Manufacturing technology
explains de Vecchi, “and then you introduce stresses. You test the actual device [digitally] in a variety of scenarios to get a sense of the limits of usage.” Results in hand, the product can be refined accordingly.
From devices to patients Increasingly, however, the interest is less in replicating devices and more in replicating the humans in which those devices will be used. By creating a digital model of an individual patient’s heart, for instance – one that reflects their unique physiology – it could become possible to understand which specific device and intervention might suit that patient best and, more broadly, to understand what devices are needed in the market.
Dr Steven Niederer first built a digital twin of a heart in the mid-2000s. A computational model of a rat heart, part of his DPhil at the University of Oxford, it took four years to build. But today, thanks to advances in AI and computational power, he and colleagues can create the basics of a human heart digital twin in the time it took the rat twin to model a single heartbeat: 48 hours. Whereas once a clinical fellow would have had to review hundreds of MRI and CT images, marking up the anatomical features and labelling different regions, machine learning can now do much of that work. And by building person- specific twins so swiftly, there’s much more potential to ultimately impact care. “We’re trying to make models of people, and then trying to use medical devices to get data from those people, and use that data to update the digital twin,” explains Niederer, National Heart and Lung Institute chair of biomedical engineering at Imperial College London and health mission lead at the Alan Turing Institute, the UK’s national institute for data science and AI. “In the longer term, once validated, that will feed into informing patient care, maybe by informing the clinician.”
For Jeff Bodner and colleagues at Medtronic, the clinician is very much the focus of digital twin work. “We’ve aligned around the idea that [a digital twin] is simulation used in the clinic, by clinicians, for a specific patient, to help optimise something about the therapy,” says Bodner, a distinguished scientist at the firm and a member of the digital twin team within its AI centre of excellence.
“Usually what that looks like is we take an image of a patient, and we use that to build a patient-specific simulation of that patient and the deployment of a device in that anatomy. Then out comes some recommendation – something that says you’ve got the right size or you need a different size, a different length, a different position, different settings. And that information is used by the clinician to make a decision. At Medtronic, when clinical decision support involves a simulation, that’s what we call a digital twin.”
www.medicaldevice-developments.com Outline of a medical digital twin Data connection
Multimodal
Visualise disease
Assess disease prognosis
Simulate treatment
Physical object (patient)
Patient-in-silico
AI interpretation AI
Interface
“The physical object is described by a plethora of different data modalities (eg, electronic health records, imaging studies, and genetic data), which are processed and combined using data fusion approaches forming the data connection. The combined information is forwarded to the patient-in-silico model to visualise the disease, assess disease prognosis and simulate treatments. The interface, with the aid of artifi cial intelligence (AI), allows the clinical team and patient to select an optimal treatment plan based on the patient-in-silico. The cycle is repeated as new patient data become available, synchronising the patient and the patient- in-silico (twin synchronisation).” Source: The Lancet. Figure created with
BioRender.com.
It’s an area the company is actively advancing. In April, for instance, it announced its completed acquisition of CathWorks, a firm using digital twins to understand blood flow through the coronary arteries and, accordingly, the best placement for stents.
“We’re trying to make models of people, and then trying to use medical devices to get data from those people, and use that data to update the digital twin.”
Dr Steven Niederer
Twins in medtech Increasingly, then, digital twins are becoming medical devices in and of themselves. They’re not solely models created and used by design engineers like Bodner to simulate situations. They’re tools used by clinicians while caring for a patient. And that means they need to have a significantly different user interface.
“Clinical support is a very different game because you can’t wait ten hours for the simulation to finish,” says Bodner. “It’s got to be nearly real time, which means the approach has to be very different. The software has to be different, because it becomes
45
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68 |
Page 69 |
Page 70 |
Page 71 |
Page 72 |
Page 73 |
Page 74 |
Page 75 |
Page 76 |
Page 77 |
Page 78 |
Page 79 |
Page 80 |
Page 81 |
Page 82 |
Page 83 |
Page 84 |
Page 85 |
Page 86 |
Page 87 |
Page 88 |
Page 89 |
Page 90 |
Page 91 |
Page 92 |
Page 93 |
Page 94 |
Page 95 |
Page 96 |
Page 97 |
Page 98 |
Page 99 |
Page 100
