ChatGPT in medical writing: A game-changer or a gimmick?

Evaluating ChatGPT-4's Performance in Identifying Radiological Anatomy in FRCR Part 1 Examination Questions

Background Radiology is critical for diagnosis and patient care, relying heavily on accurate image interpretation. Recent advancements in artificial intelligence (AI) and natural language processing (NLP) have raised interest in the potential of AI models to support radiologists, although robust research on AI performance in this field is still emerging.

Objective This study aimed to assess the efficacy of ChatGPT-4 in answering radiological anatomy questions similar to those in the Fellowship of the Royal College of Radiologists (FRCR) Part 1 Anatomy examination.

Methods We used 100 mock radiological anatomy questions from a free Web site patterned after the FRCR Part 1 Anatomy examination. ChatGPT-4 was tested under two conditions: with and without context regarding the examination instructions and question format. The main query posed was: “Identify the structure indicated by the arrow(s).” Responses were evaluated against correct answers, and two expert radiologists (>5 and 30 years of experience in radiology diagnostics and academics) rated the explanation of the answers. We calculated four scores: correctness, sidedness, modality identification, and approximation. The latter considers partial correctness if the identified structure is present but not the focus of the question.

Results Both testing conditions saw ChatGPT-4 underperform, with correctness scores of 4 and 7.5% for no context and with context, respectively. However, it identified the imaging modality with 100% accuracy. The model scored over 50% on the approximation metric, where it identified present structures not indicated by the arrow. However, it struggled with identifying the correct side of the structure, scoring approximately 42 and 40% in the no context and with context settings, respectively. Only 32% of the responses were similar across the two settings.

Conclusion Despite its ability to correctly recognize the imaging modality, ChatGPT-4 has significant limitations in interpreting normal radiological anatomy. This indicates the necessity for enhanced training in normal anatomy to better interpret abnormal radiological images. Identifying the correct side of structures in radiological images also remains a challenge for ChatGPT-4.

Large-Language Models in Orthodontics: Assessing Reliability and Validity of ChatGPT in Pretreatment Patient Education

BACKGROUND

Patients seeking orthodontic treatment may use large language models (LLMs) such as Chat-GPT for self-education, thereby impacting their decision-making process. This study assesses the reliability and validity of Chat-GPT prompts aimed at informing patients about orthodontic side effects and examines patients' perceptions of this information.

MATERIALS AND METHODS

To assess reliability, n = 28 individuals were asked to generate information from GPT-3.5 and Generative Pretrained Transformer 4 (GPT-4) about side effects related to orthodontic treatment using both self-formulated and standardized prompts. Three experts evaluated the content generated based on these prompts regarding its validity. We asked a cohort of 46 orthodontic patients about their perceptions after reading an AI-generated information text about orthodontic side effects and compared it with the standard text from the postgraduate orthodontic program at Aarhus University.

RESULTS

Although the GPT-generated answers mentioned several relevant side effects, the replies were diverse. The experts rated the AI-generated content generally as "neither deficient nor satisfactory," with GPT-4 achieving higher scores than GPT-3.5. The patients perceived the GPT-generated information as more useful and more comprehensive and experienced less nervousness when reading the GPT-generated information. Nearly 80% of patients preferred the AI-generated information over the standard text.

CONCLUSIONS

Although patients generally prefer AI-generated information regarding the side effects of orthodontic treatment, the tested prompts fall short of providing thoroughly satisfactory and high-quality education to patients.