Digital twins to predict disease

Vir­tu­al real­i­ties are a way of extend­ing and improv­ing the real world, and this is also the role of a dig­i­tal twin. In tan­gi­ble terms, they pro­vide a frame of ref­er­ence upon which dif­fer­ent dis­ci­plines can col­lab­o­rate on a com­mon project. By com­bin­ing their knowl­edge and exper­tise, they will be able to share, on the basis of the same rep­re­sen­ta­tion, sev­er­al pos­si­ble scenarios.

Nobody expects a dig­i­tal twin to be per­fect. To draw a par­al­lel with the auto­mo­tive indus­try, 95% of crash tests on cars on the mar­ket are car­ried out on com­put­ers. Thanks to the dig­i­tal twin of a vehi­cle, even if it is incom­plete, we can test its resis­tance to a crash, for exam­ple – hence, these are mod­els that can be trust­ed. The human body is obvi­ous­ly more com­plex than a car, but our objec­tive is sim­i­lar: we want to improve life expectan­cy and keep peo­ple healthy at a cost that is sus­tain­able for the health care system. 

To do this, at Das­sault Sys­tèmes we have imple­ment­ed a real flex­i­bil­i­ty between the dif­fer­ent scales and dis­ci­plines that are used by these vir­tu­al tools. Our dig­i­tal twin of the heart is defined as mul­ti-phys­i­cal. The heart is a pump, and there­fore has mechan­i­cal behav­iour that make it con­tract when stim­u­lat­ed by an elec­tri­cal sig­nal. We can dig­i­tal­ly repli­cate this phe­nom­e­non and the way the pump is con­trolled. As such, this twin is also capa­ble of repro­duc­ing the geom­e­try of the heart from an imag­ing sys­tem, but also the behav­iour of the mus­cle fibres: this is the first type of physics that we have used.

Although a mod­el is can only make use of the data you put into it, it can still be used to make predictions.

This mod­el heart con­tracts, cir­cu­lat­ing the blood. From there, we went on to call upon oth­er types of physics to repro­duce oth­er actions like flu­id mechan­ics to best char­ac­terise the haemo­dy­nam­ic behav­iour. The chal­lenge is to cus­tomise this heart mod­el to each indi­vid­ual so that we can, for exam­ple, test dif­fer­ent ways of treat­ing a patient when he or she has an ill­ness. Based on this con­fig­urable mod­el, it is pos­si­ble to cre­ate a pre­cise repro­duc­tion of the patien­t’s heart and its behav­iour, both in terms of geom­e­try but also in terms of the elec­tro­phys­i­o­log­i­cal ellipse or physics. Although a mod­el is only com­prised of the data you put into it, it can nev­er­the­less be used to make predictions. 

We are also capa­ble of a mul­ti-scale approach. For the heart, it is pos­si­ble to change the tool to repro­duce car­diotox­ic behav­iour. Some drugs will be able to enter heart cells and con­trol the opening/closing of ion chan­nels. Here we enter the field of chem­istry and mol­e­c­u­lar biol­o­gy. We can thus gen­er­ate arrhyth­mias of the heart that go as far as tor­sades de pointe [a spe­cif­ic type of abnor­mal heart rhythm], which our soft­ware is able to repro­duce accord­ing to the drug dosage. In this instance we are work­ing on a mol­e­c­u­lar scale, not just in the field of physics. 

We there­fore need to run sev­er­al dig­i­tal twins of the patient to under­stand their con­di­tion, iden­ti­fy their pathol­o­gy and be able to test dif­fer­ent inter­ven­tion sce­nar­ios. While longevi­ty has been increas­ing in recent years – accord­ing to WHO fig­ures – healthy life expectan­cy has stag­nat­ed. To over­come this plateau, this mul­ti­plic­i­ty will be nec­es­sary. For exam­ple, an inter­ven­tion sce­nario may use a drug or a med­ical device, or even both. Anoth­er dig­i­tal twin to test the mol­e­cule or med­ical device will then be required. This care will be admin­is­tered in a par­tic­u­lar envi­ron­ment: if it is an oper­a­tion, it will take place in a hos­pi­tal, but it may be a remote con­sul­ta­tion or a face-to-face con­sul­ta­tion in a doc­tor’s surgery. We will then need a dig­i­tal twin of the site.

Final­ly, mak­ing a dig­i­tal twin of the care sys­tem will also require test­ing to under­stand cash­flows and seek the best ways to man­age it. It can even be in 2D – a dig­i­tal twin does not nec­es­sar­i­ly need to be in 3D! Even if it is the most infor­ma­tive mod­el to illus­trate the tech­nol­o­gy and, in fact, we don’t need to wait until we have a com­plete dig­i­tal twin of the human body to see sol­id results. 

These advances will allow us to move into per­son­alised med­i­cine and find the right treat­ment for the right patient at the right time. There will also be sev­er­al sec­ondary ben­e­fits. First­ly, the patient will have a bet­ter under­stand­ing of what is going to hap­pen dur­ing the oper­a­tion. This will enable them to become active in their own health­care, thus pro­mot­ing more effec­tive pre­ven­tion. In addi­tion, this progress would free up doc­tors’ time. If a dig­i­tal twin is used to test dif­fer­ent con­fig­u­ra­tions that used to be done by hand, this will free up med­ical staff time with­out under­min­ing the rela­tion­ship with the patient. It also promis­es bet­ter col­lab­o­ra­tion between doctors. 

There is already a high demand from doc­tors for these dig­i­tal twins.

There is already a high demand from doc­tors for these dig­i­tal twins, espe­cial­ly as they com­plain that they spend too much time on admin­is­tra­tion and not enough time with patients. But this is only the begin­ning, and while some are con­vinced of the mer­its of this approach, they are still pio­neers in this field. The need is all the more press­ing now that med­ical knowl­edge dou­bles every 70 days. And these doc­tors see the advan­tage of being able to pool their knowl­edge to train the younger gen­er­a­tions or to dis­cuss a com­plex case with sev­er­al practitioners. 

One exam­ple is the Boston Chil­dren’s Hos­pi­tal, where the head of pae­di­atric surgery, Dr Hogan, uses a mod­el of the heart before oper­at­ing on infants with con­gen­i­tal mal­for­ma­tions. This dig­i­tal twin of the heart helps him to bet­ter under­stand the child’s ill­ness and allows him to visu­al­ly explain their child’s con­di­tion to par­ents to build confidence. 

This is proof that the tech­nol­o­gy can be used in advanced med­ical prac­tice, but we are still far from wide­spread adop­tion. While the med­ical world is keen on inno­va­tions, it also wants them to be proven. How­ev­er, these tech­nolo­gies are quite rea­son­able in terms of cost and time. In two- or three-years’ time, they will have great­ly evolved in terms of speed, ren­der­ing, results and con­sump­tion of tech­no­log­i­cal resources, which will encour­age their use through­out the med­ical profession.

Interview by Jean Zeid