Evaluating AI‐powered text‐to‐image generators for anatomical illustration: A comparative study
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Medical illustration, which involves the creation of visual representations of anatomy, has long been an essential tool for medical professionals and educators. The integration of AI and medical illustration has the potential to revolutionize the field of anatomy education, providing highly accurate, customizable images. The authors evaluated three AI-powered text-to-image generators in producing anatomical illustrations of the human skulls, heart, and brain. The generators were assessed for their accurate depiction of foramina, suture lines, coronary arteries, aortic and pulmonary trunk branching, gyri, sulci, and the relationship between the cerebellum and temporal lobes. None of the generators produced illustrations with comprehensive anatomical details. Foramina, such as the mental and supraorbital foramina, were frequently omitted, and suture lines were inaccurately represented. The illustrations of the heart failed to indicate proper coronary artery origins, and the branching of the aorta and pulmonary trunk was often incorrect. Brain illustrations lacked accurate gyri and sulci depiction, and the relationship between the cerebellum and temporal lobes remained unclear. Although AI generators tended toward esoteric imagery, they exhibited significant speed and cost advantages over human illustrators. However, improving their accuracy necessitates augmenting the training databases with anatomically correct images. The study emphasizes the ongoing role of human medical illustrators, especially in ensuring the provision of accurate and accessible illustrations.
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.
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 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.002 |
| Science and technology studies | 0.000 | 0.000 |
| 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 it