In refractive surgery, astigmatism is never just a number. Behind “−1.25 D” there is an axis, an origin, a consistency to verify against the keratometry, and a surgical choice to own. Every surgeon already runs this reasoning at every planning session: comparing axes and rereading the topography, checking the pachymetry, then setting a postoperative target. But a large part of that work stays in their head.
The support, meanwhile, is usually a spreadsheet filled in by hand. For a with-the-rule myopia, that is enough. As soon as the axes diverge, or the cornea does not account for the full cylinder, the file shows its limits. The problem is not the surgeon’s expertise: it is the tool that fails to support it.
We built a pre-LASIK astigmatism planner that embeds this reasoning into the preoperative workflow itself. This article walks through what it changes, with a public demo you can open at transtorm.ai/demos/demo-plan-lasik-astigmatism.
What the spreadsheet doesn’t tell you
The spreadsheet has one virtue: everyone knows how to use it. It also has a structural flaw: it understands nothing about what you put in it. A cell holding 80 instead of 180 for an axis is still a perfectly valid cell. A conversion formula overwritten by accident keeps displaying a plausible figure. And once several versions of the file circulate around the centre, nobody knows which one is authoritative.
| Spreadsheet-specific risk | What happens in practice |
|---|---|
| Axis entered as 80° instead of 180° | The plan is wrong and the screen displays it without complaint |
| Positive and negative cylinder mixed up | The treatment goes in the wrong direction |
| Vertex conversion to the corneal plane forgotten | The correction drifts, especially in high ametropia |
| Overwritten formula, multiple versions of the file | The reference calculation becomes untraceable |
| Scenarios compared in one’s head | The rationale for the chosen plan is never documented |
The more experienced the surgeon, the better they compensate for these weaknesses mentally. But that compensation depends on the attention of the moment, which is exactly the resource that runs out at the end of a surgical list. And the standalone calculator, open in a separate window, does not change the equation: it is one more re-entry of data, hence one more source of error, and it ends up being used only for the cases already labelled complex. Yet the cases that catch you out are rarely the ones anyone had classified as complex in advance.
A planner that follows the preoperative consultation
The planner takes the data exactly as you already have it: manifest refraction in negative-cylinder notation with the flat-meridian axis, K1 and K2 keratometry with their axes, corneal astigmatism, pachymetry, white-to-white, and the patient’s age. The demo preloads six cases representative of a real preoperative list, from a with-the-rule moderate myopia (−3.00 (−1.00 × 180°)) to a high myopic astigmatism (−7.00 (−3.75 × 15°)), by way of axis disparity, oblique corneal drift, high against-the-rule astigmatism, and lenticular astigmatism. Each case opens with a card summarising what makes it particular and what the planner contributes to it.
Below the measurements sit the surgeon’s controls: a weighting slider between the refractive target and the corneal target (60% refractive and 40% corneal by default), the vertex distance (12 mm), the optical zone (6.5 mm), and a Reset button. The plan recalculates with every adjustment. The Plan tab shows the refraction converted to the corneal plane, the ocular residual astigmatism in magnitude and axis, the planned cylinder, the target refraction, and the nomogram adjustment. At the bottom, the refraction ready to key into the laser, together with the ablation depth and the residual stromal bed in microns.
Ocular residual astigmatism, made visible
Ocular residual astigmatism, or ORA, measures the share of the refractive cylinder that the anterior cornea does not explain. On a spreadsheet, nobody computes it for every eye. Here it is calculated systematically, displayed with its magnitude and axis, and drawn on the meridian chart next to the refractive, corneal, and planned axes. The figure becomes a clinical reading of the case: you see at once what the refraction is asking for and what the cornea can actually deliver; the remainder, no laser pulse will erase.
