WO2006002626A1

WO2006002626A1 - Method and an apparatus for analyzing retinal thickness data

Info

Publication number: WO2006002626A1
Application number: PCT/DK2005/000413
Authority: WO
Inventors: Lars Loumann Knudsen; Johnny Vad
Original assignee: NORDJYLLANDS AMT
Current assignee: NORDJYLLANDS AMT
Priority date: 2004-07-01
Filing date: 2005-06-21
Publication date: 2006-01-12
Anticipated expiration: 2007-01-01

Abstract

Method and apparatus for analyzing retinal thickness data comprising receiving digital data representing the thickness of a retina at a plurality of locations of the retina, organising the data in a first digital data array, where each array entry has a number for the thickness of the retinal in units of length for a spatially resolved element of the retina, determining locations on the retina with expected potential thickness abnormalities of the retina, determining a grouping model on the basis of these locations and performing a grouping of the digital data into subgroups in accordance with the grouping model, automatically analyzing each subgroup of digital data with regard to thickness variations within the subgroup, comparing this thickness variation 2with predetermined data according to prespecified criteria in order to reveal potential disease-related thickness variations of the retina, storing in a second array those potential local disease-related thickness variations, and determining a grouping model comprising a subgrouping corresponding to ring-like structures in the retina around the foveola.

Description

Method and an apparatus for analysing retinal thickness da¬ ta

FIELD OF THE INVENTION

The present invention relates to a method and an apparatus for analysing retinal thick¬ ness data from measurements.

BACKGROUND OF THE INVENTION

Vision properties of the eye can decrease due to various retinal pathologies such as glaucoma, edema or other thickness abnormalities of the retina, for example because of diabetes and senile maculopathy. Retinal diseases lead in numerous cases relatively fast to blindness if left untreated. An early detection of changes in the retina is vital for the proper treatment thereof.

For this reason, different techniques have been developed and corresponding appara- tus, for example as described in International patent applications WO 00/04820 and WO 02/19902. by Binnun assigned to TaKa Technologies Ltd.

In International patent application WO 02/07660 by Zhou et al., a method is disclosed, where measurements of the thickness of the nerve fiber layer of the retina are used to study the influence of Alzheimer disease. To ease interpretation, artificial intelligence is employed. In US patent application No. 2003/0114740 by Essock et al., a mathos and apparatus are provided to map the shape or thickness of the retinal fiber layer in order to aid the diagnosis of glaucoma. Measurement data about the thickness of the nerve fiber layer are grouped in rings around the center of the optical disc.

For diabetic patients, investigation of the thickness of the retina is a normal procedure in Western countries. In a small country like Denmark, up to a hundred thousand of patients are investigated each year. However, though a large amount of data exists, no useful statistical evaluation has yet been performed. This implies that comparative evaluation is difficult and may contain biased uncertainties. DESCRIPTION / SUMMARY OF THE INVENTION

It is therefore the purpose of the invention to provide a method and apparatus for intel- ligent automated evaluation of such retinal thickness data in accordance with prede¬ fined criteria, where these criteria are based on medical insight of possible retinal ab¬ normalities.

This purpose is achieved by an apparatus for analysing retinal thickness data compris- ing

- means for receiving digital data representing the thickness of a retina at a plurality of locations on the retina,

- means for organising the data in a first digital data array, where each array entry has a number for the thickness of the retina in units of length for a spatially resolved ele- ment of the retina or where each array entry has a number representing the thickness of the retina for a spatially resolved element of the retina,

- means for determining locations on the retina with expected potential thickness ab¬ normalities of the retina,

- means for identifying in the first array the entries representing these locations, - means for receiving instructions for determining a grouping model and performing a grouping of the digital data into subgroups in relation to these locations and in accor¬ dance with the grouping model,

- means for automatically analysing each subgroup of digital data with regard to thickness variations within the subgroup, - means for comparing this thickness variation with predetermined data according to pre-specified criteria in order to reveal potential disease-related thickness variations of the retina,

- means for storing in a second array those potential local disease-related thickness variations, wherein the apparatus is configured to use a grouping model comprising a subgroup- ing corresponding to ring-like structures in the retina around the foveola.

