Skip to content
DraconiticResearch
Search Sources API
226,742 studies31 sourcesUpdated 56 min ago
Peer-reviewedCC BY 4.0Journal of Clinical Medicine

Distinguishing Post-LASIK Ablation Patterns on Corneal Tomography: Comparing Clinician Versus Large Language Model

5 affiliations
  1. 1College of Medicine, University of Arizona-Phoenix, Phoenix, AZ 85004, USA
  2. 2Hoopes Vision, Hoopes Moshirfar Research Center, Draper, UT 84020, USA
  3. 3Department of Ophthalmology and Visual Neurosciences, University of Minnesota, Minneapolis, MN 55455, USA
  4. 4Department of Ophthalmology and Visual Sciences, John A. Moran Eye Center, University of Utah, Salt Lake City, UT 84112, USA
  5. 5Utah Lions Eye Bank, Murray, UT 84107, USA

Abstract

Background: The purpose of this study is to assess the ability of generative AI and a refractive surgeon to distinguish post-LASIK Scheimpflug tomography images based on original excimer laser ablation and to recognize untreated corneas. Methods: Corneal tomography Pentacam images from 204 eyes, which included hyperopic ablations, myopic ablations, myopic astigmatic ablations, mixed astigmatic ablations, or untreated eyes, were compiled. These were individually queried in ChatGPT-5.5 using a multiple-choice prompt to obtain a response. A refractive surgeon was also tested on the same 204 corneas. Results: The clinician obtained higher overall accuracy (85.8% [95% CI 80.2–90.3]) compared to ChatGPT, (60.3% [95% CI 53.2–67.1], p < 0.001) and was significantly more likely to answer correctly overall based on McNemar’s test. Additionally, ChatGPT underperformed in the myopia with astigmatism, mixed astigmatism, and untreated categories, with sensitivities of 38%, 6.3%, and 76%, respectively. Conclusions: The overall accuracy of ChatGPT-5.5 did not meet that of the clinician while ability to identify myopic and hyperopic ablation patterns was comparable. Both the AI model and the clinician had decreased accuracy in myopic astigmatism and mixed astigmatism ablation patterns, though the clinician still performed significantly better. Further studies involving alternative AI models, multicenter data, and standardized evaluation protocols are warranted to establish the clinical applicability of these findings.

Related

  1. Nomogram models integrating multiregion radiomics, quantitative computed tomography, and clinical indicators for distinguishing connective tissue disease-associated interstitial lung disease from idiopathic pulmonary fibrosis and improving the prediction of gender-age-physiology stage
    Quantitative imaging in medicine and surgeryPeer-reviewedCC BY-NC-NDFull textMedical imaging & radiologyEurope PMC
  2. Electromagnetic navigation bronchoscopy-guided three-dimensional conformal microwave ablation for unresectable peripheral high-risk pulmonary nodules: a mid-term assessment with no real-time intraoperative CT/cone-beam computed tomography guidance
    Yawei Liu, Yan Huang, Xiaodan Lin +7
    Translational Lung Cancer ResearchPeer-reviewedCC BY-NC-NDFull textOncologyEurope PMC
  3. Application of a multimodal model integrating optical coherence tomography angiography (OCTA) and radiomics for the identification of high-risk Nonarteritic anterior ischaemic optic neuropathy (NAION) populations and for distinguishing its acute phase.
    Peixian Yang, Songlin Cui, Yuxin Zhang +6
    BMC ophthalmologyPeer-reviewedCC BYFull textMedical imaging & radiologyEurope PMCPMC
  4. Durability and reconnection patterns at time of repeat ablation after first-time pulmonary vein isolation by monopolar focal pulsed field ablation
    Heart rhythm O2Peer-reviewedCC BY-NC-NDFull textCardiovascular medicineEurope PMC
  5. Pulmonary Vein Stenosis After Radiofrequency Ablation for Atrial Fibrillation ― A Computed Tomography-Based Analysis of Incidence and Risk Factors ―
    Yuka Oda, Akihiko Nogami, Ryuichi Usui +1
    Circulation JournalPeer-reviewedCC BY-NC-NDFull textCardiovascular medicineJ-STAGE
BibTeX
@article{vazquezloreto2026distinguishing,
  title = {{Distinguishing Post-LASIK Ablation Patterns on Corneal Tomography: Comparing Clinician Versus Large Language Model}},
  author = {Andrea C. Vazquez-Loreto and Sanjana Molleti and Hanna Pawlowski and Calisa G. Lee-Choi and Cole S. Elsner and Mina M. Sitto and Gabriel N. Medina and Niki C. Feiz and Phillip C. Hoopes and Majid Moshirfar},
  journal = {Journal of Clinical Medicine},
  year = {2026},
  month = nov,
  doi = {10.3390/jcm15228850},
  url = {https://www.mdpi.com/2077-0383/15/22/8850},
  note = {CC BY 4.0},
}
RIS
TY  - JOUR
TI  - Distinguishing Post-LASIK Ablation Patterns on Corneal Tomography: Comparing Clinician Versus Large Language Model
AU  - Andrea C. Vazquez-Loreto
AU  - Sanjana Molleti
AU  - Hanna Pawlowski
AU  - Calisa G. Lee-Choi
AU  - Cole S. Elsner
AU  - Mina M. Sitto
AU  - Gabriel N. Medina
AU  - Niki C. Feiz
AU  - Phillip C. Hoopes
AU  - Majid Moshirfar
PY  - 2026
DA  - 2026/11/12
JO  - Journal of Clinical Medicine
DO  - 10.3390/jcm15228850
UR  - https://www.mdpi.com/2077-0383/15/22/8850
AB  - Background: The purpose of this study is to assess the ability of generative AI and a refractive surgeon to distinguish post-LASIK Scheimpflug tomography images based on original excimer laser ablation and to recognize untreated corneas. Methods: Corneal tomography Pentacam images from 204 eyes, which included hyperopic ablations, myopic ablations, myopic astigmatic ablations, mixed astigmatic ablations, or untreated eyes, were compiled. These were individually queried in ChatGPT-5.5 using a multiple-choice prompt to obtain a response. A refractive surgeon was also tested on the same 204 corneas. Results: The clinician obtained higher overall accuracy (85.8% [95% CI 80.2–90.3]) compared to ChatGPT, (60.3% [95% CI 53.2–67.1], p < 0.001) and was significantly more likely to answer correctly overall based on McNemar’s test. Additionally, ChatGPT underperformed in the myopia with astigmatism, mixed astigmatism, and untreated categories, with sensitivities of 38%, 6.3%, and 76%, respectively. Conclusions: The overall accuracy of ChatGPT-5.5 did not meet that of the clinician while ability to identify myopic and hyperopic ablation patterns was comparable. Both the AI model and the clinician had decreased accuracy in myopic astigmatism and mixed astigmatism ablation patterns, though the clinician still performed significantly better. Further studies involving alternative AI models, multicenter data, and standardized evaluation protocols are warranted to establish the clinical applicability of these findings.
SN  - 2077-0383
ER  -