• Optic nerve
• intraocular pressure
• diagnostic tests/investigation
• visual perception
• field of vision

We recently published a prospective study analysing the possibility of detecting progression in the early stages of glaucoma.1 The paper outlines certain advantages of functional procedures with respect to morphological procedures. However, it does not signal the end of the journey, but simply the state of the problem when the study was performed. Morphological methods of analysis are continuously progressing, as are functional methods. The ‘state of the art’ will change as new diagnostic and analytical procedures are developed.

Our experience lies in the design of algorithms and methods related to function. Studies such as the one published in the British Journal of Ophthalmology earlier this year1 have led to new ideas on how currently available methods may be improved. In this article we attempt to summarise these new ideas.

One of the conclusions drawn from the results of our work is that the progression of functional defects can easily be masked by false improvement brought about by training. It is customary to discount the first two tests of the patient to reduce the ‘learning effect’ and this we have done, but, disconcertingly, the patient's improvement in sensitivity may continue until the fifth test or even beyond.2 In our opinion it would be interesting to have a specific training method that should be applied before starting patient monitoring.

To this end we are designing a program that provides a quick and accurate estimate of the mean deviation (MD) of the patient, in line with another idea that we published 20 years ago.3 This preliminary step will constitute the first phase of a new perimetric strategy, called Spark, designed for use in Oculus perimeters (Oculus Optikgeräte GmbH, Wetzlar, Germany). The duration of this training phase is about 36 s and can be applied repeatedly to verify that sufficient stability …