Conversational AI Models for ophthalmic diagnosis: Comparison of ChatGPT and the Isabel Pro Differential Diagnosis Generator
Bibliographic record
Abstract
With the rapidly growing field of conversational artificial intelligence (AI), it is becoming increasingly likely that these technologies will revolutionize the way physicians diagnose and treat patients. The purpose of this study is to evaluate the use of conversational AI language models, specifically ChatGPT, for the diagnosis of ophthalmic disease, and to compare it to existing tools, namely the Isabel Pro Differential Diagnosis Generator. Prospective, comparative evaluation of ChatGPT and Isabel in formulating provisional and differential diagnoses from a set of rich text case report descriptions. Ten ophthalmology patient cases were selected at random from a publicly available online database of ophthalmic case reports. The text details of each case were input into ChatGPT and Isabel. Their ability to identify the actual diagnosis and provide relevant differential diagnoses was compared. ChatGPT identified the correct diagnosis in 9/10 cases while having the correct diagnosis listed in all 10/10 of its lists of differentials. Isabel identified only 1/10 provisional diagnoses correctly, however it included the correct diagnosis in 7/10 of its differential diagnosis lists. The median position of the correct diagnosis in the ranked differential lists was 1.0 (IQR 1.0 to 2.8) for ChatGPT versus 5.5 (IQR 3.3 to 10.0) for Isabel. Conversational AI models such as ChatGPT have potential value in the diagnosis of ophthalmic conditions, particularly for primary-care providers. As further iterations of models are deployed, additional studies investigating their capabilities are needed to determine the best ways to integrate them into practice.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.008 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".