In contrast to prior art, where the thickness of the nerve fiber layer is investigated in groupings symmetrically around the optical nerve, the invention follows a completely different approach. According to the invention, the thickness of the retina, and not only the nerve fiber layer, is measured in order to reveal early retina abnormalities such as observed in association with diabetes. In addition, a grouping of the measured data is performed symmetrically around the foveola. Why this is advantageous is ex- plained in the following. - - -

The central part foveola of the macula is rather thin as compared to the region around the foveola. Thus, the region around foveola resembles a crater-like structure with a thin centre and a high ring around the foveola before the thickness decreases again with distance from the foveola. The region around the foveola is subject to thickness changes in dependence of illnesses. However, the thickness changes are only moderate even for severe changes in the function of the macula why precise measurements and evaluations are essential. Especially a grouping in rings around the foveola has proven to reveal early changes very precisely

In contrast, a grouping of thickness data in rings around the optical nerve, similar to the method as disclosed in US 2003/0114740, does not reveal such early changes to the same degree, because the moderate changes only appear as slight thickness varia¬ tions having a random character. This is due to the fact that no circular symmetry is found around the optical nerve for thickness variations due to abnormal changes of the retina.

In a further embodiment, the apparatus has a display for displaying the first array as a first table, where each entry in the first table is a number for the thickness of the retina in units of length for a spatially resolved element of the retina, the entries being organ¬ ised spatially in the first table for reflecting an area of the retina. In this case, the above grouping model when represented in the first table comprises a ring-like struc¬ ture around a defined centre entry in the first table, the centre entry representing the foveola.

In a further embodiment, the apparatus is configured to perform the grouping into sub¬ groups in accordance with the grouping model by reading for each subgroup on the table a signal from a pointing action with a pointer on a number of entries for specify¬ ing the correspondence to the specific subgroup. In a further embodiment, the apparatus is configured to read a number for an entry in the table, where the number links the entry to a specific subgroup. - -

The grouping of the data by manual pointing may be done in correspondence with physiological understanding of the retina and with medical insight of the problem to be studied. However, an automatic evaluation in a computer is useful for especially large number of measurements. Therefore, in a further embodiment, the apparatus comprises a computer that is programmed to automatically perform the grouping in accordance with the grouping model.

In a further embodiment, the apparatus is configured for displaying and indicating the subgroups on a first table.

In a further embodiment, the apparatus is configured to analyse each subgroup of digi¬ tal data with regard to thickness variations involving the ratio between

- the maximum thickness and the minimum thickness of the retina within the sub¬ group, - the maximum thickness but one and the minimum thickness but one of the retina within the subgroup,

- an average of a few of the highest numbers and the average of a few of the lowest numbers within a subgroup.

In a further embodiment, the apparatus further comprises means for receiving selec¬ tion criteria and means for obtaining predetermined data by selecting a group of per¬ sons according to the selection criteria, means for retrieving measurement data of measurements of spatially resolved retina thickness of this group, and means for de¬ termining spatially resolved average values for the thickness of a retina typical for this group.

In a further embodiment, the apparatus is configured to obtain predetermined data as measured data from a different measurement of the same retina in a patient treatment sequence. The invention also foresees method for analysing retinal thickness data comprising

- receiving digital data representing the thickness of a retina at a plurality of locations on-the retina, - • - _ . . .

- organising the data in a first digital data array, where each array entry has a number for the thickness of the retina in units of length for a spatially resolved element of the retina,

- determining locations on the retina with expected potential thickness abnormalities of the retina,

- determining a grouping model on the basis of these locations and performing a grouping of the digital data into subgroups in accordance with the grouping model,

- automatically analysing each subgroup of digital data with regard to thickness varia¬ tions within the subgroup, - comparing this thickness variation with predetermined data according to pre- specified criteria in order to reveal potential disease-related thickness variations of the retina,

- and storing in a second array those potential local disease-related thickness varia¬ tions.

