WEBVTT

1
00:00:00.088 --> 00:00:02.250
This podcast is created by Coelis.

2
00:00:02.950 --> 00:00:08.935
I would say artificial intelligence is a really new word and everybody is interesting in it.

3
00:00:09.656 --> 00:00:10.036
Of course,

4
00:00:10.096 --> 00:00:11.277
it's a promotion fact.

5
00:00:11.778 --> 00:00:13.119
Patients really like to say,

6
00:00:13.739 --> 00:00:15.581
I use AI for your biopsy.

7
00:00:16.421 --> 00:00:18.323
And for my medical life,

8
00:00:18.343 --> 00:00:25.629
I would say it makes the biopsy much better because I'm more sure where to take the targeted biopsies.

9
00:00:26.009 --> 00:00:26.209
And

10
00:00:26.550 --> 00:00:29.232
I feel more comfortable with it because I...

11
00:00:29.576 --> 00:00:33.718
Know that I will not overlook any lesion and prostate.

12
00:00:33.718 --> 00:00:33.898
Prostate.

13
00:00:33.898 --> 00:00:34.199
Prostate.

14
00:00:34.219 --> 00:00:35.079
Prostate.

15
00:00:35.079 --> 00:00:35.379
Prostate.

16
00:00:35.379 --> 00:00:35.579
Prostate.

17
00:00:35.579 --> 00:00:35.759
Prostate.

18
00:00:35.759 --> 00:00:35.940
Prostate.

19
00:00:35.940 --> 00:00:36.140
Prostate.

20
00:00:36.160 --> 00:00:36.340
Prostate.

21
00:00:36.340 --> 00:00:36.640
Prostate.

22
00:00:37.120 --> 00:00:37.380
Prostate.

23
00:00:37.380 --> 00:00:37.540
Prostate.

24
00:00:37.560 --> 00:00:37.721
Prostate.

25
00:00:37.721 --> 00:00:37.881
Prostate.

26
00:00:37.901 --> 00:00:38.061
Prostate.

27
00:00:38.061 --> 00:00:38.221
Prostate.

28
00:00:38.221 --> 00:00:38.381
Prostate.

29
00:00:38.381 --> 00:00:38.561
Prostate.

30
00:00:38.561 --> 00:00:38.721
Prostate.

31
00:00:38.761 --> 00:00:39.021
Prostate.

32
00:00:39.021 --> 00:00:39.201
Prostate.

33
00:00:39.201 --> 00:00:39.361
Prostate.

34
00:00:39.361 --> 00:00:39.522
Prostate.

35
00:00:39.522 --> 00:00:39.702
Prostate.

36
00:00:41.102 --> 00:00:41.302
Prostate.

37
00:00:41.302 --> 00:00:41.583
Prostate.

38
00:00:41.583 --> 00:00:41.803
Prostate.

39
00:00:41.803 --> 00:00:41.963
Prostate.

40
00:00:41.963 --> 00:00:43.324
Welcome to today's episode.

41
00:00:43.364 --> 00:00:47.386
We are talking about two things that might make you do a double take.

42
00:00:47.626 --> 00:00:50.407
Prostates and artificial intelligence.

43
00:00:51.148 --> 00:00:51.388
I know,

44
00:00:51.548 --> 00:00:55.050
it sounds like the beginning of a very niche sci-movie,

45
00:00:55.050 --> 00:00:55.690
but trust me,

46
00:00:56.351 --> 00:00:58.632
AI is playing a bigger role in prostate.

47
00:00:58.732 --> 00:01:16.997
cancer care than ever before imagine a world where machines can always scan faster than a doctor with a cup of coffee in hand yeah that's the future we are heading towards we'll break down how artificial intelligence is helping doctors make quicker more

48
00:01:16.997 --> 00:01:27.860
accurate diagnose guiding treatments and maybe even taking some of the guesswork out of prostate care so what i'm not sure ai will ever be able to tell if it's really a

49
00:01:27.980 --> 00:01:32.122
prostate issue or just a stubborn case of middle age stress.

