AI Model ChatGPT Outperforms Doctors in Assessing Eye Problems, Study Suggests
Artificial intelligence (AI) has made significant advancements in recent years, with technologies like ChatGPT showing promising potential in various fields. A recent study conducted by academics from the University of Cambridge suggests that ChatGPT could be better at assessing eye problems than doctors, raising the possibility of using AI models in triaging patients and providing specialist care.
In the study, researchers tested the ability of ChatGPT 4 against the knowledge of medics at different career stages, including junior doctors and eye specialists in training. The language model was trained on 374 ophthalmology questions and then tested in a mock exam of 87 questions. The results were compared to those from expert ophthalmologists, trainee ophthalmologists, and junior doctors, as well as other language models like Llama and Palm2.
The findings of the study revealed that ChatGPT 4 scored 69% in accuracy, outperforming other language models and even some of the medical professionals. The expert ophthalmologists achieved a median score of 76%, while trainees scored 59% and junior doctors scored 43%. This suggests that language models like ChatGPT are nearing expert-level performance in assessing complex eye issues.
Lead author of the study, Dr Arun Thirunavukarasu, highlighted the potential benefits of deploying AI in triaging patients with eye problems. He stated that AI could help determine which cases require immediate specialist care, which can be seen by a general practitioner, and which do not need treatment. This could streamline the healthcare process and ensure prompt care for patients with eye issues.
While the study emphasizes the potential of AI in assisting healthcare professionals, Dr Thirunavukarasu noted that doctors will continue to be in charge of patient care. He stressed the importance of empowering patients to make individual decisions about involving computer systems in their healthcare. Ultimately, the goal is to provide patients with the best possible care while leveraging the capabilities of AI technology.
The study’s findings have been published in Plos Digital Health, showcasing the progress and potential of AI models in the medical field. As technology continues to advance, it is essential for healthcare professionals to embrace innovative solutions that can enhance patient care and improve overall healthcare outcomes. By integrating AI technologies like ChatGPT into healthcare practices, we may see significant improvements in patient triage, diagnosis, and treatment.