In a further embodiment, the method comprises displaying the first array as a first ta¬ ble on a display, where each entry in the first table is a number for the thickness of the retina in units of length for a spatially resolved element of the retina, the entries being organised spatially in the first table for reflecting an area of the retina.

In a further embodiment, the method comprises performing the grouping into sub¬ groups in accordance with the grouping model by indicating each subgroup on the table by a pointing action with a pointer on a number of entries and specifying the correspondence to the specific subgroup.

In a further embodiment, the specifying of the correspondence for an entry is per¬ formed by entering a number of the subgroup in the entry. In a further embodiment, wherein the grouping is performed automatically by a com¬ puter in accordance with the grouping model.

In a further embodiment, the method comprises indicating the subgroups on the first table. - - -

In a further embodiment, the method comprises deteπnining a grouping model which when represented in the first table comprises a ring structure around a defined centre entry in the first table, the centre entry representing the centre of the macular region, the foveola..

In a further embodiment, the method comprises deteraiining a grouping model which when represented in the first table comprises an arcuated structure with curves roughly resembling the nerve structure in the retina between the macular region and the optic nerve head.

In a further embodiment, the method comprises analysing each subgroup of digital data with regard to thickness variations involving the ratio between the maximum thickness and the minimum thickness of the retina within the subgroup. Alternatively, one could use the ratio between the maximum thickness but one and the minimum thickness but one of the retina within the subgroup, or the ration between an average of a few of the highest numbers and the average of a few of the lowest numbers within a sub group.

In a further embodiment, the method comprises obtaining predetermined data by se¬ lecting a group of persons, retrieving measurement data for measurements of spatially resolved retina thickness of this group, determining spatially resolved average values for the thickness of a retina typical for this group.

In a further embodiment, the method comprises obtaining predetermined data as measured data from a different measurement of the same retina in a patient treatment sequence. SHORT DESCRIPTION OF THE DRAWINGS

The invention will be explained in more detail with reference to the drawing, where

FIG. 1 illustrates two grouping models, -FIG. 2 -shows a table with extracted data from a measurement is shown for a number ofpatients,

FIG. 3 shows a sequence of data tables that reflect retina thickness changes for a sin¬ gle patient

FIG. 4 is a principle sketch of part of the retina FIG. 5 illustrates retina thickness measurements for selected groups of humans,

FIG. 6 illustrates time dependence of the retina thickness after medical treatment,

FIG. 7 illustrates correlation between thicknesses for the different ring-like sub¬ groups around the foveola .

DETAILED DESCRIPTION / PREFERRED EMBODIMENT

The invention facilitates statistical evaluation of large amounts of data in an efficient way as will be apparent in the following. For such statistical evaluation, measured data in a digitised form are imported, for example from a commercially available apparatus as distributed by Talia Technology Ltd. and to a large extend disclosed in International patent applications WO 00/04820 and WO 02/19902. Such data are typically stored in a database in a format which minimises storage space but which does not allow easy access to the data for further calculation or access to only part of the data.

As a first step, the invention foresees reading of the data and transforming the set of data into a proper format. A format that has turned out to be of great use is a data array where each entry corresponds to a position on the retina and contains a number for the measured thickness of the retina at this position in unit length, for example microme¬ ter.

Suitably, the array has the form as a table with rows and columns, where the data are sorted such that the array represents an area of the retina. Such an array is easy to ac¬ cess for subgroups in the array and for further evaluation, comparison and calculation. Especially, a number of such arrays are conveniently stored in the local memory of the computer for fast access to the data. Furthermore, such an array is intuitively easy for the programmer to handle and for the diagnostic personnel to handle. The array may be displayed as a table with rows and columns very much like the array itself, where the table represents an area of the retina.