50
00:01:32.903 --> 00:01:33.903
And to talk about it,

51
00:01:33.903 --> 00:01:36.365
I have the great pleasure to welcome Dr.

52
00:01:36.565 --> 00:01:39.026
Gunzel from Germany who uses

53
00:01:39.506 --> 00:01:41.067
AI in his daily practice.

54
00:01:41.668 --> 00:01:45.910
So buckle up as we navigate the world of AI and prostate cancer.

55
00:01:49.552 --> 00:01:49.752
Hi,

56
00:01:49.972 --> 00:01:50.212
Dr.

57
00:01:50.252 --> 00:01:50.573
Gunzel.

58
00:01:50.993 --> 00:01:54.255
We are really pleased to welcome you to our podcast Prostate Talk.

59
00:01:54.835 --> 00:01:56.456
We are live from the DGU.

60
00:01:56.772 --> 00:01:57.232
Congress,

61
00:01:57.412 --> 00:01:59.273
an important Congress in Germany.

62
00:02:00.014 --> 00:02:02.575
So how are you and how is the Congress going?

63
00:02:03.696 --> 00:02:03.896
Yeah,

64
00:02:04.096 --> 00:02:07.718
we have a really long night and I'm really glad to be here today.

65
00:02:08.258 --> 00:02:12.280
And I'm really looking forward to talk with you about prostate and AI.

66
00:02:13.581 --> 00:02:13.961
Thank you.

67
00:02:14.141 --> 00:02:15.402
Before to start this interview,

68
00:02:15.502 --> 00:02:17.983
could you please present yourself to our listeners?

69
00:02:18.944 --> 00:02:20.305
My name is Carsten Günzel.

70
00:02:20.405 --> 00:02:21.825
I'm urologist from Berlin.

71
00:02:22.106 --> 00:02:26.308
I work in a private practice and one day a week in a...

72
00:02:26.916 --> 00:02:42.548
in a hospital and i would say i did a lot of fusion biopsies thanks for this presentation so now let's talk about prostate first of all i have heard that you are a fervent advocate of the transpirineal approach am i right i

73
00:02:42.568 --> 00:02:52.156
would say yes we we do a lot of transpirineal biopsies and we start three years ago so i think we are really experienced with it okay

74
00:02:52.584 --> 00:02:57.407
And why did you choose to switch from transrectal to transperineal prostate biopsies?

75
00:02:57.987 --> 00:03:02.449
I think everybody knows when he's doing a lot of biopsies,

76
00:03:03.230 --> 00:03:03.830
unfortunately,

77
00:03:03.970 --> 00:03:05.011
the transrectal way,

78
00:03:05.731 --> 00:03:08.913
you have a lot of post-operative infections.

79
00:03:09.093 --> 00:03:09.333
And

80
00:03:09.873 --> 00:03:12.775
I did the biopsies on a Friday and on Monday,

81
00:03:12.775 --> 00:03:14.036
it was always the same.

82
00:03:14.156 --> 00:03:17.197
Two of my 15 biopsies had an infection.

83
00:03:17.637 --> 00:03:19.078
This was a point that I said,

84
00:03:19.258 --> 00:03:19.538
okay,

85
00:03:19.699 --> 00:03:22.180
let's stop with the transrectal way and we choose a more

86
00:03:22.672 --> 00:03:23.853
safe way for the patient.

87
00:03:24.554 --> 00:03:25.935
And you mentioned the infection.

88
00:03:26.055 --> 00:03:27.936
What kind of infection can happen?

89
00:03:29.177 --> 00:03:29.898
Little infection.

90
00:03:29.898 --> 00:03:32.620
It means just an urinary tract infection,

91
00:03:32.680 --> 00:03:35.343
but it could be a sepsis as well.

92
00:03:35.443 --> 00:03:39.686
And a sepsis is really a danger for the life of the patient.

93
00:03:39.866 --> 00:03:40.847
And this could be a big,

94
00:03:40.927 --> 00:03:41.488
big problem.

95
00:03:42.168 --> 00:03:42.428
Okay.

