Dr. Abhishek Mandal, Ph.D.

Founder, Vision Science Academy, London, United Kingdom


Vision Science Academy Exclusive

The unlimited scope of data science is expected to transform the global perception of healthcare research and its diverse applications. With an unprecedented amplification in the digital footprint of information in healthcare, the cumulative role of data scientists also continues to evolve in the medical profession.

Eye care constitutes a multidisciplinary component of modern healthcare and includes Ophthalmology, Optometry, and other allied Vision Sciences. This feature allows eye care to boast extensive exabytes of data which pertain to arithmetical values (e.g., lens power, depth of the anterior chamber, or intraocular pressure), fundoscopic images, and ultrasonographic (US) or three-dimensional computed tomographic (CT) scans of the eye (Cheng et al., 2020). Similar to other fields within healthcare, patient records in eye care have also been vastly transformed into the electronic version. The entirety of digitised information pertinent to a particular patient can now be conveniently accessed through a universal electronic medical record (EMR) system. In the niche of eye care, data scientists have successfully implemented several digital databases which can store up to millions of patient records. One perfect example is the Intelligent Research in Sight or IRIS software developed by the American Academy of Ophthalmology, which is capable of accommodating a multitude of relevant clinical statistics from more than 50 million patients simultaneously (Pershing & Lum, 2022). The analytical tools of IRIS are efficient in correlating the relative efficacy of different therapeutic modalities in real-time and in determining clinical associations between various ocular pathologies. More importantly, such comprehensive clinical registries can map out the timing and duration of various ocular treatments and hence, improve the clinical practice in Ophthalmology (Lee et al., 2021).

A detailed statistical evaluation of extensive data domains can significantly boost the overall applicability of clinical outcomes when compared to small-sized data registries. Larger datasets, collectively termed as the big data, are more effective in terms of capitalising on the effect of randomness by uncovering subtle associations which could remain overlooked when analysing smaller study samples. Moreover, biostatisticians can utilise the real-time data from the exhaustive EMR databases to underline the causative factors of an ocular pathology, and subsequently, help devise preventive strategies to curb these clinical indicators (Yang et al., 2021). In addition, data analysis of larger samples in eye care can better address the clinical gaps in the diagnosis and management of complex ocular conditions. For instance, an unanticipated discovery of novel disease markers and isolation of diagnostic features via 3D ophthalmic imaging has only been made possible through complex databases in eye care (Clark et al., 2016).

At the moment, we are on the verge of a potential breakthrough as multidimensional findings continue to stem through the applications of data sciences in eye care. The immense growth in EMR has been coupled with the development of electronic tools based on artificial intelligence (AI). Machine learning and AI programs are exceptional when it comes to the recognition of intricate data patterns invisible to the human eye. In line with this, screening systems utilising AI can scour retinal images and video recordings to accurately identify the markers of progression in retinal complications e.g., diabetic or hypertensive retinopathy, and macular disorders (Ting et al., 2019).

Keeping this in mind, it won’t be inaccurate to predict that data science will utterly revolutionise our approach in the primary or higher-level clinical settings in eye care.



Cheng, C.-Y., Soh, Z. D., Majithia, S., Thakur, S., Rim, T. H., Tham, Y. C., & Wong, T. Y. (2020). Big Data in Ophthalmology. The        Asia-Pacific Journal of Ophthalmology, 9(4), 291-298. https://doi.org/10.1097/apo.0000000000000304

Clark, A., Ng, J. Q., Morlet, N., & Semmens, J. B. (2016). Big data and ophthalmic research. Surv Ophthalmol, 61(4),
443-465. https://doi.org/10.1016/j.survophthal.2016.01.003

Lee, C. S., Brandt, J. D., & Lee, A. Y. (2021). Big Data and Artificial Intelligence in Ophthalmology: Where Are We Now?       Ophthalmol Sci, 1(2), 100036. https://doi.org/10.1016/j.xops.2021.100036

Pershing, S., & Lum, F. (2022). The American Academy of Ophthalmology IRIS Registry (Intelligent Research In Sight):       current and future state of big data analytics. Curr Opin Ophthalmol, 33(5), 394-398.       https://doi.org/10.1097/icu.0000000000000869

Ting, D. S. W., Pasquale, L. R., Peng, L., Campbell, J. P., Lee, A. Y., Raman, R., Tan, G. S. W., Schmetterer, L., Keane, P. A., &       Wong, T. Y. (2019). Artificial intelligence and deep learning in ophthalmology. British Journal of Ophthalmology, 103(2),       167-175. https://doi.org/10.1136/bjophthalmol-2018-313173

Yang, Y. C., Islam, S. U., Noor, A., Khan, S., Afsar, W., & Nazir, S. (2021). Influential Usage of Big Data and Artificial       Intelligence in Healthcare. Comput Math Methods Med, 2021, 5812499. https://doi.org/10.1155/2021/5812499