ARTIFICIAL INTELLIGENCE FOR GLAUCOMA SPECIALISTS

Artificial intelligence (AI) is the simulation or replication of human intelligence processes by machines, especially computer systems. These processes are basically cognitive functions seen in humans and animals, such as: learning (the acquisition of information and rules for using the information), reasoning (using rules to reach approximate or definite conclusions) and self-correction. 

Simply put, AI is the intelligence which is shown by machines, unlike the “natural intelligence” shown by humans/animals. The aim of AI is to make the machines/computers reach such a level of intelligence that they can perform almost all human functions. AI was founded on the premise that human intelligence “can be so precisely described that a machine can be made to simulate it.”

A scene from "The Terminator" movie franchise

AI was introduced in 1956, but was followed by some long periods during which no significant contributions were made (known as “AI winters”). Prior to that, in 1936 a British mathematician Alan Turing presented his paper entitled: “On Computable Numbers, with an Application to the Entscheidungsproblem” at the London Mathematical Society. According to the Church-Turing Hypothesis, symbols as simple as 0 and 1 could simulate any conceivable act of mathematical deduction. In other words, according to this hypothesis, digital computers can simulate any process of formal reasoning. Turing went on to comment:”If a human could not distinguish between responses from a machine and a human, the machine could be considered ‘intelligent’”. This was a historical development in the establishment of AI.

"The Imitation Game", a hard-hitting movie on Alan Turing

A typical AI perceives its environment and takes actions that maximize its chance of successfully and efficiently achieving the goals set for it. Such AIs are dependent on the use of “algorithms”. An algorithm is a set of unambiguous instructions that a computer can comprehend and execute. Algorithms can be “simple” or “complex”; the latter built on top of simpler algorithms. AIs can themselves be “weak”, when they perform straightforward tasks like retrieving information or acquiring images; or “strong”, capable of all and any cognitive functions that a human may have, and is in essence no different than a real human mind. 

A machine can be presented with vast amounts of data (“Big data”). However, machines have to work intelligently by using only useful data out of what is presented to it. Many algorithms are capable of extrapolating from data. They take in the available data and parsing it, evaluating it and comparing different data pieces come up with results. Parsing is the analysis of a string or text into logical syntactic components. Algorithms allow the machine to learn from its operations. Algorithms can enhance themselves by learning new heuristics (strategies or rules of thumb, which have worked well in the past) or can themselves write other algorithms. Simply put, an algorithm can respond automatically by following from results of past experiences available to it (the rule of thumb) or as the machine runs, it works on sets of data which are new for it and reaches conclusions from scratch. Every new experience is used to improve the performance.

“Machine Learning” is a subfield of AI that “gives computers the ability to learn without being explicitly programmed”. In machine learning the computer is designed to optimize a performance criterion using data from past experience. “Data mining” algorithms look for characteristic patterns in the information available to them. “Machine Learning” does the same thing but improves upon this ability: the program modifies its behavior based on its learning experience.

Relationship of AI and other subfields

When the machine has to be trained, data is fed into it. This is called the “training set”. This is the initial baseline data which is used to compare with subsequently presented data sets. If the machine has to be trained to, say identify certain images, the machine is trained to look for certain markers or properties which are called “labels”. The machine can be fed a label to identify an image (for example, a picture of the optic nerve head is labeled “ONH” and fed to the machine so that the machine can identify the structure by looking at the label, i.e. “ONH”. On the other hand if labels are not available, the machine can be provided with some structural landmarks to identify an optic nerve head and when the machine is presented with an image, the machine assesses the image for those landmarks which identify it as optic nerve head and then concludes that the image is indeed of the optic nerve head or not). 

Artificial Neural Networks (ANN or simply “neural nets”) function like machine learning algorithms with biological models applied to them. ANNs are defined as a software setup that seeks to imitate the behavior of the human brain through the use of layers of artificial neurons, which are digital constructs with weighted inputs, activation functions and outputs. We shall revisit this concept in a subsequent post on “Tools in AI”.

AI systems are already available or in development for the detection of various ophthalmic conditions such as: diabetic retinopathy, age-related macular degeneration, and glaucoma. In subsequent posts THE GLOG will look into the tools used in AI and the application of AI in glaucoma.

AUTOMATED GONIOSCOPY

GONIOSCOPE GS-1

• The Gonioscope GS-1, being introduced by Nidek, is an automated gonioscopy system with 360⁰ color imaging.

• The GS-1 incorporates a multi-mirror prism lens. 

• It has a white LED light for illumination.
• 16 surfaces present in the system are able to visualize the entire 360⁰ in a single image.
• A full capture includes 272 images (17 foci in 16 areas).

