Home
Perspectives
When a “child prodigy” in physics comes of age
When a “child prodigy” in physics comes of age
Silvia Arroyo Camejo became a scientific author already as a teenager. Today, she's a data scientist helping to make MRI scanners even smarter, powered by artificial intelligence. Our #Futureshaper series kicks off with a feature portrait of this passionate physicist.
When many in her age group were thumbing through teen fan magazines, she was poring over books on theoretical physics. Silvia Arroyo Camejo was all of 17 when she decided to write a book herself: on the cryptic world of quantum physics – the shear unintelligibility of which had captured her imagination.
She received various awards while still in secondary school, including recognition in 2005 from the German Physical Society (DPG) for outstanding achievements in the field of physics. Silvia's book Skurrile Quantenwelt ("Crazy Quantum World") published in 2006 was translated from German into four other languages. She herself now has only one last copy of the book – in Japanese: "Over time I gave away the remaining copies to family and friends when they asked me," she notes, laughing.
That Silvia, after her long journey through theoretical physics, was then drawn to applied medical technology, was certainly due in part to her father, a Spanish-born vascular surgeon: "I want to use my abilities in ways that people actually benefit from directly. I can achieve significantly greater impact in medical technology than by pursuing basic research in physics."
#Futureshaper @ Siemens Healthineers
Whether technical developers, creative business managers or product designers: in our #Futureshaper series, we present employees who, with their innovative ideas, are helping to shape the future of healthcare.
Predevelopment work – as the term suggests – is a preparatory stage prior to plunging into actual product development. It's precisely the right place for nerdy Silvia, the tinkerer, whose light green eyes start to gleam when she talks about her work: "It's like a giant playing field. We have diverse technological components that we have to correctly unite into a whole."
Apparently, I have tendency to develop a certain obsession for topics that I find really fascinating.
Silvia Arroyo Camejo, Data Scientist Magnetic Resonance Imaging at Siemens Healthineers
What do our customers really need?
Sylvia's fingers dance across her computer keyboard. Cryptic-looking, multi-colored combinations of characters string together on the computer screen. Silvia is coding using the Python programming language. Her task scope includes programming complex algorithms, as well as coordinating the technical collaboration among her team members.
Silvia says that an in-depth needs analysis was conducted by directly exchanging ideas and information with international customers before starting the predevelopment project. What do these various hospitals, clinics, practices, and practice chains need? What are they lacking? And what would be just the right technical tool to help them?
A nagging shortage of highly qualified personnel
The smart scanning function that Silvia and her team are researching aims to make MRI scans considerably less dependent on the level of experience of attending MTRAs, and to do so using artificial intelligence – on the basis of Deep Learning algorithms, for instance. You can imagine it similarly to autonomous driving, explains Silvia. Here, as well, various levels of automation are at play.
From a physical standpoint, magnetic resonance imaging is a highly complex and, for this very reason, incredibly beautiful and interesting process.
Silvia Arroyo Camejo, Data Scientist Magnetic Resonance Imaging at Siemens Healthineers
As near as possible to "autonomous driving"
Some MRI scanners from Siemens Healthineers already operate today using what are called "Dot Engines" which, in part, perform intelligent functions as part of scan preparation. For example, they can correctly document the anatomy of patients and, based on this log, automatically adjust initial scan parameters accordingly.
The smart scanning project that Silvia is pursuing aims to go one step further by significantly simplifying the operation of MRI scanners by increasing the level of automation: "We seek to get as close as possible to autonomous driving, so to speak – while remaining fully aware that the MTRA always has the final say, and can intervene at any time."
The new smart scanning function the team is working on could make it possible in future for image quality controls to be performed automatically: artificial intelligence could recommend, for example, whether scans need to be repeated, or what particular scan steps should come next in the sequence.
Another advancement in precision medicine
While it may sound contradictory at first, precisely this standardization of processes would mark another step towards precision medicine: there would be no "one size fits all", and all scanning steps could be individually adapted to the given patient. As Silvia puts it, "The scanner watches quasi continuously over the shoulder of the MTRA, as an assistant, noting to the MTRA for example, "You know what? For this patient I'd recommend that we also run the tumor protocol, as there's something suspicious-looking there."
Images generated more precisely, which undergo automatic image quality control on the scanner, would ensure that fewer patients have to be called back in for repeat scans. The potentially life-saving "time to diagnosis" factor could be shortened. For hospitals and clinics, the process could substantially relieve the shortage of skilled personnel, enable workflows to be designed more efficiently, cut costs, and reduce the load on the healthcare system as a whole.
And apropos the future: Does Silvia plan to author more books? "The idea does indeed tempt me. Perhaps someday later I'll find the time again to do so. But", she notes, "it's also important to leave the computer behind now and then, right?" – as that slightly cryptic smile appears again on her face.