96
00:03:43.089 --> 00:03:44.090
And as for today,

97
00:03:44.250 --> 00:03:47.292
would you get back to transrectal approach sometimes?

98
00:03:47.452 --> 00:03:47.753
Never.

99
00:03:48.033 --> 00:03:48.353
Never.

100
00:03:49.094 --> 00:03:51.576
I'm sure it would help doctors who hesitate to change their

101
00:03:51.844 --> 00:04:13.660
practices something else come in my mind i have heard that you are performing more than 1000 biopsy a year is that true you are right really 1000 oh my god that's mean that you must prepare a lot of mri images right what is your process for this preparation because it's really impressive actually this is a

102
00:04:13.700 --> 00:04:19.584
big task to pre firm all of the biopsies of course but as well to prepare the mris because as

103
00:04:20.188 --> 00:04:20.768
Urologists,

104
00:04:20.788 --> 00:04:25.431
you have to deal with a really inhomogeneous quality of these MRIs.

105
00:04:25.992 --> 00:04:28.193
And I think for fusion biopsies,

106
00:04:28.573 --> 00:04:31.335
it's really essential to have a good

107
00:04:31.976 --> 00:04:35.258
MRI and puncture the right lesions in the MRI.

108
00:04:35.898 --> 00:04:40.581
And how do you be sure that the MRI that you will have from the patient is good?

109
00:04:40.761 --> 00:04:44.003
Do you work with a specific radiologist department at your hospital?

110
00:04:44.203 --> 00:04:44.924
What is the process?

111
00:04:45.556 --> 00:04:50.480
Actually in Berlin the situation is that we have a lot of radiological institutes.

112
00:04:50.660 --> 00:04:52.601
And this means there are three,

113
00:04:52.761 --> 00:04:58.766
four really good and high quality radiological institutes with a really high quality of MRIs.

114
00:04:59.466 --> 00:05:00.547
But the rest is...

115
00:05:01.039 --> 00:05:02.440
the quality is not really high.

116
00:05:02.480 --> 00:05:04.582
So you have to deal with this problem.

117
00:05:04.922 --> 00:05:05.642
And for this,

118
00:05:05.702 --> 00:05:10.205
you need a lot of experience or you use assisted software.

119
00:05:10.846 --> 00:05:11.426
Interesting.

120
00:05:11.947 --> 00:05:12.867
Doing my research,

121
00:05:13.047 --> 00:05:19.832
I have noticed that you have recently published a study called Diagnostic Utility of Artificial Intelligence Assisted

122
00:05:20.232 --> 00:05:23.555
Transparenal Biopsy Planning in Prostate Care Suspected Men.

123
00:05:23.975 --> 00:05:29.118
Could you explain us how artificial intelligence can help you in these procedures?

124
00:05:29.519 --> 00:05:29.879
Of course.

125
00:05:30.351 --> 00:05:30.591
Actually,

126
00:05:30.591 --> 00:05:34.695
the point is a multi-parametric MRI is a really big data set.

127
00:05:34.895 --> 00:05:37.377
You have to evaluate a lot of data,

128
00:05:37.517 --> 00:05:38.658
a lot of sequences.

129
00:05:38.898 --> 00:05:39.699
And for this,

130
00:05:39.899 --> 00:05:44.362
you have to be experienced in the evaluation of these MRIs.

131
00:05:44.543 --> 00:05:46.965
We all know there are a lot of information in the MRI.

132
00:05:47.045 --> 00:05:55.492
And we know that computer-aided diagnostic systems have the possibility to analyze such a big data set really fast.

133
00:05:55.752 --> 00:05:56.773
And maybe really,

134
00:05:57.433 --> 00:05:58.294
really sure,

135
00:05:58.354 --> 00:05:59.595
to be sure that...

136
00:05:59.815 --> 00:06:02.636
you find the right lesion in the prostate.

137
00:06:02.916 --> 00:06:08.979
And this was a start that we thought about the idea to use such a CAD software,

138
00:06:09.439 --> 00:06:12.320
which is artificial intelligence assisted,

139
00:06:13.060 --> 00:06:16.002
to analyze our MRIs before a biopsy.

