Artificial intelligence (AI) technology is increasingly playing major roles in data digitization, prediction analytics, and interoperability of digital healthcare data. Data digitization and integration of that data with structured and external data sets that offer a 360-degree view of the patient can provide actionable insights to providers, payers, and patients.
When it comes to data digitization, AI technologies like NLP (Natural Language Processing) and computer vision help with the process. NLP supports speech-to-text and vice versa, document and data conversions, patient notes, processing of unstructured data, and query support systems. Computer vision includes augmented reality (AR), virtual reality (VR), telehealth, and digital radiology.
ML algorithms facilitate more accurate detection of errors in billing and coding, leading to reduced claims denials. ML algorithms also are used for optimization of the supply chain for pharmaceuticals.
Deep learning and cognitive computing tools accelerate the processing of huge data sets, helping to inform precise and comprehensive risk forecasting and providing recommended actions that improve patient outcomes.
Predictive AI versus Generative AI
While predictive AI and Generative AI both use ML to learn from data, they do so in different ways and have different goals.
- Predictive AI is used to predict future outcomes by identifying patterns in historical data and then using those patterns to forecast future trends. For example, a predictive AI model can be trained on a dataset of patients’ longitudinal health records (LHRs) and then used to predict which patients are most likely to run into a specific health issue scenario in the future.
- Generative AI, on the other hand, can create new content, such as text, images, code, etc. This is done by learning from existing datasets and then generating new content, similar to the training data. For example, Generative AI can be used to analyze a patient’s tumor tissue and DNA and identify the genetic mutations driving the cancer. Based on this information, clinicians can recommend a personalized treatment plan that targets specific genetic mutations.
Direct applications of Gen AI in VBC
Value-based care (VBC) delivers rewards based on outcomes, not volume, as is the case in fee-for-service models. There are operational, financial, data complexity/accessibility, and technology-related challenges that need to be overcome to implement a value-based program effectively.
Below are several applications for Gen AI in VBC scenarios:
- Intelligent contract builder process(es). The VBC contract process today is very time-consuming and manual to develop, review, and put into action. Gen AI processes can help streamline this approach.
Using past contracts against which we can build and run large language models (LLMs), Gen AI can generate the new contract based on past patterns. Individual components of these contracts, such as different variables and their values, pricing information, attributes of different clauses, expiry dates, etc., can be extracted out of complex and lengthy contracts within seconds and presented to the user with a simple-to-use workflow in which users can finalize the contract within days.
- Improvements in care management process(es). Care management strategies for patients center around the effective use of data, processes, and systems by a team usually comprised of physicians, nurses, CBO (community-based organization) workers, care managers, and social workers. The basic concept is to have timely interventions for patients to reduce health risks and decrease the total cost of care.
Personalized care plans for patients broadly fall under four steps:
- Population stratification using risk stratification techniques
- Alignment of care management services to the needs of the patient (i.e., created while interacting with the patient in a personalized manner to ensure buy-in into the plan)
- Preparation of care plan and device monitoring for the patient for proactive care
- Association of appropriate personnel to establish care plan team for execution, follow-ups, etc.
Gen AI isn’t needed for patient risk stratification, which can be achieved using simple data analytics, post any data digitization (if needed for unstructured datasets), using a patient LHR. The challenge starts with contacting and engaging the patient. The communication protocols (emails, phone calls, SMS/MMS messages, snail mail) require persistent efforts to yield results. Gen AI can help personalize outgoing communication (conversational AI) based on past patient interactions, including any language translation preferences and level of education of the individual to keep the communication simple to understand.
Once the patient has been engaged, a care team can prepare the care plan and put the device monitoring/data collection protocols in place. Non-clinical and administrative steps like medication reminders, scheduling appointments on time, scheduling check-ins for a telehealth conversation, creation of alerts and notifications when things do not go as planned, Rx refills, prompting for daily exercise under the care plan – all can be personalized and automated using Gen AI.
For the Internet of Medical Things (IoMT), Gen AI could help companies create more personalized and patient-centered devices – incorporating software that allows for preventive maintenance and repairs.
The last part would be to help the care team navigate the complexity of the healthcare system – different workflows, assignment of the appropriate personnel based on their availability and expertise, providing insights to the care team about patients who are not yet that sick but could be if meaningful interventions don’t happen on time.
Other healthcare use cases for Gen AI
A good number of use cases are being worked on using Generative AI. Some are in research/concept stages, while a few are being deployed into production. These use cases fall under the following categories:
- Administrative tasks. Companies like Doximity, Abridge, and DeepScribe are working on solutions that automate administrative processes, such as documentation, claims handling, preauthorization and appeals, patient onboarding, and scheduling. These solutions will help reduce the administrative burden on physicians, nurses, and the healthcare staff as well as reduce human error.
- Prevention of costly medical errors. Lapses in patient safety kill thousands of people on an annual basis. Gen AI with direct video stream surveillance can observe physicians, nurses, and hospital staff, compare their actions to evidence-based guidelines, and warn clinicians when they’re about to commit an error.
- Medical education. We are in an age in which patients are very savvy and can easily find medical information online. Many times, this information is confusing, and might be incorrect. Gen AI-powered bots could be a reliable source of information for patients.
- Clinical decision support. A clinical decision support system can simulate how an experienced physician would make the decision based on the data available.
Pharma use cases
Gen AI is being used by pharmaceutical companies for:
- Drug discovery. NVIDIA now offers a set of Generative AI cloud services from which any company can do customization of AI foundation models to accelerate drug discovery and research work in the fields of genomics, molecular and cellular biology, and chemistry. This is already being utilized by startups, such as Evozyne and Insilico Medicine, as well as larger firms such as Amgen.
- Cancer research. Gen AI is being used to analyze patient tissue samples and employ functional precision oncology to improve patient outcomes.
- Clinical trials and precision medicine. Gen AI can accelerate and improve clinical trials and precision medicine therapies.
The success of VBC is heavily predicated on the ability of healthcare organizations to share and leverage data to optimize patient and population outcomes. Gen AI can be deployed across numerous use cases in healthcare that support VBC initiatives and goals, including contract-building, developing personalized patient care plans, reducing medical errors, clinical decision support, and streamlining administrative tasks. In addition, pharmaceuticals can use Gen AI to accelerate drug discovery, improve cancer treatments, and support clinical trials.
Rahul Sharma is chief executive officer of HSBlox, which assists healthcare stakeholders at the intersection of value-based care and precision health with a secure, information-rich approach to event-based, patient-centric digital healthcare processes – empowering whole health in traditional care settings, the home and in the community.