By studying the local thickness variations in such a table, retina abnormalities may be found by far easier and more precise than by evaluation of the average thickness of the retina, because small local thickness variation only have small influence on the aver¬ age thickness of the retina. For an automatic evaluation in a computer, it has turned out that a grouping of the data is highly useful. Such a grouping may be done in corre¬ spondence with physiological understanding of the retina and with medical insight of the problem to be studied. An example of such a grouping is illustrated in FIG. 1, where each entry for simplicity has a coordinate such as (25,39) reflecting the row number of 25 and the column number of 39.

In the large table in the lower part of the illustrated display, a colour pattern is used for illustration of zones for the grouping of data into subgroups. The zones in the form of square rings are offset form the centre of the display simply because the screen only displays part of the table. The central part of the table is thus the square from coordi- nate (25, 38) to coordinate (27,41). This central zone is defined to contain the central reading spot of the retina, the foveola. Around this central rectangular zone, rings of square form with a width of two rows and two columns are used for the subgrouping in the fovea and the macula.

A sub grouping in accordance with the shown zones is advantageous due to the fol¬ lowing argument. The thickness of the retina is under normal conditions thinner at the foveola than the average thickness of the retina and gets thicker than the average to¬ wards the fovea before it gets almost constant in thickness in the macula at a greater distance from the foveola. Thus, an almost circular symmetric thickness variation is expected around the foveola. For the grouping, thickness variations in each sub group are small for healthy retinas.

In the case with local thickness variations, the average thickness for a group may dif¬ fer largely from an expected average value indicating risk for disease. A better indica- tion for abnormalities has turned out to be the ratio between the highest and the lowest thickness number in a sub group or the ratio between the average of a few of the high¬ est numbers and the average of a few of the lowest numbers.

Illustration of the zones may be combined with a photograph of the retina as an over¬ lay in the display such that thickness variation can be localised instantaneously. For the practitioner, this reveals calibration mistakes when the zones are not symmetrically around the foveola. In fact, the computerised evaluation of the data may be used for calibration. Taking into account that the foveola is thinner than the macula, a spatial calibration may be performed of the data, such that the central zone always covers the thinnest region. This ensures a proper zone placement and sub group generation.

In FIG. 2, extracted data from a measurement is shown for a number of patients. The first column shows the mean value of the thickness data the maximum and the mini- mum of the data and the number of data points used for the average value. The col¬ umns from the fifth column through column number 15 show data from subgroups with arbitrary numbers 1 through 11. For each subgroup, the average thickness is indi¬ cated together with the number of data points

Each row corresponds to the data from one patient. For the first patient in the first row, it is seen that the maximum measured value of 272 is far higher than the average and far higher than the average in the shown columns. This may after further evaluation turn out to be a local thickness increase. For the third patient with data in row 3, the thickness in the first zone/sub group is larger than the thickness of the other sub groups illustrating an abnormal thickening of the foveola. The fourth patient for which data are found in the fourth row, the retina has an overall thickness which is much larger than the other patients. The method according to the invention performed when by a computer can automatically check for such abnormalities and extract the relevant numbers in order to facilitate the scientific evaluation or the daily diagnostic work of the practitioner.

A different gouping into subgroups is shown as a small table in the upper left part of the display of FIG.1. As can be seen, the zones as ring like structures - analogous to the foregoing example — are in addition parted into four sectors such that asymmetries of thickness variation can be detected.

Once a first investigation has been performed and an impression of the thickness variations in the retina has been obtained-, the same measurement data may be further analysed with a second sub grouping model in order to obtain a thorough testing of the data in different directions.

Also, in order to improve the system, artificial intelligence like in neural network computers may be used which may result in a learning process of the system yielding a higher effectivity and reliability in the long run.