140
00:06:16.562 --> 00:06:17.342
Very interesting.

141
00:06:17.642 --> 00:06:22.124
Does it impact the quality of the image in a good or maybe a bad way?

142
00:06:22.604 --> 00:06:22.844
Actually,

143
00:06:22.844 --> 00:06:26.366
when we look in the data we published in the European Neurology Focus,

144
00:06:26.406 --> 00:06:29.467
we can see that it helps to analyze the pictures.

145
00:06:29.707 --> 00:06:31.207
or the sequences of the MRI.

146
00:06:31.548 --> 00:06:37.749
What we found is that we are really safe to find all of the right indices,

147
00:06:38.049 --> 00:06:38.449
lesions,

148
00:06:39.090 --> 00:06:41.310
and we will find more cancer,

149
00:06:41.350 --> 00:06:42.250
additional cancer,

150
00:06:42.310 --> 00:06:45.591
maybe which is overlooked by the radiological report.

151
00:06:45.791 --> 00:06:51.573
We'll find other additional lesions in the prostate with prostate cancer,

152
00:06:51.853 --> 00:06:54.214
which was not described in a radiological report.

153
00:06:54.254 --> 00:06:59.315
So I think it's really an advantage for the biopsy to use such a software.

154
00:07:00.227 --> 00:07:00.487
Okay,

155
00:07:00.847 --> 00:07:03.609
and I'm thinking about the radiologist as well.

156
00:07:03.929 --> 00:07:07.271
What is the incidence on your collaboration with him?

157
00:07:08.572 --> 00:07:08.912
Actually,

158
00:07:09.352 --> 00:07:10.433
I'm really experienced.

159
00:07:10.433 --> 00:07:14.175
This means I don't have a really collaboration with a radiologist.

160
00:07:14.755 --> 00:07:18.097
I always analyze the sequences by myself,

161
00:07:18.157 --> 00:07:23.460
but I think this collaboration is really important because first of all,

162
00:07:23.520 --> 00:07:27.742
we need a really good and proper radiological report.

163
00:07:28.082 --> 00:07:28.923
On the other hand,

164
00:07:29.283 --> 00:07:29.403
the

165
00:07:29.863 --> 00:07:31.584
radiologists need the feedback,

166
00:07:31.724 --> 00:07:33.445
the histopathological feedback,

167
00:07:33.485 --> 00:07:35.186
because if they have this,

168
00:07:35.546 --> 00:07:39.488
they can evaluate their radiological report to say,

169
00:07:39.568 --> 00:07:39.788
okay,

170
00:07:39.808 --> 00:07:40.788
it was the right lesion,

171
00:07:40.828 --> 00:07:42.509
which I described in the report.

172
00:07:43.210 --> 00:07:44.730
So in another way,

173
00:07:44.830 --> 00:07:46.851
it's like a teamwork with your radiologist.

174
00:07:46.891 --> 00:07:47.272
Of course.

175
00:07:47.652 --> 00:07:47.872
Okay.

176
00:07:48.292 --> 00:07:50.933
And how does it impact your own daily work?

177
00:07:51.734 --> 00:07:52.654
This new software,

178
00:07:52.714 --> 00:07:53.775
this new technology?

179
00:07:54.775 --> 00:08:00.758
I would say artificial intelligence is a really new word and everybody is interesting in it.

180
00:08:01.479 --> 00:08:01.859
Of course,

181
00:08:01.899 --> 00:08:03.060
it's a promotion fact.

182
00:08:03.600 --> 00:08:04.941
Patients really like to say,

183
00:08:05.561 --> 00:08:07.402
I use AI for your biopsy.

184
00:08:08.242 --> 00:08:10.143
And for my medical life,

185
00:08:10.163 --> 00:08:17.427
I would say it makes the biopsy much better because I'm more sure where to take the targeted biopsies.

186
00:08:17.828 --> 00:08:23.891
And I feel more comfortable with it because I know that I will not overlook.

