TEDxKrakow
Tal Golesworthy: How I repaired my own heart
Filmed:
Readability: 4.6
1,418,296 views
Tal Golesworthy is a boiler engineer -- he knows piping and plumbing. When he needed surgery to repair a life-threatening problem with his aorta, he mixed his engineering skills with his doctors' medical knowledge to design a better repair job.
Tal Golesworthy - Engineer and artist
Tal Golesworthy is an engineer and entrepreneur, working in research and development of combustion and air pollution control -- until he decided to innovate in his own health. Full bio
Tal Golesworthy is an engineer and entrepreneur, working in research and development of combustion and air pollution control -- until he decided to innovate in his own health. Full bio
Double-click the English transcript below to play the video.
00:15
I'm a process engineer.
0
0
2000
00:17
I know all about boilers and incinerators
1
2000
3000
00:20
and fabric filters and cyclones and things like that,
2
5000
3000
00:23
but I also have Marfan syndrome.
3
8000
2000
00:25
This is an inherited disorder.
4
10000
3000
00:28
And in 1992
5
13000
2000
00:30
I participated in a genetic study
6
15000
2000
00:32
and found to my horror, as you can see from the slide,
7
17000
3000
00:35
that my ascending aorta was not in the normal range,
8
20000
3000
00:38
the green line at the bottom.
9
23000
2000
00:40
Everyone in here will be between 3.2 and 3.6 cm.
10
25000
3000
00:43
I was already up at 4.4.
11
28000
3000
00:46
And as you can see,
12
31000
2000
00:48
my aorta dilated progressively,
13
33000
2000
00:50
and I got closer and closer
14
35000
2000
00:52
to the point where surgery was going to be necessary.
15
37000
3000
00:55
The surgery on offer was pretty gruesome --
16
40000
3000
00:58
anesthetize you, open your chest,
17
43000
3000
01:01
put you on an artificial heart and lung machine,
18
46000
3000
01:04
drop your body temperature to about 18 centigrade,
19
49000
3000
01:07
stop your heart, cut the aorta out,
20
52000
3000
01:10
replace it with a plastic valve and a plastic aorta,
21
55000
3000
01:13
and, most importantly,
22
58000
2000
01:15
commit you to a lifetime of anticoagulation therapy,
23
60000
3000
01:18
normally warfarin.
24
63000
2000
01:20
The thought of the surgery was not attractive.
25
65000
3000
01:23
The thought of the warfarin
26
68000
2000
01:25
was really quite frightening.
27
70000
2000
01:27
So I said to myself, I'm an engineer, I'm in R and D,
28
72000
3000
01:30
this is just a plumbing problem.
29
75000
2000
01:32
I can do this. I can change this.
30
77000
2000
01:34
So I set out
31
79000
2000
01:36
to change the entire treatment
32
81000
2000
01:38
for aortic dilation.
33
83000
3000
01:41
The project aim is really quite simple.
34
86000
4000
01:45
The only real problem with the ascending aorta
35
90000
3000
01:48
in people with Marfan syndrome
36
93000
2000
01:50
is it lacks some tensile strength.
37
95000
2000
01:52
So the possibility exists
38
97000
2000
01:54
to simply externally wrap the pipe.
39
99000
4000
01:58
And it would remain stable and operate quite happily.
40
103000
3000
02:01
If your high-pressure hose pipe,
41
106000
2000
02:03
or your high-pressure hydraulic line, bulges a little,
42
108000
3000
02:06
you just wrap some tape around the outside of it.
43
111000
2000
02:08
It really is that simple in concept,
44
113000
3000
02:11
though not in execution.
45
116000
3000
02:14
The great advantage of an external support for me
46
119000
3000
02:17
was that I could retain all of my own bits,
47
122000
3000
02:20
all of my own endothelium and valves,
48
125000
2000
02:22
and not need any anticoagulation therapy.
49
127000
3000
02:25
So where do we start?
50
130000
2000
02:27
Well this is a sagittal slice through me.
51
132000
3000
02:30
You could see in the middle
52
135000
2000
02:32
that device, that little structure, squeezing out.
