Erica Seccombe

There was a news article circulating this week about Ron Sveden of Massachusetts who thought he had lung cancer but luckily for him the growth turned out not to be a tumour but a pea sprouting; proving that those hardy crop seeds will grow in the most extreme conditions!

Thanks to the team, I now have the first image and the last image of the seed sprouting sequence. So here you can see what the two end captures look like if they their transfer functions are overlayed.

The first and last 3D volumes of the seed spouting sequence rendered in Drishti. The third image on the right is the two volumes overlaid as transfer functions.

Tim selected and set the base volume so that each new frame can be positioned in a template. Later I watched Ajay manipulate the transfer functions in order to isolate the beans from the bed of agar. He made some suggestions and showed me how he would do it by using the crop tools in combination with the transfer function. He does it expertly and I follow his methods on my own from scratch, taking me a full day to correctly capture one bean compared to Ajay’s 15 minutes.

I met two Palaeontologists in Vizlab this week, Dr Gavin Young and Alice Clement who are working with Tim on the XCT capturing the most amazing volume data of a lung fish, comparing aspects of the nervous system between live species and ancient fossils. They had come in for a lesson on Drishti with Tim, and we shared a moment commiserating on our bumbling efforts in comparison to the masterly skills of Ajay. One of the difficulties experienced by infrequent users (or even frequent users) is that Ajay constantly improves Drishti’s tools, renovating the user facilities.  So is important to be able to understand what the updates are and how they work for the data at hand. I have learnt to just persevere while referring to Ajay’s handy online instructions; then it is possible to succeed no matter how clumsily. Drishti is freeware and is now available to down load to PC or Mac so researchers are not limited to working at Vizlab, but there I have the added advantage being a stone’s throw away from Ajay.

a bean sprout - imaged rendered in Drishti.

I think it is amazing that we are looking inside a tiny bean. We predict that the difficulty we will encounter with the final bean sequence, when we proceed with animating the full dynamic process of sprouting, is that every volume in the sequence has a unique transfer function. For the moment it is not such an issue with only a few volumes but it will begin to cause difficulties with more than 5. We may have to rethink how they are linked together. However for the moment its giving me an excellent opportunity to hone my Drishti skills. To get a cleaner image I isolate one bean and clean it up as much as possible so the interior is visible through the fine outer skin. I have rendered an image (above) of the bean and its sprout overlayed, the two volumes differentiated by the colour yellow and pink. And below are two screen shots to compare the transfer functions of the first and last volumes before they merge.

screen shot of first bean in Drishti illustrating the transfer function as a histogram

If you take note of the white shadows in the histogram you can see how I’ve selected the space in between. This is the thin corridor of material between the agar jelly and the bean’s starchy interior. It allows me to examine the transparent outer shell while viewing internal cavities.

A screen shot capturing the end sprout volume in Drishti.

In this screen shot you can observe the comparative transfer function. It is different because the fragile sprout is not contrasted enough with the agar. You can see how the end of the sprout is a bit messy from being embedded with the base.

Finally the mungs make their debut on the XCT. They were ready, soaking over night; two mungs and one little radish for good luck. Amazingly they all sprout and don’t die halfway through the sequence – nine hours of 64 captures in 360°. Will the data be any good though? The acquisition of data is now in the hands of Dr Andrew Kingston to process the information for each shot. One data set takes a couple of hours to compile manually and the process is about to be automated in order to get through the enormous amount of work that the facility is generating.

3D tomograph of beans in Drishti import.

But now for a sneak preview. This is what the volume looks when it is first imported into Drishti to create a subvolume. The import main window is divided into two– a histogram window and an image window. The histogram window is used for display of 1D histogram of the loaded image. Having to understand 3D data as a histogram has been a mind bending experience but I’ve managed to plod along. The image window displays the currently selected slice of volume, like any tomographic image, but you can run through from top to bottom by using the left slider. Sizing the parameters of the volume cuts out most of the junk around the image, reducing the size of the volume.

