Dr. William Trattler sheds light on big data’s applications in the field of cataract and corneal diseases—among others
How is big data reshaping eye care? Let’s explore its impact on cataract and refractive surgeries, patient selection, tailored treatment, and the ethical considerations it brings to clinical settings.
Big data has emerged as a transformative force in ophthalmology, reshaping our understanding, diagnosis, and treatment of various corneal conditions. Dr. William Trattler, a respected refractive, corneal, and cataract surgeon from the United States, explored the current applications and promises of big data in the field of cataract and corneal diseases.
Enhancing cataract and refractive surgery outcomes
According to Dr. Trattler, big data has been instrumental in significantly enhancing outcomes in cataract surgery. “Numerous intraocular lens (IOL) calculation formulas currently available to cataract surgeons leverage big data and artificial intelligence (AI) to enhance the reliability and accuracy of IOL formulas. Notable examples include the Hill-RBF formula by Warren Hill and IOL power calculations developed by ZEISS, as well as Graham Barett’s suite of formulas. With each passing year, IOL formulas continue to improve, leading to better outcomes overall, he noted.
Besides refining intraocular lens calculations, big data has the potential to help surgeons identify the optimal techniques when performing cataract surgeries, such as crunching data on energy usage during phacoemulsification to help optimize the efficiency of our procedure, according to Dr. Trattler. “Leveraging big data can aid in predicting the most suitable type of intraocular lens for individual patients. One of the major challenges we face is that while we have a very high rate of success when patients choose to receive a range of vision IOLs, there are still a small percent of patients who are not satisfied with their outcome. The hope is that by utilizing big data insights, we can help optimize IOL selection leading to improved patient satisfaction,” he continued.
For refractive procedures, big data is being utilized in laser-assisted in situ keratomileusis (LASIK), photorefractive keratectomy (PRK), and small incision lenticule extraction (SMILE). Companies are aggregating results and employing optimization strategies to refine outcomes in these procedures.
Big data has indeed facilitated significant advancements in LASIK surgery. Dr. Trattler cited an example — Surgivision, a company with an online platform developed by eye surgeon Dr. Guy Kezirian, where practitioners can upload results from their procedures. This aggregated data enables surgeons to make more informed decisions and adjust parameters to achieve the best visual outcomes for their patients.
Data-driven decision making
Dr. Trattler also mentioned that incorporating findings from largescale corneal registries into his clinical decision-making process facilitated more informed patient assessment and treatment approaches. This ultimately improved the quality of care provided to his patients.
“I believe that studies examining data from large patient cohorts provide valuable insights into patient assessment and treatment approaches. A recent study* that caught my attention focused on screening children aged three to 18 to determine the prevalence of keratoconus. This is significant because keratoconus, despite being a treatable corneal disease, is challenging to diagnose at early stages since patients may have no visual symptoms or have any signs on slit lamp exam,” he explained. “This screening, conducted in Chicago, Illinois, revealed a significantly higher prevalence of keratoconus among children than previously estimated.”
Dr. Trattler continued: “So, as a collective effort in our community, we aim to identify patients with keratoconus at the earliest possible stage to ensure that we can treat their condition before they experience significant vision loss This type of information (provided by the study) has the potential to reshape how healthcare professionals view their patient demographics. Historically, optometrists and ophthalmologists might have waited until patients experienced reduced vision before initiating screening tests. However, this study indicates that conducting this study supports that conducting topography to screen for keratoconus, even in the absence of symptoms or visual impairment, could be beneficial,” he further explained.
When big data and AI work together
Dr. Trattler noted that the essence of AI lies in its ability to unearth solutions to unasked questions, thereby enhancing our ability to achieve better outcomes.
“We have high hopes for the role of AI in delving into big data and uncovering trends and answers to questions we may not have previously considered,” he noted. “By scrutinizing vast amounts of data, AI can identify patterns that may elude human observation, providing deeper insights and enabling us to utilize the findings more effectively. For instance, AI may help surgeons determine which patients are better candidates for EDOF vs trifocal IOLs, based on factors such as preop corneal steepness, preop corneal regularity, anterior chamber depth, axial length, gender, etc. While our current results with range of vision IOLs are quite good, AI can potentially further improve patient satisfaction,” he said.
Moreover, AI has the potential to guide patient selection in ways we may not have envisioned, offering insights into nuanced subsets within larger demographics.
“For example, it may suggest that individuals aged 65 to 85 generally fare well with certain types of range of vision lenses, whereas those aged 85 to 100 may achieve optimal outcomes with a different lens type. Such granular analyses based on big data trends hold promise for improving overall treatment outcomes,” Dr. Trattler elaborated.
Big data and private practice: Bridging the gap
While big data holds immense promise, integrating it into clinical practice presents some challenges. Dr. Trattler highlighted accessibility as a main hurdle.
“While there’s a plethora of studies and analyses available, the crucial step is disseminating this information effectively to surgeons and eye care professionals to enhance patient care. We need to devise better methods for conveying the findings from these extensive data analyses. I believe effective communication is key. With the abundance of emails and text messages we receive daily, it’s about finding a platform that enables doctors to integrate the insights from big data seamlessly into their patient management practices,” he explained.
Dr. Trattler suggested these could involve seminars, webinars, or educational sessions, with both leaders and practitioners playing pivotal roles in educating others and driving implementation.
In terms of understanding the epidemiology or treatment outcomes of corneal diseases with big data, Dr. Trattler believes that we are still in the process of making significant advancements.
“While we haven’t experienced major breakthroughs yet, progress is underway. I anticipate that as researchers analyze data and pose better questions, we’ll see significant developments. Ultimately, the key is determining the right questions to ask when utilizing big data—what information do we need to improve treatments and make a meaningful impact?” he said.
Addressing ethical considerations, Dr. Trattler emphasized that our foremost priority is to ensure that big data is de-identified to prevent bias. “Essentially, we aim to anonymize the data to prevent individuals from being targeted or personally impacted. Additionally, it’s critical to mitigate biases that could unfairly affect particular groups based on factors like ethnicity or religion. The objective is to maintain unbiased data, devoid of any potential harm or discrimination towards specific demographics,” he remarked.
Reference
* Harthan JS, Gelles JD, Block SS, et al. Prevalence of Keratoconus Based on Scheimpflug Corneal Tomography Metrics in a Pediatric Population From a Chicago-Based School Age Vision Clinic. Eye Contact Lens. 2024;50(3):121-125.
Editor’s Note: A version of this article was first published in CAKE Magazine Issue 23.