53
137000
3000
02:35
Now that's a left ventricle
54
140000
2000
02:37
pushing blood up through the aortic valve --
55
142000
2000
02:39
you can see two of the leaflets of the aortic valve working there --
56
144000
4000
02:43
up into the ascending aorta.
57
148000
2000
02:45
And it's that part, the ascending aorta,
58
150000
2000
02:47
which dilates and ultimately bursts,
59
152000
4000
02:51
which, of course, is fatal.
60
156000
3000
02:54
We started by organizing image acquisition
61
159000
3000
02:57
from magnetic resonance imaging machines
62
162000
3000
03:00
and CT imaging machines
63
165000
2000
03:02
from which to make a model
64
167000
3000
03:05
of the patient's aorta.
65
170000
3000
03:08
This is a model of my aorta.
66
173000
3000
03:11
I've got a real one in my pocket,
67
176000
2000
03:13
if anyone would like to look at it and play with it.
68
178000
3000
03:16
You can see, it's quite a complex structure.
69
181000
2000
03:18
It has a funny trilobal shape at the bottom,
70
183000
3000
03:21
which contains the aortic valve.
71
186000
2000
03:23
It then comes back into a round form
72
188000
2000
03:25
and then tapers and curves off.
73
190000
2000
03:27
So it's quite a difficult structure
74
192000
2000
03:29
to produce.
75
194000
3000
03:32
This, like I say, is a CAD model of me,
76
197000
2000
03:34
and this is one of the later CAD models.
77
199000
2000
03:36
We went through an iterative process
78
201000
2000
03:38
of producing better and better models.
79
203000
2000
03:40
When we produced that model
80
205000
3000
03:43
we turn it into a solid plastic model,
81
208000
3000
03:46
as you can see,
82
211000
2000
03:48
using a rapid prototyping technique,
83
213000
2000
03:50
another engineering technique.
84
215000
2000
03:52
We then use that former
85
217000
2000
03:54
to manufacture a perfectly bespoke
86
219000
2000
03:56
porous textile mesh,
87
221000
2000
03:58
which takes the shape of the former
88
223000
2000
04:00
and perfectly fits the aorta.
89
225000
2000
04:02
So this is absolutely personalized medicine
90
227000
2000
04:04
at its best really.
91
229000
2000
04:06
Every patient we do
92
231000
2000
04:08
has an absolutely bespoke implant.
93
233000
4000
04:12
Once you've made it, the installation's quite easy.
94
237000
3000
04:15
John Pepper, bless his heart,
95
240000
3000
04:18
professor of cardiothoracic surgery --
96
243000
2000
04:20
never done it before in his life --
97
245000
2000
04:22
he put the first one in, didn't like it, took it out, put the second one in.
98
247000
2000
04:24
Happy, away I went.
99
249000
2000
04:26
Four and a half hours on the table and everything was done.
100
251000
3000
04:29
So the surgical implantation actually was the easiest part.
101
254000
4000
04:33
If you compare our new treatment to the existing alternative,
102
258000
3000
04:36
the so-called composite aortic root graft,
103
261000
3000
04:39
there are one of two startling comparisons,
104
264000
2000
04:41
which I'm sure will be clear to all of you.
105
266000
3000
04:44
Two hours to install one of our devices
106
269000
2000
04:46
compared to six hours
107
271000
2000
04:48
for the existing treatment.
108
273000
2000
04:50
The existing treatment requires, as I've said,
109
275000
2000
04:52
the heart and lung bypass machine
110
277000
2000
04:54
and it requires a total body cooling.
111
279000
2000
04:56
We don't need any of that; we work on a beating heart.
112
281000
3000
04:59
He opens you up, he accesses the aorta while your heart is beating,
113
284000
3000
05:02
all at the right temperature.
114
287000
2000
05:04
No breaking into your circulatory system.
115
289000
2000
05:06
So it really is great.
116
291000
2000
05:08
But for me, absolutely the best point
117
293000
3000
05:11
is there is no anticoagulation therapy required.
118
296000
3000
05:14
I don't take any drugs at all
119
299000
2000
05:16
other than recreational ones that I would choose to take.
120
301000
2000
05:18
(Laughter)
121
303000
2000
05:20
And in fact, if you speak to people who are on long-term warfarin,
122
305000
3000
05:23
it is a serious compromise to your quality of life.
