Clinical Practice

SER is the international standard formula that combines sphere and cylinder into a single number to express myopia severity:

The IMI (International Myopia Institute) 2019 white paper recommends SER as the standard for all myopia research. This enables direct comparison across studies, treatment efficacies, and populations.

In MiyoLab, every sphere and cylinder value entered is automatically converted to SER. All progression calculations, BHVI projections, and treatment efficacy analyses are based on this value.

IMI 2019 myopia classification:

• No myopia: SER > −0.50 D (pre-myopia additionally requires a validated onset-risk model)
• Mild myopia: −0.50 D ≥ SER > −3.00 D
• Moderate myopia: −3.00 D ≥ SER > −6.00 D
• High myopia: SER ≤ −6.00 D

Measurements taken while accommodation (the focusing reflex) is active in children may overestimate myopia. The gold standard is autorefraction performed after temporarily paralyzing accommodation with a cycloplegic agent (typically cyclopentolate 1%).

MiyoLab asks you to record the measurement method (cycloplegic / non-cycloplegic / subjective) on entry. Because the individual difference varies, no universal fixed offset is applied to non-cycloplegic measurements. Clinical progression analysis uses a comparable cycloplegic series, and decision support warns when methods are inconsistent.

The IMI 2021 Clinical Management Guidelines strongly recommend cycloplegic refraction for myopia diagnosis and monitoring in children.

The foundation of myopia management is accurately measuring the rate of progression. MiyoLab calculates annual progression rate (D/year) from each patient's SER measurements across multiple time points using ordinary least squares (OLS) linear regression.

Calculated metrics are the annual cycloplegic SER slope (D/year), R², standard error, and data quality. Clinical classification requires at least three comparable measurements per eye spanning one year; a low-confidence slope generates no automatic action.

The absolute cycloplegic rate is summarized descriptively. An age-reference ratio alone is not a threshold for starting or changing treatment.

Implausible progression rates (e.g., > −2.0 D/year or large positive shifts) are automatically flagged as implausible and data entry review is recommended.

BHVI (Brien Holden Vision Institute) Model

This line is not an externally validated individual prognosis. Exact age-18 values, individual confidence bands, and high-myopia probabilities are not shown in clinical summaries. Treatment lines show only direct RCT absolute SER group differences within their study horizons.

Myopia onset and course are multifactorial. Family history and lifestyle are retained separately as counselling context; no externally unvalidated 0–100 score, individual probability, or projection coefficient is produced:

• Parental myopia: family-history context associated with onset, not an individual progression coefficient
• Outdoor time: protective association especially for prevention of myopia onset (He et al. 2015)
• Screen / near work time: > 3 hours/day of near work increases risk (Huang et al. 2015)
• Current myopia level and progression rate
• Axial length and AL/CR ratio
• Onset age: Early onset (< 8 years) carries higher risk of ultimate myopia
• Ethnicity: East Asian children have 2-3× higher prevalence (IMI 2021)

These factors generate no automatic risk score, recall interval, or treatment-initiation threshold; the measured SER/AL course and examination take priority.

MiyoLab's CDS engine, based on IMI 2021 Clinical Management Guidelines, continuously monitors over 12 clinical parameters and generates recommendations at three levels:

• Info: A note to review measurement method, data sufficiency, or axial elongation in context
• Warning: Measured high myopia/long AL, anisometropia, or a finding requiring retinal assessment
• Critical: Reserved for genuinely urgent, examination-based safety findings in the relevant module

Measurement variability limits interpretation of small differences. MiyoLab's absolute information thresholds are not diagnostic or treatment gates.