Progress Report

Requirements:

i. The holder must be designed to accommodate as wide a range of egg size as possible. This is necessary since both during the building of the holder and more importantly during its subsequent use for imaging a wide range of egg sizes will be used. During building it is anticipated that egg fragility will lead to many breakages and so many different eggs will be used for the production of the mould; for imaging it is only natural that researchers using the holder will want to image a large number of samples. It should be noted that there is a general lack of UK or EU regulation concerning the standardisation of egg size and so it will not be possible to follow a strict guideline concerning size during egg selection. The British Egg Industry Council [2] provides only four broad categories of egg size (small, medium, large, extra large) based on the mass of the eggs.

ii. The sources and detectors that are to be attached to the mould must be rigidly fixed in position and the fixed spatial coordinates of all sources and detectors must be determined prior to any imaging. This is necessary because it forms an essential requirement for the subsequent image reconstruction using the temporal optical absorption and scattering algorithm (TOAST).

iii. Provisions must be made in the design so that each detector can be optically isolated from direct illumination by sources in neighbouring detectors as well as from all external sources of light.

Proposed building process

The outer shell of the egg holder was to be constructed in two equal halves, an upper and a lower half, from low-temperature thermoplastic. The lower half was to be mounted on rods that in turn will be mounted on a flat horizontal rectangular or square base. Elevating the lower half of the holder on rods would make the holder more stable (in the absence of any support the holder would roll around on any surface it is placed) and provide access to its entire surface area. This would facilitate the placement of sources/detectors over the entire surface. The two halves were to come together and attach in such a way so that irrespective of the containing egg the spatial coordinates of the sources/detectors would remain fixed. Small sockets mounted onto the thermoplastic shell at predetermined positions were to be used to attach source/detector connectors.

Various methods were to be considered to determine the positioning and spatial coordinates of the sources/detectors required for image reconstruction. One possibility would involve a more or less random positioning of the sources/detectors and the subsequent use of a 3D digitising arm preferably while the egg is within the holder but if not possible then either before or after (since as mentioned previously, theoretically, the design of the holder will be such that the presence or not of an egg will not affect the spatial coordinates of the sources/detectors). Another possibility would involve a more orderly positioning of the sources and detectors as well as more careful construction. The maximum number of source/detector rings and the maximum number of sources/detectors that could be placed on each ring would be estimated. The sources/detectors on each ring would then be placed at equal angular separation as shown in figure 1 below.

The inner surface of the holder would be lined with soft NIR-absorbing foam of thickness to be determined. The foam would serve a dual purpose: On the one hand, since it is compressible it would provide the necessary padding to accommodate a wide range of egg sizes while the spatial coordinates of the sources/detectors remain fixed (requirement 1). On the other, its NIR absorbing properties would ensure that detectors were optically isolated (requirement 3). Holes would be made in the foam through which the connectors would be translated bringing the sources/detectors up to the egg surface.

Figure 2 shows the proposed model.

Actual building process and extent to which requirements were met

The outer shell of the holder has indeed been constructed using two sheets of thermoplastic as originally planned. It was decided to build two holders differing originally in the way in which they would be lined with NIR foam. Four pieces of thermoplastic of dimensions were cut from a larger sheet. One of the sheets was heated at its centre using a heat gun and made to take the shape of the egg by placing the egg at its centre and pulling down on the edges of the sheet until it had reached the middle of the egg as shown in figure 3:

The mould was then turned over so that its top side was now facing downwards. The bottom, still exposed side, was covered with very thin foil paper so as to prevent the two sheets from sticking together and thermally isolate them preventing deformation of either sheet when the heat gun was subsequently used to make small changes in the shape of the mould. The bottom half of the mould was formed in exactly the same way as the top half and is shown in figure 4:

Figure 5 shows a series of pictures of the mould being formed.

It has not been possible to fulfil requirement 1 to the extent that was originally thought. While it was originally believed that the holder would be able to hold small, medium and large eggs various constraints have meant that it can only hold a wide range of large eggs:

The original mould was created using a hard boiled medium sized egg by the method described above. It was originally thought that a combination of a mould formed from a medium sized egg and the compression of the subsequently added NIR foam would be suitable and sufficient for the holder to accommodate the 3 different egg sizes. However, once the mould had been formed it was found that it would accommodate precisely only the single egg from which it had been formed with the possibility of a small variation in size. It was therefore decided that the egg would have to incorporate the thickness of the foam (~1 cm) so that when the inner surface of the mould would subsequently be lined with foam it would be of the correct size to tightly fit the egg as shown in figure 6 below:

To form the “enlarged” mould an attempt was made to surround an egg with NIR foam and proceed as described above. This approach however did not work because of the small size and curved nature of the egg; The foam was available in strips of width approximately 2.5cm so covering the egg meant using a couple of strips around the central part and many small pieces for the top and bottom. As a result the egg lost its well defined shape and consequently the mould created was incorrectly shaped. Using thinner strips did not overcome the problem.