• An automated “Angle Detection” system provides guidance in order to capture the anterior chamber angle irrespective of any color differences in the region.

• The GS-1 captures the images and then “stitches” them into composite linear or circular images.

• Each area is automatically captured in 17 different foci and upto 15 images in each area can be saved.
• The machine uses high resolution to detect peripheral anterior synechiae, neovascularization, pigmentation and the position of MIGS.
• The captured data is saved in an inbuilt SSD, without need for an external PC connection, enhancing safety and ease of retrieval.

• The prism lens has a slideback safety mechanism to prevent excessive pressure on the cornea.
• The lens uses a gel immersion technique to prevent injury to the corneal epithelium.
• The machine has a 9-inch, color, tiltable, touch screen monitor with intuitive operations such as pinch-and-zoom as well as swipe.

FUCHS UVEITIS SYNDROME (FUS)

Fuchs described the condition in 1906. The clinical features comprised of: mild anterior uveitis; heterochromia; cataracts; and occasional glaucoma.

A similar condition was previously described by Lawrence in 1843.

Also known as: Fuchs Heterochromic Iridocyclitis; Fuchs Heterochromic Cyclitis; or Fuchs Heterochromic Uveitis.

Usually unilateral.

Typical age of onset: 3^rd-4^th decades of life.

Male: Female ratio is same.

The International Uveitis Study Group describes FUS as: A chronic, unilateral, nongranulomatous inflammation, mainly involving the anterior uvea, insidious in onset, low grade in activity, affecting both genders equally, preponderance in those between 20-45 years old, unresponsive to corticosteroid therapy, absence of systemic disorders and generally with a good prognosis except for the development of cataract and glaucoma. 

However, 7.8-10% patients have bilateral disease.

Etiologic theories: Likely a true inflammation of immunologic origin, although no clear HLA association has been described. It is possibly related to depression of suppressor T-cell activity. Autoantibodies against corneal epithelium have been reported in almost 90% cases. The condition was also reported in a father-son duo afflicted with retinitis pigmentosa. A few cases with associated congenital Horner’s syndrome have been reported. Clinical association with Toxoplasmosis, due to chorioretinal scars and sarcoidosis has been suggested, but no clear evidence is available so far. Chronic rubella infection is also reported to be associated with FUS. Adrenergic denervation and sympathetic paralysis suggested as patho-genetic mechanisms as well. 

Uveitis is generally mild. Patients may have a single episode with a protracted course. Although, the condition may appear intermittent initially.

Characteristic fine stellate KPs are usually seen in the lower half of cornea (occasionally involve the upper half).

Iris has extensive stromal atrophy. Transillumination of the iris shows a characteristic faint, uniform translucence. (Translucency of surrounding ocular coats also reported).

Prior to the development of heterochromia, iris changes may occur in the form of smoothening of the stroma and loss of corrugations seen in a normal iris.

Heterochromia of the iris is variable and an inconsistent feature. It is more obvious in blue eyes and subtle or absent in dark irides due to equal pigmentation of the anterior and posterior layers. In some patients the anterior layer is lost and a heavily pigmented posterior layer becomes visible. This gives rise to a darker appearance of the iris (Reversed Heterochromia). In some, especially bilateral cases, heterochromia is absent. 

Iris heterochromia (Left eye)

Nodules are often present on the pupillary borders (similar to Koeppe nodules), though they may occur on the iris surface (Busacca nodules). The nodules in FUS are typically small and translucent. Small, refractile iris crystals are also typically seen (They represent Russell bodies and can occur in other uveitic conditions). The finding of unilateral nodules in dark individuals is especially useful, as the heterochromia is less apparent in these individuals.

Koeppe nodules

Iris crystals

The anterior chamber angle is open and characteristically free of synechiae. Fine, branching, unsheathed, meandering vessels are present in the angle, often extending upto the trabecular meshwork. These vessels often give rise to filiform hemorrhages. These fine vessels may bleed spontaneously or during Tonometry, gonioscopy, mydriasis or surgery. These hemorrhages/bleeds are not associated with fibrosis or secondary angle closure glaucoma.

Vessels in the angle

Anterior segment neovascularization may occur in the absence of retinal ischemia. This is also possible in chronic uveitic conditions and iris melanoma.

Vitreous cells are common. They can be dust-like to stringy or membranous veils.

Cataracts are quite frequent to develop during the course of the disease. They start as posterior subcapsular cataracts and rapidly progress to mature or hypermature stages. Surgery for cataracts may require synechialysis and complicated with postoperative marked anterior uveitis, corneal edema, hyphema, raised IOP and cystoid macular edema. Patients should be covered with peri-operative corticosteroids. Visual prognosis is usually good. 

