“New and inventive methods have played a huge role as enablers of scientific discovery. When maturing, several of the methods go commercial, further increasing their societal impact,” said Hans Hertz of the Division of Bio-Opto-Nano Physics at the KTH Royal Institute of Technology, Sweden.
And Hertz should know. He is also co-founder and shareholder in the company Excillum which 20 years ago pioneered the liquid-metal-jet microfocus source − the brightest microfocus X-ray source worldwide.
“Research for novel methods of scientific discovery is often undervalued,” he continued. “Method’s development shapes the way scientists think. If there is no method to test something, it’s not worth having. It takes a long time for methods to show up in science – so we assume we’ve always had them. For example, Moderna and Pfizer are credited with developing COVID-19 vaccines in nine months but they were based on very long-term work by others.”
“We need to continue to investigate out of scientific interest and curiosity,” he added.
He started by explaining some of the highlights in the history of X-rays.
“The visible light range is small,” he said. “The discovery of X-rays in 1895 by German physicist Wilhelm Röntgen opened a massive new window.”
Interestingly, he pointed out that Röntgen made the discovery on 8 November 1895, the scientific paper was published in December and by 5 January the story was on the first page of a daily newspaper in Vienna followed by similar dailies in other European capitals.
“By 1900 there were X-ray boxes used for entertainment on the streets of Paris.”
In 1912 Max von Laue's experiments showed that X-rays could diffract through crystals. This led to the development of the Bragg equation by father and son Bragg that provided a mathematical framework for interpreting these diffraction patterns. This earned them the Nobel Prize in Physics in 1914 and 1915 and initiated the discipline of X-ray crystallography. Today, X-ray crystallography is fundamental for determining the structures of complex biological molecules, including proteins and DNA.
“It was an important basis for Watson and Crick’s 1953 paper which postulated that the genetic code was a double helix. We didn’t know before that how genetic material was copied. The contribution to this work of Rosalind Franklin was only recognised much later,” said Hertz.
He noted that another major breakthrough was the development of the CAT or computed axial tomography scan in 1971 by Godfrey Hounsfield which is still the most used imaging method in medicine because it is relatively inexpensive and fast.
“All of this heralded a new era of biomedicine.”
Ongoing developments
Most people know about X-rays because of their use in medicine but they are also used across a range of disciplines including esoteric topics like paleopathology which aims to diagnose the pathological conditions in ancient remains. In fact, Hertz’s first involvement with STIAS was in working with Egyptologist and fellow Salima Ikram imaging mummies.
But the challenge with X-rays across the over a century of their use has been to increase and enhance the brightness and resolution of the images in particular for soft-tissue diagnostics.
“The basics of X-ray imaging is resolution and contrast,” explained Hertz.
Excillum, which was established in 2007 and from small beginnings now includes 100 personnel, has been a leader in this area and is currently involved in developing sources enabling high-spatial-resolution X-ray bioimaging.
“Such imaging requires a small spot high-flux source,” explained Hertz. “This enables several biomedical phase contrast and other imaging applications where high spatial resolution, high contrast and short exposure time are critical.”
Hertz explained that Excillum’s liquid-metal-jet microfocus X ray sources have ten times higher brightness than classical micro-focus sources; give a clean spot with high spatial stability; and, enable high speed, resolution and contrast.
They are also working towards a number of new technologies including ex-vivo clinical propagation-based imaging – an X-ray histology with cellular and subcellular resolution and, hopefully, in future also intra-operative 3D resection-margin assessment; in-vivo clinical propagation based imaging – which is currently being tested in mouse lung and computational imaging and virtual clinical trials; and, also in-vivo molecular X-ray imaging which aims for X-ray fluorescence from targeted nanoparticles.
He explained that surgeons removing cancer tumours always have the challenge of knowing if they are removing enough. Being able to image the resection margin for the presence of any tumour cells is therefore vital. “Right now, we are able to do this in 3D but it still takes too long,” said Hertz. “Ideally you would hope to be able to do this and come back with a result while the patient is still on the operating table so that all the tumour cells can be removed in one operation. We eventually hope to have a version that takes only an hour, but we are still not there.”
The in-vivo propagation-based imaging currently being trialled in mouse and computer models could eventually be used to diagnose small airways problems – like Chronic Obstructive Pulmonary Disease and Asthma.
“We hope our systems for soft-tissue imaging will eventually compete favourably as regards speed, resolution and complexity of system with the present standard for soft-tissue imaging, magnetic resonance imaging,” he said.
Lessons from start-ups
Hertz also gave some insights from his two decades of experience into the challenges of establishing a science or technology start-up.
“In the high-tech sector it is good to start with a unique product but a good rule of thumb is to ask if it is 10 times better than everything else?” he said. “The most important is that you need more capital than you expect. You also need to be able to hire people who will take the risk to work in a start-up. Luckily, this is something that has changed hugely in Sweden in the last 30 years. Young people now are much more prepared to jump from a start-up to a start-up. They don’t have as much expectation of keeping a job for life.”
“Timing is also difficult – you can’t start too late or too early. And, it always takes longer than expected.”
“Failures are usually due to technical, market or capital issues,” he added. “80% fail – this is often not talked about and you don’t necessarily understand why.”
“You usually start by selling one machine a year often to academic institutions where it isn’t a major crisis if it breaks down; then it’s ten a year and the customers are not so nerdy; then it’s a 100 a year into the industrial environment where it’s another game because a breakdown stops a production line and you are no longer dealing with academics but with CFOs. The commercial sector is brutal – it counts money. You can only really start playing when you learn that.”