The foveola is connected with the optical nerve head by small nerves that mainly ex¬ tend in arcs mainly between the foveola or fovea and the optical nerve head, which is illustrated in principle in FIG.3. In order to study thickness variations of such neural regions, other sub groupings can be used. For example, a grouping for covering arcu¬ ate regions may be chosen which resemble those areas of the retina where such nerves are expected. The principle of such a grouping is illustrated in FIG.3 showing one zone for an arcuate sub group expected to cover a major nerve. For example, changes around the optical nerve may be evaluated in order to reveal early stages of glaucoma.

The setting up of zones in a table as in FIG. 1 in order to achieve a grouping of the data into subgroups may be made manually on a table as indicated in FIG. 1. How¬ ever, in an easy manner, this is achieved automatically by simple computer routines that part the data array into appropriate sub groups illustrated on display as a table with coloured zones for presentation, for example if additional control is desired. For this task, a number of predetermined intelligent zone patterns may be stored in a memory for different purposes.

Once data from a measurement have been organised in an array as described before, further evaluation is facilitated. For example, for a high number of measurements, corresponding arrays can be used for statistical calculations. Such statistical evaluation may be performed for pre-selected groups of patients or healthy people. For example, average thickness variations may be investigated for a large number of diabetic pa- tients and compared to a healthy control group in order to understand the development of retina related diseases for diabetic patients.

Thus, the invention is a useful tool for scientific investigation of retina diseases. For this reason, different groups of people may be exposed to retina measurements and the data subsequently statistically evaluated in order to understand tendencies of retina thickness changes in relation to different diseases. Such statistical evaluation can then later be used for diagnostic work in order to find indication for certain types of dis¬ eases at very early stages. By investigating measurements from selected groups of people, the method according to the invention may not only be used to define novel reference values and conditions dependent on diseases but also to define reference values for normally healthy states, for example during pregnancy or in stress related situations.

Apart from statistical evaluation, the simple format of the data allows an easy visuali¬ sation of changes of the thickness of the retina on a table in dependency on the medi¬ cal treatment. This is illustrated in FIG. 3 a.

The table in the lower part of the displayed screen represents an area of the retina, where each entry depicts the change of thickness of the retina between two measure¬ ments. The first measurement was taken on the date 27.05.2002, whereas the second was taken at 04.06.2002. Apart from the two encircled red regions the retina has be¬ come substantially thinner. In the period from the 04.06.2002 to the 26.06.2002, the retina has become thicker, as illustrated in FIG. 3b, such that an overall change for the complete treatment period could be diagnosed in the area of the foveola and in a lower region as illustrated in FIG. 3 c, whereas the thickness was reduced in the right and left part of the retina. Thickness variations within relatively short time may be due to dif¬ ferent kinds of treatment, for example treatment with drugs or laser treatment.

Calculations and comparisons are facilitated because of the ordering of the data in arrays, which not only makes the computer method fast but also easy to illustrate on a display, as the array has substantially the same principle format as a table on the dis¬ play. Also, the data format in tabular like arrays has the advantage that it can be imported into spread sheet programs and image analysis programs, if desired. As such lands of programs are frequently changed into new versions, it is an advantage that this part is not strictly incorporated in one single program of the invention. Though the basic fea- ^■ tares may advantageously be included in one program, further sophisticated evaluation- may be performed by other related program packages, for example in spread sheet programs with statistical evaluation routines.

The data format also facilitates the formation of catalogues for patients in a hospital. Data for each patient may be part of the patient file or simply be linked to electronic patient files and follow the patient through different systems, even through systems across country borders, making the data instantaneously available to any authorised person. The digital nature of the data permits fast transmission and quick evaluation without the need of time consuming study of images where the result is biased by the person who has to judge from a reproduced image with possibly distorted appearance due to aging of photographic or printed material.

Furthermore, the nature of the data allows easy comparison to other patient related data if contained in a patient file. Thus, measurement data for retinal thickness can be related to any other registered patient information and be used later for an overall pa¬ tient evaluation.