Raised intra-ocular pressure (IOP) in FUS is not as common as in PSS. It may occur as a serious complication in late stages. The mechanism for the elevated IOP is assumed to be increased outflow resistance at the level of the trabecular meshwork. The elevated IOP typically persists after resolution of the uveitis.

IOP does not respond well to corticosteroids. Occasional and short term therapy can be used to control symptomatic exacerbations. Long term steroids are ineffective and promote glaucoma/cataract formation. 

A steroid trial may also help to differentiate FUS from other conditions, such as Posner Schlossman Syndrome and other acute uveitic conditions with trabeculitis, which respond well to steroids.

IOP control can be attempted medically, failing which surgical management is required. Laser trabeculoplasty should be avoided due to the inflammatory nature of the condition.

POSNER SCHLOSSMAN SYNDROME

INTRODUCTION: 

It is also known as: “Glaucomato-cyclitic Crisis”

It is an uncommon type of open-angle glaucoma.

First described by Posner and Schlossman in 1948.

It is characterized by: Recurrent episodes of mild, non-granulomatous anterior uveitis and markedly elevated intra-ocular pressure (IOP).

It is usually a self-limited condition.

ETIOLOGY:

It has been associated with a number of systemic disorders including:

•       Gastrointestinal (Especially peptic ulcer).
•      Allergic disorders (HLA-Bw54 was discovered in 41% patients in a study, suggesting a role for immunogenetic factors).
•    Infection (Herpes simplex DNA was found in aqueous specimens of 3 patients with PSS; evidence of Cytomegalovirus [CMV] infection as the inciting agent has also been reported).

EPIDEMIOLOGY:

Usually seen in young-middle aged individuals, 20-50 years of age (though may occur in elderly also).

Commoner in males.

SYMPTOMS:

Slight ocular discomfort or pain.

Blurred vision.

Colored halos. (Halos last several hours to a few weeks or even longer. Usually recur on monthly or yearly basis).

SIGNS:

Mild ciliary flush.

Corneal epithelial edema.

Small to mid size, non-pigmented keratic precipitates in central and inferior cornea.

Trace flare and cells.

Pupillary constriction.

Hypochromia of iris (Reported in upto 40% of patients; this may cause confusion with Fuch’s Heterochromic Iridocyclitis [FHI]). 

IRIS FLUORESCEIN ANGIOGRAPHY:

Early segmental iris ischemia with late congestion and leakage has been described.

GONIOSCOPY:

Normal, open angles with occasional debris and characteristic absence of synechiae.

TONOMETRY:

IOP elevation may precede or follow the anterior chamber reaction.

IOP is out of proportion to the inflammatory process. Usually ranges from 40-60 mmHg. It coincides with the appearance and duration of the inflammatory process.

IOP and facility of outflow return to normal between the attacks. Severe cases with optic nerve and visual field (VF) changes have been reported.

PATHOGENESIS:

Glaucoma is attributable to inflammatory changes in the trabecular meshwork. Histology of the trabecular meshwork during the attack revealed numerous mononuclear cells in the tissue.

Other theories of mechanism include increased aqueous production due to elevated levels of prostaglandins in the aqueous. There is also an association with chronic open angle glaucoma.

DIAGNOSTIC EVALUATION:

Anterior chamber tap for viral PCR and analysis of specific antibody production (for CMV and herpes simplex).

VF testing / RNFL analysis for glaucomatous damge.

Ancillary tests to exclude other causes of unilateral uveitis and secondary glaucoma.

DIFFERENTIAL DIAGNOSIS:

Atypical cases of Fuchs Hetrochromic Iridocyclitis

Non specific hypertensive Iridocyclitis

Herpetic anterior uveitis

Acute angle closure glaucoma

CMV-induced anterior uveitis

Sarcoidosis

Tuberculosis

Multiple sclerosis

TREATMENT:

Aim of treatment is controlling the inflammation and raised IOP.

Topical corticosteroids are usually effective for the uveitic component.

In confirmed CMV infection, specific treatment can be initiated and given for 2-3 months.

IOP can be controlled with aqueous suppressants. Apraclonidine was found to be particularly effective. The efficacy of PG-analogues has not been well established.

Prophylactic therapy in between the attacks is not recommended.

Glaucoma filtering surgery may be required in medically resistant cases.

PROGNOSIS:

Recurrences decrease with age.

Visual prognosis is usually good.

Chronically high IOP with glaucomatous optic nerve degeneration has been reported in 25% cases in a study.

Patients with 10 years or more of PSS have a 2.8 times higher risk of developing glaucomatous optic nerve and VF damage, when compared to patients with less than 10 years of the disease.

PSS has also been linked to Non-Arteritic Anterior Ischemic Optic Neuropathy (NA-AION) during acute elevation of IOP in patients with small cups.