123
308000
3000
05:26
And even worse,
124
311000
2000
05:28
it inevitably foreshortens your life.
125
313000
2000
05:30
Likewise, if you have the artificial valve option,
126
315000
3000
05:33
you're committed to antibiotic therapy
127
318000
2000
05:35
whenever you have any intrusive medical treatment at all.
128
320000
3000
05:38
Even trips to the dentist require that you take antibiotics,
129
323000
3000
05:41
in case you get an internal infection on the valve.
130
326000
3000
05:44
Again, I don't have any of that, so I'm entirely free.
131
329000
3000
05:47
My aorta is fixed, I haven't got to worry about it,
132
332000
3000
05:50
which is a rebirth for me.
133
335000
3000
05:54
Back to the theme of the presentation:
134
339000
2000
05:56
In multidisciplinary research,
135
341000
2000
05:58
how on earth does a process engineer used to working with boilers
136
343000
3000
06:01
end up producing a medical device
137
346000
2000
06:03
which transforms his own life?
138
348000
2000
06:05
Well the answer to that is a multidisciplinary team.
139
350000
3000
06:08
This is a list of the core team.
140
353000
3000
06:11
And as you can see,
141
356000
2000
06:13
there are not only two principal technical disciplines there,
142
358000
3000
06:16
medicine and engineering,
143
361000
2000
06:18
but also there are various specialists
144
363000
2000
06:20
from within those two disciplines.
145
365000
2000
06:22
John Pepper there
146
367000
2000
06:24
was the cardiac surgeon who did the actual work on me,
147
369000
3000
06:27
but everyone else there had to contribute one way or another.
148
372000
3000
06:30
Raad Mohiaddin, medical radiologist:
149
375000
2000
06:32
We had to get good quality images
150
377000
2000
06:34
from which to make the CAD model.
151
379000
2000
06:36
Warren Thornton, who still does all our CAD models for us,
152
381000
3000
06:39
had to write a bespoke piece of CAD code
153
384000
3000
06:42
to produce this model
154
387000
2000
06:44
from this really rather difficult input data set.
155
389000
3000
06:49
There are some barriers to this though. There are some problems with it.
156
394000
3000
06:52
Jargon is a big one.
157
397000
2000
06:54
I would think no one in this room understands
158
399000
2000
06:56
those four first jargon points there.
159
401000
3000
06:59
The engineers amongst you
160
404000
2000
07:01
will recognize rapid prototyping and CAD.
161
406000
2000
07:03
The medics amongst you, if there are any, will recognize the first two.
162
408000
3000
07:06
But there will be nobody else in this room
163
411000
2000
07:08
that understands all of those four words.
164
413000
2000
07:10
Taking the jargon out was very important
165
415000
2000
07:12
to ensure that everyone in the team
166
417000
2000
07:14
understood exactly what was meant
167
419000
2000
07:16
when a particular phrase was used.
168
421000
2000
07:18
Our disciplinary conventions were funny as well.
169
423000
3000
07:21
We took a lot of horizontal slice images through me,
170
426000
3000
07:24
produced those slices and then used those to build a CAD model.
171
429000
4000
07:28
And the very first CAD model we made,
172
433000
2000
07:30
the surgeons were playing with the plastic model,
173
435000
3000
07:33
couldn't quite figure it out.
174
438000
2000
07:35
And then we realized that it was actually a mirror image
175
440000
2000
07:37
of the real aorta.
176
442000
2000
07:39
And it was a mirror image
177
444000
2000
07:41
because in the real world we always look down on plans,
178
446000
3000
07:44
plans of houses or streets or maps.
179
449000
3000
07:47
In the medical world they look up at plans.
180
452000
3000
07:50
So the horizontal images were all an inversion.
181
455000
3000
07:53
So one needs to be careful with disciplinary conventions.
182
458000
3000
07:56
Everyone needs to understand
183
461000
2000
07:58
what is assumed and what is not assumed.
184
463000
3000
08:01
Institutional barriers
185
466000
2000
08:03
were another serious headache in the project.
