CHEST 2024 late-breaking data showcased the potential of AI-based systems in diagnosing and managing pulmonary embolism.
Artificial intelligence (AI) is emerging as a transformative tool in health care, with the possibility to enhance clinical decision-making and disease management. At the CHEST 2024 annual meeting in Boston, late-breaking data showcased the potential of AI-based systems in diagnosing and managing pulmonary embolism (PE).
The study, presented by Zeina Morcos, MD, an internal medicine resident at Staten Island University Hospital, compared tw2o AI models: ChatGPT-3.5 and DoximityGPT (utilizing GPT-4, tailored for health care applications). The team of investigators aimed to evaluate these models' effectiveness in diagnosing PE against traditional methods, particularly when CT angiography (CTA) results were not immediately available.
"We chose ChatGPT-3.5 and DoximityGPT to assess their ability to diagnose PE using a checklist derived from the Wells score and Pulmonary Embolism Severity Index (PESI)," Morcos explained during an interview with The American Journal of Managed Care® (AJMC®). The study involved 10 patients selected from the Pulmonary Embolism Response Team (PERT) at the Staten Island University Hospital.
Key findings from the study revealed that both AI models demonstrated significant alignment with clinical diagnoses made using CTA despite not having access to these imaging results initially. ChatGPT-3.5 matched the clinical diagnosis in 5 out of 10 cases, while DoximityGPT matched in 7 out of 10 cases. The difference, Morcos noted, lies in DoximityGPT's specialization in health care scenarios, potentially enhancing its diagnostic accuracy in clinical settings.
Beyond diagnosis, both AI models provided risk stratification and management recommendations that aligned closely with clinicians' assessments. "Even in cases where the AI models didn't match our initial diagnosis, they offered differential diagnoses that were logical and clinically relevant, such as considering concurrent conditions like COPD exacerbation or myocardial infarction,” Morcos added.
The results underscore AI's evolving role in augmenting clinical decision-making, especially in resource-limited environments or situations requiring immediate assessments. However, Morcos acknowledged concerns about the rapid pace of AI development and its implications for future health care practices.
"As AI continues to advance, integrating seamlessly into our daily practices, we must navigate carefully," she said. "While AI enhances efficiency and accuracy, its integration must be guided by rigorous evaluation and ethical considerations."
These late-breaking data from CHEST 2024 highlight a pivotal step towards leveraging AI for enhanced health care delivery, improved diagnostic precision, and patient care outcomes in the management of conditions like PE. As AI technologies grow in prominence and ability, ongoing research and collaborative efforts will be crucial in harnessing the full potential while ensuring patient safety and clinical efficacy.
Reference
Morcos Z, Achkar M, Grabie Y, Acharya S, Rotblat D. AI revolutionizes PE diagnosis: a comparative analysis of two AI models. CHEST. 2024;166(4):A6362.
Pediatric Alopecia Areata Risks and Disparities Vary by Race and Ethnicity
October 30th 2024Alopecia areata is more prevalent in Hispanic, Asian/Pacific Islander, and Black children than in non-Hispanic White children, with some Asian subgroups having a higher risk compared with those of Chinese ethnicity.
Read More
Exploring Racial, Ethnic Disparities in Cancer Care Prior Authorization Decisions
October 24th 2024On this episode of Managed Care Cast, we're talking with the author of a study published in the October 2024 issue of The American Journal of Managed Care® that explored prior authorization decisions in cancer care by race and ethnicity for commercially insured patients.
Listen