By Satish Kumar
In recent years, digital technologies have been widely adopted across health institutions. It has helped Health care Organizations to reduce efficiency, improve access and improve quality of care. Digital healthcare is also helping to prevent chronic diseases. This is one of the key technologies driving change artificial intelligence. It is transforming the entire value chain of healthcare delivery, including complex clinical and administrative processes. Many of these use exciting AI applications Natural Language Processing.
Natural Language Processing (NLP) is a sub-branch of AI that deals with algorithms and models to process language as humans do. NLP uses deep learning algorithms to process unstructured data into structured data. Also in Clinical Notes and Physician’s Opinions Electronic Health Records (EHR) systems are typically in unstructured text. NLP systems can analyze this unstructured data and derive useful medical insights. Other AI technologies such as NLP and genome analytics used with record linkage can help deliver personalized medicine.
NLP systems can shift through newly generated medical knowledge in articles and research papers can create useful summaries of it. The availability of these summaries, linked to the EHR, will help clinicians make more informed clinical decisions and optimize treatment plans.
Healthcare is transitioning to value-based care, and one of the critical success factors in this transition is the patient experience. A medical virtual assistant can help collect patient information such as demographics, insurance, etc. It can navigate the patient to enter self-measured physiological parameters. In a question-and-answer or chat format, this visual assistant asks the user about a symptom or illness. Once the assistant gets an answer, it matches the possible diagnosis and creates possible scored results for the next step – which could be another question, forwarding to a human agent, or telling the patient what to do next. Creating this intelligence requires that the agent use NLP to tokenize the answer, create feature sets, and perform data matching using pre-trained deep learning models.
Training these deep learning models requires large amounts of data about disease causes, symptoms, and diagnoses. Most large hospitals and healthcare providers have electronic health records, which must be identified before the model can be used for training and validation. While virtual assistants have greatly improved their accuracy over the past few years, the best results can still be achieved by pairing with a human physician.
NLP-based assistants can also be used for routine administrative tasks such as self-scheduling/re-scheduling/cancelling an appointment at a healthcare facility by the patient. For better adoption, they can be made as part of mobile apps and voice-enabled.
NLP and AI can also play a big role in disease diagnosis and disease prevention. AI systems are used in surveillance and outbreak investigation. These are the main functions of public health and important to protect the population from diseases and types of infection. AI systems can help prioritize medications or vaccinations when capacities are limited.
In pandemics, hospitals have used pre-hospital digital triage to direct patients to appropriate care settings such as home quarantine, respiratory clinics, testing sites, surgery facilities, emergency rooms, etc.
NLP and other AI technologies are playing an important role in public health’s transition to precision public health. This will help deliver the right interventions to the right population at the right time and includes consideration of the social determinants of health.
By analyzing structured/semi-structured data in EMR/EHR systems using AI, risk and vulnerable populations can be identified. This can help deliver tailored public health messages and other public-health measures.
Using NLP to analyze social media feeds, newspaper clips, and other media content can aid in environmental scanning and situational awareness. Public health notifiable disease information along with clustering and spatial analysis will also help identify new disease or pathogen outbreaks.
These two datasets can be cleaned, processed and combined with other datasets to train context-aware machine learning models to estimate disease prevalence and burden. The prevalence and burden of disease and their projected values in the future will help the authorities to plan for excess capacity in medical supplies and hospital capacity.
NLP powered question-answer platforms/chatbots can provide personalized support or advice on relevant topics. They can also help identify patients’ social needs during hospital visits.
NLP can also help with topic modeling and sentiment analysis, measuring compliance/acceptance of public health measures and public concerns.
The development of this NLP model will require different datasets owned by many different organizations that need to be connected. There is also a need to exchange data from testing laboratories and healthcare providers.
Data exchange between different healthcare organizations and public health systems should be done through the HL7/ISO data exchange protocol.
Changing disease prevalence and burden will require new public health and clinical guidelines. Deep learning models need to be continuously trained with new/updated data and new/updated public health guidelines.
For NLP systems to be consistent, this entire process, from data acquisition to training of deep learning models, must be automated. Automation of this will ensure that Q&A virtual assistants are always up to date.
The adoption of AI in organizations has increased manifold in the last two years. For sure, AI is going to change healthcare jobs for the better. Insights and Visualizations. It will enhance the decision making of healthcare workers. The future application of AI will be more towards explainable AI (XAI). Whether patients or clinicians need to understand the reasoning behind decisions, XAI is as important to increasing trust in medical AI as humans. XAI is also going to accelerate the adoption of AI in public health and healthcare organizations.
By Satish Kumar, CEO Superna System
(Disclaimer: The views expressed are solely those of the author and not necessarily subscribed to by ETHealthworld. ETHealthworld.com shall not be responsible for any loss directly or indirectly caused to any person/organization.)