In FIG. 5, a retinal thickness analysis is illustrated. The lower curve shows averaged thickness data measured as rings with increasing distance from the foveola. As indi- cated, the first ring indicates a thickness of around 150 micrometer, whereas the sec¬ ond, third and fourth ring show the increase of thickness which resembles the crater- like thickness variation, where the thickness again decreases with further distance from the foveola. The other curves show thickness abnormalities for three groups of humans with increasing changes of the macula. It is remarkable that thickness changes with even severe effect for the person only implies a rather moderate thickness abnor¬ mality.

For a number of diseases, only minor areas of the macula are influenced, and the effect of a specific medical treatment does not necessarily influence the entire macula. An evaluation of local changes is therefore essential. In FIG. 6, thickness measurements for a number of rings is shown for a group of humans suffering diabetes and having received medical treatment for thickening of the macula by steroid injections behind the eye. The different diagrams represent different time lapse after the treatment. The curves show measurements before treatment and at a certain time after the treatment. It can be observed that after 7-19 days, a thickening of the outer rings is observed. After 20-55 days, 56-120 and 121-365 days, the thickness of the inner rings decreases. After 366-700 days, the thicknesses of the inner rings are comparable to the thickness before treatment.

The other rings do not show changes in thickness in dependence of the time after the treatment. This implies that the ring-like grouping of the data reveals a period where the treatment does have a negligible effect, a period where the effect is pronounced and a period after termination of the effect. However, as the effect is pronounced only in the inner rings, an average over the entire macula would not reveal this effect. This illustrates the advantage of the method according to the invention.

The retina contains 90% water, such that thickness variations can be expected to be due to changes in the stored amount of water. There are two blood vessel systems regulating the exchange of nourishing components and water in the retina - the retinal circulation for supply to the inner part of the retina apart from the foveola, and an outer choroidal circulation for supply of the outer part of the retina and the foveola. Until recently, it has not been possible to reveal which of the two vessel systems are affected by different diseases. In addition, the effect of a specific treatment has not been possible to evaluate with certainty and with respect to the vascular systems. Us¬ ing the invention as demonstrated in connection with FIG. 6, it is evident that the inner rings are changing thickness. As a conclusion, it is highly likely that it must concern the inner circulation. As a result, a better treatment can be obtained as the treatment can be specifically directed towards the vessel system that is influenced by illness.

FIG. 7 illustrates for the group of patients in FIG. 6 how the thicknesses of the differ¬ ent rings are correlated. The result reveals that the thicknesses of the inner rings are mutually correlated and the thicknesses of the outer rings are mutually correlated, but the thicknesses of the inner rings are not correlated with the thicknesses of the outer rings. This supports the significance of the inner, choroidal vessel circulation and the outer retinal circulation.

The foregoing examples illustrates the advantages when using a method and an appa- ratus according to the invention.

Claims

1. Apparatus for analysing retinal thickness data comprising

- means for organising the data in a first digital data array, where each array entry has a number for the thickness of the retina in units of length for a spatially resolved ele¬ ment of the retina or where each array entry has a number indicative of the thickness of the retina for a spatially resolved element of the retina, - means for determining locations on the retina with expected potential thickness ab¬ normalities of the retina,

- means for identifying in the first array the entries representing these locations,

- means for receiving instructions for determining a grouping model and performing a grouping of the digital data into subgroups in relation to these locations and in accor- dance with the grouping model,

- means for automatically analysing each subgroup of digital data with regard to thickness variations within the subgroup,

- means for comparing this thickness variation with predetermined data according to pre-specified criteria in order to reveal potential disease-related thickness variations of the retina,

2. Apparatus according to claim 1, wherein the apparatus has a display for displaying , the first array as a first table, where each entry in the first table is a number for the thickness of the retina in units of length for a spatially resolved element of the retina, the entries being organised spatially in the first table for reflecting an area of the ret- ina.

3. Apparatus according to claim 2, wherein the apparatus is configured to perform the grouping into subgroups in accordance with the grouping model by reading for each subgroup on the table a signal from a pointing action with a pointer on a number of entries for specifying the correspondence to the specific subgroup.

