI allows this trend to continue to evolve and strengthen during experience in person,” she continued. “This maintains the suppliers focused on the family and the complex dynamics in play. This also helps to relieve the burden of the documentation.
Positive results potential
Looking beyond suppliers, this technology shows a potential for nursing care and other clinical applications in pediatric health care that contributes to positive results, she added.
“This includes the use of AI for clinical decision-making assistance,” she said. “Stanford has been at the forefront of clinical IT and decision -making tools for more than a decade – and the idea of helping to accelerate and evolve these applications to provide almost real time support is promising.
“We also see early signs that AI can be an effective tool to help assess images, videos and other forms of visual data that add to the richness of each commitment of patients and providers,” she continued.
“Likewise, in the environments where providers cannot directly observe a child outside the examination room or while taking care of other patients, AI can help process and assess information to strengthen the care continuum” always on '' without adding load to the supplier. “
Stanford Medicine Children's Health also has a team to seek how AI can support the diagnosis of early autism and improve the development of therapy for children.
Follow Bill's successful coverage on LinkedIn: Bill Siwicki
Send him an email: bsiwicki@himss.org
Healthcare It News is a publication of the Himss media.
Watch now: Generative advice from the AI of a superior officer of chief ia ↓ armpits