First volume of beans in Drishti

As soon as a subvolume is saved I can drag it across to render. This is the full dataset, the beans in their container on a bed of agar jelly. In Drishti one would normally reduce the volume first by cropping in the low-res mode, but I’ve prepared an animation demonstrating the use of clipping planes to reduce the volume in high res. The process of cropping and clipping is usual for any 3D data set. It is about refining the image and working out which bit of the data is the important information. For instance, if you were examining closely the lung of a rat, you wouldn’t necessarily need its tail. This is what makes Drishti important as an volume exploration tool as it has the function to identify and select the data according to its material density ie; skin, flesh, fat, tissue, muscle, veins, nerves, bone, marrow…

the first bean

So what we have come to realise by looking at this first datset is that the agar jelly is too close in density to the starch based bean. This makes it difficult to use the transfer function to isolate the bean. It was easier with the plastic objects, the data was so straight forward. All I had to do was crop the stems they were attached to. The bean is different, a bit messy, no clean lines and so the object will not be whole in the way that I have imagined it. Drew in Vizlab explains that this is typical of working with living or organic matter, it is never how you want it to be and is always messy. Bones, rock and plastic are easier to render. But for now, this is all I have so I practice stripping of all the spongy muck down to the bare, transparent skin. There is a delicate shape to the bean, and a glimmer inside of the tiny shoot beginning to develop. I like it. I can’t wait to play with the rest of the sequence.

Tim suggests that in the meantime we work out another more contrasting base that will give the bean sustenance, but will not be too dense. I do a web search on such possible materials hoping that I’ll stumble across some new plant growing technology. I spend a bit of time reading about hydroponic systems and pages posted by bromeliad and orchid fanatics. Tim is an expert on all of this anyway as he has maintained his childhood passion for carnivorous plants; two enormous baskets of twisting vines hang from the ceiling of his office resplendent with snappy blooms. Tim decides to foam up the agar so it sets with air. Adding air to agar jelly makes me think of whisks, this makes me think of marshmallow which leads to Pavalova. Its not such a silly idea apparently. Perhaps whipped egg white with agar might do the trick, as long as it retains a marshmallow texture, the tiny bubbles of eggwhite might contrast with the starchy bean.

whisking egg whites in the wet lab

preparing a contrasting base for beans

I xxx
I am fond of making the dessert Pavalova and it has been an interest between my friend Rowanne and I for some time now; we have even considered blogging (Pav lovs blog) our efforts for every time we create one the result is unexpected – hit & miss. So for me whipping up eggwhite in the wet lab is seriously amusing. Lab coat, eye protection, glass test tube etc.  I get to measure out the agar with nice little measuring spoons but I bring in my own bar mix. There are magnetic mixers in the lab but it’s easier with the electric whisk. I make a few bases in Petri dishes and put them in the fridge for later. I put a few more beans on the soak.

How do you make a mung bean remain still and at the same time keep it from drying out? The difficulties sprouting a mung bean for 3D tomography is that they flip over when they turn their sprout down. Mung beans are a hardy variety and they move quickly once their sprout gets going. Because of the rotating bed of the XCT, the object has to be secured on the same axis to remain in focus all 360 degrees. To achieve the clearest image, the base material also needs to contrast from the bean. Water is not an option as surprisingly it is too dense, so Tim proposes to try first a bed of agar jelly. We go to the wet lab to mix it up.

There is something crazy about the wet lab in Applied Maths; not the extreme sterile environment you would expect,  but it has a fantastic experimental atmosphere of tinkering around among countless glass vials. I made a documentation of it in 2006 to try and capture the essence of it and it certainly hasn’t changed much since then.

fluoroscein c=o, 2006,

Recently I viewed the work of German artist, John Bock’s film Fischgrätenmelkstand kippt ins Höhlengleichnis Refugium, (2008 video) at the MCAs, The Beauty of distance: songs of survival in a precarious age, at the 17th Biennale of Sydney. (I am yet to find an official English translation of this title, but my step son fed it into a translation tool with truly hilarious results.)  This work reminded me of the wet lab possibly because I’m viewing it from an artist’s or outsider’s point of view; I’m quite sure the Department does not view their laboratory in quite the same way!

Bock’s absurd and humorous film features an 18^th century wigged and powdered inventor and his beautiful but trapped accomplice. Highlighting the complexity of language, narrative, history, performance and sculpture in the pursuit of scientific endeavor, the film is a satire evoking Descarte’s treatise Discourse on the Method of Rightly Conducting One’s Reason and of Seeking Truth in the Sciences (1637). I’ve found the most current review with images of Bock’s work  at de de ce blog.