4. Apparatus according to claim 3, wherein the apparatus is configured to read a num- ber for an entry in- the table, where the number-links the entry to a specific subgroup.

5. Apparatus according to claim 1 or 2, wherein the apparatus comprises a computer that is programmed to automatically perform the grouping in accordance with the grouping model.

6. Apparatus according to claim 2, 3 or 4 or according to claim 5 when dependent on claim 2, wherein the apparatus is configured for displaying and indicating the sub¬ groups on a first table.

7. Apparatus according to any preceding claim, wherein the ring like structure in addi¬ tion is parted into a number of sectors representing further subgroups, preferably four sectors, such that asymmetries of thickness variation can be detected.

8. Apparatus according to any preceding claim, wherein the apparatus is configured to analyse each subgroup of digital data with regard to thickness variations involving the ratio between

- the maximum thickness and the minimum thickness of the retina within the sub¬ group,

- the maximum thickness but one and the minimum thickness but one of the retina within the subgroup,

9. Apparatus according to any preceding claim, further comprising means for receiv- ing selection criteria and means for obtaining predetermined data by selecting a group of persons according to the selection criteria, means for retrieving measurement data of measurements of spatially resolved retina thickness of this group, and means for determining spatially resolved average values for the thickness of a retina typical for this group.

10. Apparatus according to any preceding claim, wherein the apparatus is configured to obtain predetermined data as measured data from a different measurement of the same retina in a patient treatment sequence.

11. Method for analysing retinal thickness data comprising

- receiving digital data representing the thickness of a retina at a plurality of locations on the retina,

- automatically analysing each subgroup of digital data with regard to thickness varia¬ tions within the subgroup,

- comparing this thickness variation with predetermined data according to pre- specified criteria in order to reveal potential disease-related thickness variations of the retina,

- storing in a second array those potential local disease-related thickness variations,

- determining a grouping model comprising a subgrouping corresponding to ring-like structures in the retina around the foveola.

12. Method according to claim 11, wherein the ring like structure in addition is parted into a number of sectors representing further subgroups, preferably four sectors, such that asymmetries of thickness variation can be detected.

13. Method according to claim 12, wherein the method comprises displaying the first array as a first table on a display, where each entry in the first table is a number for the thickness of the retina in units of length for a spatially resolved element of the retina, the entries being organised spatially in the first table for reflecting an area of the ret¬ ina.

14. Method according to claim 13, wherein the method comprises performing the grouping into subgroups in accordance with the grouping model by indicating each subgroup on the table by a pointing action with a pointer on a number of entries and specifying the correspondence to the specific subgroup.

15. Method according to claim 14, wherein the specifying of the correspondence for an entry is performed by entering a number of the subgroup in the entry.

16. Method according to claim 12 or 13, wherein the grouping is performed automati- cally by a computer in accordance with the grouping model.

17. Method according to any preceding claim, wherein the method comprises indicat¬ ing the subgroups on the first table.

18. Method according to any preceding claim, wherein the method comprises deter¬ mining a grouping model which when represented in the first table comprises a ring structure around a defined centre entry in the first table, the centre entry representing the focal point of the retina.

19. Method according to any one of the claims 12-17, wherein the method comprises determining a grouping model which when represented in the first table comprises an arcuated structure with curves roughly resembling the nerve structure in the retina between the foveola and the optic nerve.

20. Method according to any preceding claim, wherein method comprises analysing each subgroup of digital data with regard to thickness variations involving the ratio between

21. Method according to any preceding claim, wherein the method comprises obtain¬ ing predetermined data by selecting a group of persons, retrieving measurement data for measurements of spatially resolved retina thickness of this group, determining spa¬ tially resolved average values for the thickness of a retina typical for this group.

22. Method according to any preceding claim, wherein the method comprises obtain¬ ing predetermined data as measured data from a different measurement of the same retina in a patient treatment sequence.