Friday planning session. A young patient, −3.00 (−2.50 × 90°) on manifest refraction. On the spreadsheet, nothing unusual: you would treat the 2.5 D of cylinder. In the planner, the keratometry tells a different story: the cornea is nearly spherical, barely 0.5 D. ORA reads 1.76 D and the Checks tab flags the mismatch. The vector plan treats only about 1.7 D: ablating the full 2.5 D of the manifest would mean sculpting into the cornea an astigmatism it never had.
The Vectors tab unfolds the calculation. The double-angle plot places each astigmatism at twice its clinical axis, with the vector’s length giving the power in dioptres: two astigmatisms that sit close on this diagram are genuinely close, whatever the notation. The J0/J45 component table lines up the refractive, the corneal, their difference, and the chosen plan. What used to demand a dedicated worksheet and a good memory for trigonometric formulas now reads at a glance.
Built-in checks, never blocking
The Checks tab first qualifies the astigmatism with a badge (for instance “Against the rule”), then runs through the safeguards: corneal astigmatism, pachymetry, white-to-white, residual stromal bed. When the refraction and the cornea disagree, a non-blocking alert appears: high ORA above 1.0 D, treating the refraction alone can leave residual astigmatism, and the weighting slider arbitrates between the two. It informs; the decision remains surgical.
That is the whole difference with occasional calculation. Because ORA and the vectors are computed for every eye, the mismatch flags itself, including in the chart you thought was routine.
Adjusting without losing coherence
Faced with the same eye, several strategies are defensible: treat the manifest, favour the cornea, look for a vector compromise, apply your personal nomogram. The spreadsheet leaves you to compare them in your head. The planner shows the consequences of each one, live: the planned cylinder, the departure from the manifest refraction, the ablation depth, the stromal margin that remains.
Next case, a high against-the-rule astigmatism. The surgeon moves the weighting slider toward the cornea and watches the plan respond: the planned cylinder shifts, the target refraction recalculates, the ablation depth gains a few microns, the residual stromal bed stays green. Two adjustments later, he is back at the 60/40 weighting. It was the right compromise, but this time he saw it instead of assuming it.
The tool standardises what should be standardised, without touching what belongs to medical judgement.
| What the planner standardises | What the surgeon keeps |
|---|---|
| Structured encoding of the measurements | The final clinical choice |
| The conversions (corneal plane, double-angle plot) | How to weight refraction against cornea, case by case |
| Systematic calculation of ORA and J0/J45 | Their personal nomogram |
| Visualising the meridians and vectors | The medical interpretation of the case |
| The plan ready to key in, with its traceability | Sign-off and responsibility |
From isolated calculation to preoperative workflow
The goal was never to add one more window. The calculation lives inside the preoperative workflow: measurements come in structured, the result is interpreted alongside the rest of the case, and scenarios are compared on screen rather than from memory. In production, the planner runs on the same configurable engine as our other clinical tools. The nomogram and the alert thresholds adapt to your practice, and every validated plan generates a preoperative report that records the input data, the conversion, the vector analysis, the options compared, and the rationale for the plan that was chosen.
The decision logic thus becomes a usable trace. You come back to it for an enhancement or an audit, and you discuss it between surgeons in the same centre. Junior colleagues read the reasoning instead of guessing it, and the nomogram is refined on your own outcomes rather than on averages published elsewhere.
What this demo shows, and what it doesn’t
The public demo runs on demonstration data, without real authentication, and is not a medical device. It shows the clinical core, the same vector calculation as the oph4py pre-LASIK planner, in an interface available in five languages. The generated preoperative report and the customisable nomogram belong to the production version; the demo deliberately keeps to the scope of the calculation so it stays self-contained.
Try it, or talk to us
Open the demo at transtorm.ai/demos/demo-plan-lasik-astigmatism, load the lenticular astigmatism case, and move the weighting slider: you will watch the compromise shift in front of you. If you would like a planner shaped around your own practice (your nomogram, your thresholds, your preoperative workflow), book a conversation with transtorm.ai below.
Book a call