In terms of my project I found another work at the MCA, Victoria, (2008 video) very relevant and moving. By the Finnish artist Salla Tykkä, Victoria is a ten-minute timelapse on the life cycle of a giant lily as it unfurls its petals in the dark, changing colour from pure white to a red hue as the morning light arrives. This species has very large leaves that can span up to three meters in diameter and the flowers can open up to forty centimeters wide on seven to eight meter stalks. Coincidently this summer I had visited the Adelaide Botanic Gardens and discovered for the first time the Amazon Waterlily Pavilion which is used to care for and propogate the Victorian Amazonica which was introduced into the gardens in 1868. The genus was named after Queen Victoria by John Lindley in October 1837, when the first known specimen was returned from a exploration of the South America’s Amazon (formerly British Guiana). Sadly the lily was not in flower during my visit, however, in the late 1800s the very first bloom of a Victorian Amazonica caused an enormous sensation in Adelaide. So for me, Tykkä’s video incorporated so eloquently all that has passed in terms of the history of observation of the natural world, including the beauty and the wonder.

I quite like this review, The Beauty of Distance in Kant’s aesthetics by Daniel Wilson, especially in consideration of Salla Tykkä’s companion film Airs Above the Ground, which was projected next to Victoria. Both films examine the colour white and the human fascination with beauty and the desire to capture and control nature.

Anyway, Tim and I mix up the agar and put it in the fridge to wait over night. I put a few beans on to soak to be ready; we have joined the queue. Getting on the XCT is the tricky part, the facility is in great demand. Meanwhile I spend time observing and learning the process of Micro 3D X-ray.

Dr Michael Turner preparing samples for the XCT

Dr Michael Turner is in there mostly and he tolerates my presence. We discovered we are in fact distant cousins, so he is obliged to put up with me, but Michael is really very inclusive and I’ve come to understand more about the process by sitting next to him and just watching. There is a lot of adjusting. When I began my initial residency the XCT was the first and only machine in the facility. Since then the team has advance the program so they have a bit more external control – before they had to walk in and out and re-program every time. Now they have included a new XCT with a helical base, this means that they can capture more information and larger sections of data. For example, a length of bone or rock, rather than a tiny sample.  But this second XCT is not yet fully operational. They are also working on a third room and it already has its lead lined walls installed and painted.

I felt very anxious about starting again in the Department of Applied Maths. Last time it seemed so easy; those plastic objects were perfect. Back then I was more worried how I was going to make work with 3D data, that was the challenge. But what to X-ray now? The team are looking at dynamic systems and that means rock and bone. I was really worried that Tim might think my sprouting seed idea rather simplistic. The Department meets for morning coffee and welcome me back. Naturally I am asked what I’m planning to do this time round. I felt so self-conscious I waited for everyone to go back to work before I proposed the idea to Tim; will 3D X-rays of sprouting seeds work? I don’t know why I was worried. Tim had of course thought about it before and can’t wait to give it a go.

My original idea was to X-ray Australian native seeds as they germinate – as they have curious structures I thought they would be visually interesting – but to do this I would have to become an expert on how they sprout. For the moment we need something immediate. Mung beans are the obvious choice; they are so willing to grow they will sprout happily in space. Infact, when I look them up there are a some sites proffering school experiments on how to sprout mung beans rotating on record tables to simulate anti-gravity. Without gravity, where do mungs direct their sprouts? In space, do they know what is up or down?

So first things first. I go to Dickson to the health food shop in search of seeds. A nice girl in a loose cotton shift helps me choose a box of organic bio-snacky mung beans. She looks pleased and makes a comment about my healthy choice. I’m not sure if I can tell her and I feel strangely guilty when I hand over the cash; its not like I’m about to torture mice…

But how do mung beans grow? There are excellent sources available for timelapse photography and this example was filmed by the science technicians of Reigate college in the UK over aprox 10 days.

Mung bean germination – HD timelaspse

I start growing a few in a bowl on the kitchen sink at home. In theory I know what the seeds do but now I have to be a bit more observant. They expand considerably in the first 12 hours of soaking. I rinse the water and keep them moist. I have to be out all day so when I come home they have expanded some more. The next morning they have split their green jackets and I can see a tiny protrusion. hmm, I have to say it does look a little bit like a penis. It only takes a few days and all of the mung beans have three to four centimetre sprouts. I wonder what they will look like X-rayed.