|
 |
| LETTER TO THE EDITOR |
|
| Year : 2021 | Volume
: 14
| Issue : 3 | Page : 203-204 |
|
|
Artificial intelligence to predict ocular manifestation of COVID-19
Dimple Nagpal1, Nayan Gupta2
1 Chitkara University Institute of Engineering and Technology, Chitkara University, Rajpura, Punjab, India
2 Chitkara School of Health Sciences, Chitkara University, Rajpura, Punjab, India
| Date of Submission | 14-Dec-2020 |
| Date of Decision | 23-May-2021 |
| Date of Acceptance | 24-May-2021 |
| Date of Web Publication | 20-Oct-2021 |
Correspondence Address:
Nayan Gupta
Chitkara School of Health Sciences, Chitkara University, Rajpura, Punjab
India
 Source of Support: None, Conflict of Interest: None  | Check |
DOI: 10.4103/ojo.ojo_464_20
How to cite this article:
Nagpal D, Gupta N. Artificial intelligence to predict ocular manifestation of COVID-19. Oman J Ophthalmol 2021;14:203-4 |
Sir,
The surge of coronavirus across the globe has enforced clinicians, health-care workers, scientists, and governments to work together to strategize the preventive measures against this pandemic. It is now established that coronavirus and its variants have various systemic association along with ocular manifestations. Various published literatures report ocular manifestations in the middle phase of the disease.[1],[2],[3] Most of these anterior segment signs such as conjunctival congestion, epiphora, and increased secretions due to coronavirus are reported among patients with severe pneumonia. Other ocular abnormalities presented in animal models include anterior uveitis, optic neuritis, and retinitis. These occurrences of the ocular symptoms may appear as the initial sign or during the spread of COVID-19. Mucormycosis is another sight-threatening sign reported among COVID patients.[4]
Important measures for a safe ophthalmic practice during this pandemic include various measures such as stratification of patients for clinical visits, protection of medical workers, surveillance of medical workers, sterilization, and diagnostic techniques. [Figure 1] illustrates the steps for safe ophthalmic practice.
Ocular symptoms can be diagnosed through varying imaging modalities for screening of COVID-19-related symptoms through computer-aided diagnosis. The collaboration between clinicians and health-care industry workers is a boon for diagnosing COVID-19 through artificial intelligence (AI) and machine learning (ML) techniques. Medical industries are engaged in the diagnosis and various AI applications that have been implemented for screening and providing decision support for clinicians, for diagnosing the disease with ease.
COVID-19 has the possibility to influence the retinal microvasculature. It can be recognized by taking the fundus images which includes various morphological changes present in the retina. These morphological changes can be distinguished either by tedious manual examination or by computer-aided diagnosis (CAD) that can assist the ophthalmologist with recognizing the issue. The flowchart of the proposed methodology is represented in [Figure 2].
- Data acquisition: data acquisition of COVID-19 patients can also be done by wearing protective shields along with necessary equipment for checking the retinal images
- Region investigation of the existing approach: region investigation assumes a significant part in deciding any kind of sickness. The locale can likewise be indicated utilizing different highlights, for example, optic disc, blood vasculature, and so forth
- Image preprocessing (IP): IP is a significant factor which influences the image quality. IP incorporates the handling of the picture by eliminating commotion, shading change for the detection of COVID-19
- Image segmentation: segmentation of the region can be done by focusing on specific part of the region which shows significant changes in the retina. The changes that can be seen due to COVID-19 are neovascularization, presence of microaneurysms, hemorrhages, etc.
- Feature extraction: feature extraction is an evaluative advance as the classifier will perceive yield from the information include design
- Classification: the classification and grading can be done based on the presence of changes in retinal images
- Performance evaluation: Assessment of the infection can be anticipated by different boundaries, for example, receiving operator curve, area under curve, sensitivity, specificity, and accuracy.
Hence, AI and ML through CAD are the recent advancements which should be considered for ophthalmic diagnostics.
Financial support and sponsorship
Nil.
Conflicts of interest
There are no conflicts of interest.
 References | |  |
| 1. | Seah I, Agrawal R. Can the coronavirus disease 2019 (COVID-19) affect the eyes? A review of coronaviruses and ocular implications in humans and animals. Ocul Immunol Inflamm 2020;28:391-5.  |
| 2. | Bertoli F, Veritti D, Danese C, Samassa F, Sarao V, Rassu N, et al. Ocular findings in COVID-19 patients: A review of direct manifestations and indirect effects on the eye. J Ophthalmol 2020;2020;4827304. |
| 3. | Invernizzi A, Torre A, Parrulli S, Zicarelli F, Schiuma M, Colombo V, et al. Retinal findings in patients with COVID-19: Results from the SERPICO-19 study. EClinicalMedicine 2020;27:100550. |
| 4. | Dyer O. Covid-19: India sees record deaths as “black fungus” spreads fear. Br Med J 2021;22:53. |
[Figure 1], [Figure 2]
|
