Table 2 Exemplary applications of LLMs

Connor & O’Neill³⁹

Supporting initial diagnosis and triaging of patients by fine-tuning LLMs on a specialised dataset of electronic medical records, clinical notes, sports science and medicine literature.

Stewart et al.⁵²

Using traditional and modern natural language processing to triage patients on arrival based on structured data and unstructured free-text history of presenting complaint to predict risk stratification. This includes predictions on the likelihood of admission to hospital, prediction of critical illness, prediction of triage score, prediction of provider-assigned chief complaint, prediction of investigation, and prediction of infection.

Buzzaccarini et al.⁶⁰

Enhancing patient consultations by providing accurate and reliable information on aesthetic procedures, their risks, benefits and potential outcomes, enabling well-informed decisions and improved treatment outcomes.

Currie⁸⁰

Providing language translation and helping health professionals to communicate with patients speaking foreign languages; helping health professionals to educate their patients and empower patients to take an active role.

Schmälzle & Wilcox⁵⁰

Using LLMs to create an AI-guided message creation system to disseminate health-related information via social media.

Cheng et al.⁶¹

Using ChatGPT to monitor news and social media platforms for signs of outbreaks of disease clusters and to alert health professionals to potential threats.

Agbavor & Liang³⁵

Using GPT 3 to distinguish individuals with Alzheimer's disease from healthy controls and to infer cognitive testing scores based on linguistic features. It is shown that this approach outperforms conventional approaches and performs comparable to specifically fine-tuned models. Usable as a web app in a doctor's office.

Rau et al.⁴⁹

Supporting radiologists' diagnostic performance by providing imaging recommendations in accordance with recent guidelines.

Arslan⁵⁸

Using ChatGPT to provide personalized recommendations on topics such as nutrition, exercise and psychological support in obesity treatment.

Cheng et al.⁶¹

Using ChatGPT to provide treatment recommendations based on patients clinical presentations, disease severity, and comorbidities.

Yeo et al.⁵⁵

Using ChatGPT as an informational plattform to comprehend and to respond to cirrhosis related questions in different languages, adressing barriers that may impact patient care.

Knebel et al.⁴⁵

Using ChatGPT for the assessment of acute ophtalmological conditions with regard to triage accurracy and recommendations for preclinical measures.

Hosseini et al.⁴⁴

Using LLMs to increase efficiency in note-taking through prepopulation of forms, voice recording and converting recordings into clinical notes, or synthesizing existing patient notes to save clinicians' time.

Gottlieb et al.⁸²

Using Conversational AI to create study documents by translating complex concepts into simpler ones or designing informed consent documents for patients.

Guo et al.⁷⁹

Using a ChatGPT-like (ProteinGPT) systems to accelerate protein research. The model is aimed at learning and understanding protein 3D structures. ProteinGPT enables users to upload proteins, ask questions, and engage in interactive conversations to gain insights.