A New Tool in the War Against Cataract?

Cataracts are widely cited in medical literature as the leading cause of reversible visual impairment,¹ a statement that fails to articulate the devastation and damage this condition can have on patients. 

Those diagnosed with cataracts are more likely to develop dementia, more likely to die younger, and more likely to suffer depression. It is by all accounts an immiserating tunnel of darkness. More positively, it doesn’t have to be this way: Cataract surgery is one of the most cost-effective healthcare interventions that effects improvement physically, as well as psychologically.² 

But there is a further problem. Combined with aging populations, a rise in conditions like diabetes, alcohol abuse and global backlogs caused by COVID-19, demand for cataract surgery is at an all-time high. So how can artificial intelligence (AI), and the subsets of AI like machine learning (ML) and deep learning (DL) assist clinicians?

Crunching the Numbers

Machine learning has been employed to optimize biometry and intraocular lens (IOL) power calculations and to produce better refractive outcomes. It could also be a promising aid in determining accurate corneal power and prediction outcomes for post-refractive surgery patients.¹

The power of deep learning can also assist in surgical training. This employs the already existing technologies of computer-assisted corneal topography with virtual reality and AI. This powerful method of learning can be used in tandem to develop intelligent teaching systems for cataract surgical training. 

The viability of such an approach for microsurgical training has already been assessed in the neurosurgical discipline where trainee surgeons were assessed on a virtual neurosurgical procedure against experienced surgeons. This introduces an almost gamified element to learning.¹ Similarly, commercially available ophthalmological simulation-based training machines, such as Eyesi (Haag-Streit, Köniz, Switzerland), can harness AI in a similar fashion to provide a comprehensive training experience, allowing trainees to gain proficiency before actual patient exposure.¹

Superior Surgical Workflows

Machine learning can also improve surgical workflow in ways never imagined. Chunking each part of the operation into what’s known as phase classification and detection could potentially evolve into phase specific assessments. AI may be applied to predict the risk of intraoperative complications and optimize surgical workflows¹ — and these improved workflows could result in more efficient use of clinical resources and spaces, more treatments and better patient outcomes.

As noted by Gutierrez et al., the detection of different phases of cataract surgery (e.g., capsulorrhexis, cortical removal, lens insertion) can potentially translate into phase-specific assessments of surgical technical skills and enable procedure-tracking during surgery. This will allow real-time feedback and augment intraoperative decision-making.¹

Meanwhile a paper from 2020 analyzed the motivation behind AI development in ophthalmic surgery. These include: augmenting the information available to surgeons, accelerating intraoperative pathology, and recommending surgical steps.³ It seems likely as the technology continues to evolve, ophthalmologists will have a potent assistant in improving surgical outcomes. 

Three-Dimensional Surgery

Virtual reality and AI could combine not only during training but also during the procedures themselves. For example, 4K virtual reality surgery systems are being tested for cataract procedures.⁴ The surgeon wears 3D glasses which enable enhanced 3D imagery of the eye and thus, leveraging superior focus and depth of vision. This methodology results in less neck and back strain for the surgeon. 

In phacoemulsification cataract surgery, a 2022 paper⁵ noted that “computer vision approach using deep neural networks was able to pupil track, identify the surgical phase being executed, and activate surgical guidance tools … this proof-of-concept investigation suggests that a pipeline from a surgical microscope could be integrated with neural networks and computer vision tools to provide surgical guidance in real time.”

With all the technological advancements that are pushing the boundaries of ophthalmic patient care, you’d be forgiven for thinking that AI was just another tool in the toolkit — albeit one with a great degree of hype. But it’s becoming increasingly clear that AI — broadly in ophthalmology, and specifically within cataract treatment — holds promise, not just for better outcomes but also a new paradigm in clinical practice. There is light at the end of this tunnel…

References