ChatGPT shows superior diagnostic skills over radiologists
The study highlights how radiologist expertise affects ChatGPT’s diagnostic performance, with accuracy ranging from 60% to 80%.
A study from Osaka Metropolitan University revealed that ChatGPT, based on OpenAI’s GPT-4, has surpassed radiologists in diagnosing brain tumours. Researchers compared the diagnostic abilities of ChatGPT and radiologists using 150 MRI reports. ChatGPT achieved a 73% accuracy rate, slightly ahead of neuroradiologists at 72% and general radiologists at 68%.
The AI’s accuracy varied depending on the report’s author. It performed best with neuroradiologist reports, reaching 80% accuracy, while general radiologist reports saw the AI’s accuracy drop to 60%.
Researchers aim to explore its use in other diagnostic fields. They hope to enhance diagnostic precision and reduce the burden on medical professionals through AI integration. The study points to a future where AI might play a crucial role in preoperative tumour diagnoses.
Lead researcher Yasuhito Mitsuyama believes that these results indicate the potential of AI in improving diagnostic processes. The team is optimistic about its future applications in medical education and imaging technologies.
About author
DW Team
The GIP Digital Watch Observatory team, consisting of over 30 digital policy experts from around the world, excels in the fields of research and analysis on digital policy issues. The team is backed by the creative prowess of Creative Lab Diplo and the technical expertise of the Diplo tech team.
