OR WAIT null SECS
Findings suggests implementation could save time and money, but needs to be evaluated carefully.
According to a recent study on artificial intelligence (AI) by JAMA Cardiology, model applications show promise for improving the accuracy and speed of event classification in multicenter clinical trials. The study examined the potential for natural language processing (NLP) models to accurately classify heart failure (HF) hospitalizations compared with clinical adjudication. Although the use of AI in medicine is still a relatively new concept, it is already revolutionizing the industry with new applications being announced on a daily basis.1,2
“NLP is a promising strategy for identifying clinical events from medical record text at scale. In this study, the single-center [community care cohort project (C3PO)] NLP model for adjudication of HF hospitalizations agreed with the gold-standard human [clinical events committee (CEC)] 87% of the time in a multicenter clinical trial. Fine-tuning the C3PO NLP model or training a new NLP model further improved agreement with the CEC to 93%, which was similar to human reviewer reproducibility,” explained the authors of the study.2 “These results represent a first step toward application of NLP to streamline the identification of clinical outcomes scalably in clinical trials and observational cohorts, although further validation studies including other EHR systems, health systems, and countries and integration with methods to obtain medical records are needed.”
Despite its potential, there are some stakeholders who believe implementing AI comes with risk. According to Forbes, the biggest risk with AI in clinical trials is the potential for accidental exposure of patient data and data generation bias. If it’s working on input data that may be biased against particular patient profiles—be it gender, economic resources, or geography—it is likely that AI algorithms could pick up on this bias. As a result, patient data will be compromised, and profiles could face rejection. The article emphasizes the importance of setting the correct metrics in AI to ensure ethical and unbiased decision-making.3
There remain numerous ways that AI can continue to aid clinical trials as technology advances. However, authors of the study published by JAMA Network Open said that each case needs to be evaluated with great care.1
“Importantly, each use case will require careful evaluation, and as use cases become more complex, more comprehensive review may be needed. In the case of event adjudication, it is reasonable to evaluate AI-generated output against the traditional adjudication approach,” said the authors of the study, titled The Future of Clinical Trials: Artificial to Augmented to Applied Intelligence. “A framework for oversight is needed to ensure that historical biases do not get baked into future trials—and thus future health care—purely because of model momentum. With risks appropriately managed, benefits can be safely and ethically realized.”
The importance of greater use of AI in healthcare is evident when examining the global cost of clinical research and trials, which is more than $50 billion and projected to balloon to more than $85 billion by 2030.1
“Given that up to 20% of clinical trial costs is related to the manual transfer and verification of data from the electronic medical record to a data capture system, as well as the site monitoring of such data, all of which are readily automated tasks, the application of AI to clinical trials could be revolutionary and is likely necessary,” the authors wrote.
The researchers added that AI has significant potential to enhance clinical trials across multiple areas; however rigorous validation and regulatory oversight must be in place to facilitate safe, effective, and ethical deployment across the clinical trials ecosystem.1
They note that these model outputs must produce an accurate assessment of the health state used to analyze the treatment benefit and risk, and must address risks regarding data privacy, security, and bias.
“Because these new technologies also carry significant risks, including the risk of exacerbating inequities, promoting open science and improving understanding of the ‘black box’ of AI is a must,” the authors wrote.1 “As we learn, we must find ways to share weaknesses, flaws, or failures without undermining the integrity of the process. Indeed, the conduct of high-quality evaluation of AI tools using rigorous designs and analysis and published in peer-reviewed publications is critical to generating high-quality evidence on the benefits and risks of using AI to enhance clinical trials.”
1. The Future Of Clinical Trials: Artificial to Augmented to Applied Intelligence. JAMA Network. November 11, 2023. Accessed November 14, 2023. https://jamanetwork.com/journals/jama/fullarticle/2811988
2. Natural Language Processing for Adjudication of Heart Failure in a Multicenter Clinical Trial: A Secondary Analysis of a Randomized Clinical Trial. JAMA Cardiology. November 11, 2023. Accessed November 14, 2023. https://jamanetwork.com/journals/jamacardiology/fullarticle/2811977
3. AI For Efficient Clinical Trials: It's Risky, But Optimism Rules. Forbes. November 14, 2023. Accessed November 14, 2023. https://www.forbes.com/sites/forbestechcouncil/2023/11/14/ai-for-efficient-clinical-trials-its-risky-but-optimism-rules/?sh=4e6e75f9318c