186
468000
3000
08:06
The Brompton Hospital was taken over
187
471000
2000
08:08
by Imperial College's School of Medicine,
188
473000
2000
08:10
and there are some seriously bad relationship problems
189
475000
3000
08:13
between the two organizations.
190
478000
2000
08:15
I was working with Imperial and the Brompton,
191
480000
2000
08:17
and this generated some serious problems with the project,
192
482000
3000
08:20
really problems that shouldn't exist.
193
485000
3000
08:23
Research and ethics committee: If you want to do anything new in surgery,
194
488000
3000
08:26
you have to get a license from your local research and ethics.
195
491000
3000
08:29
I'm sure it's the same in Poland.
196
494000
2000
08:31
There will be some form of equivalent,
197
496000
3000
08:34
which licenses new types of surgery.
198
499000
3000
08:37
We didn't only have the bureaucratic problems associated with that,
199
502000
3000
08:40
was also had professional jealousies.
200
505000
2000
08:42
There were people on the research and ethics committee
201
507000
2000
08:44
who really didn't want to see John Pepper succeed again,
202
509000
3000
08:47
because he's so successful.
203
512000
2000
08:49
And they made extra problems for us.
204
514000
3000
08:53
Bureaucratic problems:
205
518000
2000
08:55
Ultimately when you have a new treatment
206
520000
3000
08:58
you have to have a guidance note going out
207
523000
2000
09:00
for all of the hospitals in the country.
208
525000
2000
09:02
In the U.K. we have the National Institute for Clinical Excellence, NICE.
209
527000
2000
09:04
You'll have an equivalent in Poland, no doubt.
210
529000
3000
09:07
We had to get past the NICE problem.
211
532000
3000
09:10
We now have a great clinical guidance out on the Net.
212
535000
3000
09:13
So any of the hospitals interested
213
538000
2000
09:15
can come along, read the NICE report
214
540000
3000
09:18
get in touch with us and then get doing it themselves.
215
543000
3000
09:23
Funding barriers:
216
548000
2000
09:25
Another big area to be concerned with.
217
550000
3000
09:29
A big problem with understanding one of those perspectives:
218
554000
3000
09:32
When we first approached
219
557000
2000
09:34
one of the big U.K. charitable organizations that funds this kind of stuff,
220
559000
4000
09:38
what they were looking at was essentially an engineering proposal.
221
563000
3000
09:41
They didn't understand it; they were doctors, they were next to God.
222
566000
2000
09:43
It must be rubbish. They binned it.
223
568000
2000
09:45
So in the end I went to private investors
224
570000
2000
09:47
and I just gave up on it.
225
572000
2000
09:49
But most R and D is going to be institutionally funded,
226
574000
3000
09:52
by the Polish Academy of Sciences
227
577000
2000
09:54
or the Engineering and Physical Sciences Research Council or whatever,
228
579000
4000
09:58
and you need to get past those people.
229
583000
2000
10:00
Jargon is a huge problem when you're trying to work across disciplines,
230
585000
3000
10:03
because in an engineering world,
231
588000
2000
10:05
we all understand CAD and R.P. --
232
590000
2000
10:07
not in the medical world.
233
592000
2000
10:09
I suppose ultimately the funding bureaucrats have really got to get their act together.
234
594000
3000
10:12
They've really got to start talking to each other,
235
597000
2000
10:14
and they've got to exercise a bit of imagination,
236
599000
3000
10:17
if that's not too much to ask --
237
602000
3000
10:20
which it probably is.
238
605000
3000
10:23
I've coined a phrase "obstructive conservatism."
239
608000
3000
10:26
So many people in the medical world don't want to change,
240
611000
3000
10:29
particularly not when some jumped-up engineer has come along with the answer.
241
614000
3000
10:32
They don't want to change.
242
617000
2000
10:34
They simply want to do whatever they've done before.
243
619000
2000
10:36
And in fact, there are many surgeons in the U.K.
244
621000
2000
10:38
still waiting for one of our patients
245
623000
3000
10:41
to have some sort of episode,
246
626000
2000
10:43
so that they can say, "Ah, I told you that was no good."
247
628000
3000
10:46
We've actually got 30 patients.
248
631000
2000
10:48
I'm at seven and a half years.
