|
| |
| LETTER TO THE EDITOR |
|
| Ahead of print publication |
|
|
Comment on: “Understanding required to consider AI applications to the field of ophthalmology”
Thiago Goncalves dos Santos Martins
Department of Ophthalmology, University of Coimbra, Coimbra, Portugal
| Date of Submission | 05-Jun-2022 |
| Date of Acceptance | 20-Jun-2022 |
| Date of Web Publication | 25-Aug-2022 |
Correspondence Address:
Thiago Goncalves dos Santos Martins,
Botucatu Street, 821 Vila Clementino, São Paulo 04023-062
Portugal
 Source of Support: None, Conflict of Interest: None DOI: 10.4103/2211-5056.354538
How to cite this URL:
Martins TG. Comment on: “Understanding required to consider AI applications to the field of ophthalmology”. Taiwan J Ophthalmol [Epub ahead of print] [cited 2023 Mar 23]. Available from: https://www.e-tjo.org/preprintarticle.asp?id=354538 |
Dear Editor,
In the response to the article titled, “Understanding required to consider AI applications to the field of ophthalmology” published in your esteemed journal, which is a well-thought-out and written paper, I would like to raise few points regarding this study.
In the article, the author reports some performance limitations of diagnostic imaging of artificial intelligence.[1]
We should remember the limitations and discussions about ethical aspects in the development of artificial intelligence. Some biases that can occur in data collection can affect the training and development of the algorithm. Data must be validated in a geographically distinct population and validated by independent researchers to avoid bias in their development.[2]
Algorithms need to be further developed for the studies of multimodal and three-dimensional images. The analysis of images in three dimensions allows a better analysis of the patient's pathology.[3] The algorithm with three-dimensional analysis can be useful in planning retinal surgeries and monitoring intraocular pathologies. These technologies are already used in other areas of medicine, such as in the analysis of bone pathologies.[4]
Artificial intelligence and ophthalmology works in this way as a solution to barriers in the ophthalmic care of the population that would contribute to the reduction of visual impairment.
Financial support and sponsorship
Nil.
Conflicts of interest
The authors declare that there are no conflicts of interests of this paper.
| References | |  |
| 1. | Tabuchi H. Understanding required to consider AI applications to the field of ophthalmology. Taiwan J Ophthalmol 2022;12:123-9.  [Full text] |
| 2. | Faes L, Liu X, Wagner SK, Fu DJ, Balaskas K, Sim DA, et al. A clinician's guide to artificial intelligence: How to critically appraise machine learning studies. Transl Vis Sci Technol 2020;9:7. |
| 3. | Zhuge Y, Ning H, Mathen P, Cheng JY, Krauze AV, Camphausen K, et al. Automated glioma grading on conventional MRI images using deep convolutional neural networks. Med Phys 2020;47:3044-53. |
| 4. | Tajdari M, Maqsood A, Li H, Saha S, Sarwark JF, Liu WK. Artificial intelligence data-driven 3D model for AIS. Stud Health Technol Inform 2021;280:141-5. |
|