As an alternative, a papier mâché technique was considered: Fine paper was cut into long thin strips and flour and water were mixed together to form a paste like consistency. Each strip of paper was soaked in the mixture and wrapped around a large egg. This proved to be a successful way of increasing the overall size of the egg by the required ~2cm but along the way papier mâché -ing of a real hard boiled egg was replaced by papier mâché -ing of a porcelain egg that had been acquired for the following reasons:

i. this would save a lot of time given that the porcelain egg was already much closer to the required size,
ii. there would be no possibility of the porcelain egg rotting as could be the case with the real egg.

Once the papier mâchéd porcelain egg had dried it was sprayed with a fast dry enamel paint that would act as a protective layer preventing paper layers from being pulled off when the thermoplastic was to be separated from the porcelain egg.

Figure 7 show pictures of the real papier mâchéd egg and the porcelain papier mâchéd egg after it had been sprayed with enamel paint.

The porcelain egg was subsequently used to form the mould as described on pages 2 & 3. Having used an egg with dimensions analogous to a large egg surrounded with foam, means that the holder will now only accommodate large eggs. This possibly imposes a restriction on future use of the holder but it is believed that it is an acceptable compromise for two reasons:

i. The size of the egg to be imaged should not affect the image obtained and since a wide range of large eggs can be imaged the original requirement (requirement 1) that it should be possible to image many samples is satisfied.
ii. An alternative method (described below) that would theoretically overcome this restriction would be more complicated to build, require the assistance of a workshop technician, take much longer than originally planned and, in practice, very likely not work.

For the alternative method, instead of lining the inside of the mould with NIR foam, patches of foam would be attached to the end of small hollow rods that would be translated radially inwards and outwards up to and away from the egg within the holder and supported in place by small screws. The fibres would be passed through the hollow rods and brought up to the egg. This is shown in figure 8. This way the NIR foam would be in intimate contact with any egg size (requirement 1) supporting it in place as well as isolating the neighbouring fibres (requirement 3). This arrangement would also satisfy requirement 2.

The above solution would be very difficult to implement because of the very small surface area over which optodes can be placed. Even in the absence of the holding mechanisms the distance between adjacent optodes is very small and it is therefore unlikely that there would be enough room to position the holding mechanisms.

As stated on page 3, it was originally decided to build two holders differing in the way in which they would be lined with NIR foam. One would be lined using continuous strips while the other would be lined with patches as shown in figure 7 but with the patches actually glued to the inner surface of the mould. In both cases holes would be made through the foam for the light to penetrate. The former of these lining techniques was abandoned in favour of both moulds (each having a slightly different shape) being lined with NIR foam cut into patches. The advantage of this method was that the small patches could be placed at any angle producing a tighter fit around the egg and reducing the risk of any light travelling through air gaps and reaching the detectors without passing through the egg.

It was also proposed, as part of the project outline (c.f. Fig 1), that the exact positions of the optodes could possibly be determined prior to actual positioning. This proposal was also abandoned since apart from the complexity of determining the positions for such an irregularly shaped 3D object, experimental errors would lead to the actual positionings being different from those calculated. Nonetheless the holes for the optodes have been made to form rings around the mould and the maximum number of holes forming each ring have been equally distributed. By trial and error it has been found that the maximum number of holes/optodes that can be fitted is 26 in the arrangement shown in figure 9.

To make the holes and hold the optodes in position (requirement 2) 3 different methods were tried:

i. The heat gun was used to make the area surrounding the position where an optode was to be placed pliable. The plastic channel was then simply pushed through and held in position until the plastic became hard again.
ii. Holes were cut into the thermoplastic at the positions where the optodes were to be placed.
iii. A hand drill with a drill bit of 6.1mm was used to make round holes at the positions where the optodes were to be fixed. A reamer was used to slightly increase the size of the hole and push the edges inwards as shown in figure 10.

Out of the 3 methods, method (ii) was found to be the least effective with the cut holes being either too small for the plastic channels to pass through or too large. Method (i) was better than (ii) – it gave a tight fit for the optodes but as was the case when trying to form the mould, using the heat gun even on isolated areas caused the thermoplastic to deform in an undesirable way. Method (iii) was found to be the most effective – it did not deform the thermoplastic and with a choice of correct drill size and some reaming the fit for the optodes was very tight.

Figure 11 is a picture of the base and holder with NIR foam lining and an egg in position.