249
633000
2000
10:50
We've got 90 post-op patient years between us,
250
635000
2000
10:52
and we haven't had a single problem.
251
637000
2000
10:54
And still, there are people in the U.K. saying,
252
639000
2000
10:56
"Yeah, that external aortic root, yeah, it'll never work, you know."
253
641000
3000
10:59
It really is a problem. It really is a problem.
254
644000
2000
11:01
I'm sure everyone in this room has come across arrogance
255
646000
3000
11:04
amongst medics, doctors, surgeons at some point.
256
649000
4000
11:08
The middle point is simply
257
653000
2000
11:10
the way that the doctors protect themselves.
258
655000
3000
11:13
"Yeah, well of course, I'm looking after my patient."
259
658000
3000
11:16
I think it's not good, but there you are, that's my view.
260
661000
3000
11:19
Egos, of course, again, a huge problem
261
664000
2000
11:21
If you're working in a multidisciplinary team,
262
666000
2000
11:23
you've got to give your guys the benefit of the doubt.
263
668000
3000
11:26
You've got to express support for them.
264
671000
2000
11:28
Tom Treasure, professor of cardiothoracic surgery:
265
673000
3000
11:31
incredible guy.
266
676000
2000
11:33
Dead easy to give him respect.
267
678000
2000
11:35
Him giving me respect? Slightly different.
268
680000
3000
11:38
That's all the bad news.
269
683000
2000
11:40
The good news is the benefits are stonkingly huge.
270
685000
4000
11:44
Translate that one. I bet they can't.
271
689000
2000
11:46
(Laughter)
272
691000
2000
11:48
When you have a group of people
273
693000
2000
11:50
who have had a different professional training, a different professional experience,
274
695000
3000
11:53
they not only have a different knowledge base,
275
698000
2000
11:55
but they have a different perspective on everything.
276
700000
2000
11:57
And if you can bring those guys together
277
702000
2000
11:59
and you can get them talking and understanding each other,
278
704000
3000
12:02
the results can be spectacular.
279
707000
3000
12:05
You can find novel solutions, really novel solutions,
280
710000
3000
12:08
that have never been looked at before
281
713000
2000
12:10
very, very quickly and easily.
282
715000
2000
12:12
You can shortcut huge amounts of work
283
717000
3000
12:15
simply by using the extended knowledge base you have.
284
720000
3000
12:18
And as a result,
285
723000
2000
12:20
it's an entirely different use of the technology
286
725000
3000
12:23
and the knowledge around you.
287
728000
3000
12:26
The result of all this
288
731000
2000
12:28
is that you can get incredibly quick progress
289
733000
3000
12:31
on incredibly small budgets.
290
736000
2000
12:33
I'm so embarrassed at how cheap it was
291
738000
2000
12:35
to get from my idea to me being implanted
292
740000
3000
12:38
that I'm not prepared to tell you what it cost.
293
743000
2000
12:40
Because I suspect there are
294
745000
2000
12:42
absolutely standard surgical treatments
295
747000
2000
12:44
probably in the USA
296
749000
2000
12:46
which cost more
297
751000
2000
12:48
for a one-off patient
298
753000
3000
12:51
than the cost of us getting from my dream
299
756000
3000
12:54
to my reality.
300
759000
2000
12:56
That's all I want to say, and I've got three minutes left.
301
761000
3000
12:59
So Heather's going to like me.
302
764000
2000
13:01
If you have any questions, please come up and talk to me later on.
303
766000
3000
13:04
It would be a pleasure to speak with you. Many thanks.
304
769000
3000
ABOUT THE SPEAKER
Tal Golesworthy - Engineer and artistTal Golesworthy is an engineer and entrepreneur, working in research and development of combustion and air pollution control -- until he decided to innovate in his own health.
Why you should listen
Tal Golesworthy is an experienced engineer and entrepreneur who branched out into bio-engineering in response to his own dilated aorta. His company now funds the research and development of devices that will help people with aortic dilation avoid major surgery and lifelong drug therapy.
He also has considerable expertise over a wide range of combustion and process plant including conventional and novel combustion systems, and his background has also included several years as an information scientist, and hot gas cleaning.
Tal Golesworthy | Speaker | TED.com