187
00:08:24.111 --> 00:08:25.531
any lesion in the prostate.

188
00:08:26.452 --> 00:08:26.712
Okay,

189
00:08:27.272 --> 00:08:27.992
very interesting.

190
00:08:28.292 --> 00:08:32.473
And how do you see prostate care management in the next few years?

191
00:08:32.974 --> 00:08:34.374
From diagnosis to treatment,

192
00:08:34.454 --> 00:08:36.415
how artificial intelligence,

193
00:08:36.555 --> 00:08:36.835
AI,

194
00:08:37.195 --> 00:08:40.296
or just technology will impact your practice in the future?

195
00:08:41.516 --> 00:08:42.436
I think the point is,

196
00:08:42.797 --> 00:08:44.517
when we talk about prostate cancer,

197
00:08:44.537 --> 00:08:46.838
we know there is a really specialized,

198
00:08:47.118 --> 00:08:49.959
or there are a lot of specialized therapy modalities.

199
00:08:50.459 --> 00:08:50.939
For this,

200
00:08:51.303 --> 00:08:53.924
We need a really accurate diagnostic.

201
00:08:54.184 --> 00:08:57.465
We should use all of the assistance programs,

202
00:08:57.545 --> 00:08:58.025
softwares,

203
00:08:58.125 --> 00:09:00.626
everything to make such a diagnostic,

204
00:09:00.646 --> 00:09:02.426
the initial diagnostic,

205
00:09:02.826 --> 00:09:03.946
as proper as we can.

206
00:09:04.327 --> 00:09:04.447
So

207
00:09:04.967 --> 00:09:08.288
I would say with a really accurate diagnostic,

208
00:09:08.508 --> 00:09:16.250
you can give the patient the best therapy modality to save his life or to make his life better.

209
00:09:16.790 --> 00:09:18.891
So the heart of the matter is the diagnostic.

210
00:09:19.251 --> 00:09:19.931
Of course,

211
00:09:19.951 --> 00:09:20.331
I would say.

212
00:09:20.655 --> 00:09:23.157
And it also benefits the patient at the end.

213
00:09:23.418 --> 00:09:23.818
Of course.

214
00:09:24.338 --> 00:09:26.380
Because we are more accurate with it.

215
00:09:26.761 --> 00:09:37.350
So it's better for the patient because he can be sure that we will find the cancer and we will find the cancer maybe with the highest grade,

216
00:09:37.750 --> 00:09:38.911
what is really important.

217
00:09:39.412 --> 00:09:39.892
Definitely.

218
00:09:40.433 --> 00:09:41.614
Before ending this interview,

219
00:09:42.114 --> 00:09:48.440
what would you tell other physicians to convince them to use or at least not to fear artificial intelligence?

220
00:09:49.076 --> 00:09:53.879
I would say there is still a little bit of distance to use an MP

221
00:09:54.220 --> 00:09:55.480
MRI for urologists.

222
00:09:55.480 --> 00:10:00.664
There are not really a lot of urologists who like to evaluate the MRI.

223
00:10:01.264 --> 00:10:03.446
With such an assistance program,

224
00:10:03.686 --> 00:10:05.207
which is AI-assisted,

225
00:10:05.467 --> 00:10:08.589
you are able to evaluate the MRI by yourself.

226
00:10:08.689 --> 00:10:11.471
And it's a training way,

227
00:10:11.591 --> 00:10:12.112
of course,

228
00:10:12.372 --> 00:10:13.192
but you have to do it.

229
00:10:13.252 --> 00:10:15.814
And you will see when you get the assistance,

230
00:10:16.074 --> 00:10:16.695
for example,

231
00:10:17.075 --> 00:10:17.975
by a CAD software,

232
00:10:17.975 --> 00:10:18.796
there are a lot of people who are using it.

233
00:10:19.144 --> 00:10:24.726
you will get a really good start to start to evaluate your own MRIs.

234
00:10:25.727 --> 00:10:27.808
Thanks a lot for these precious informations.

235
00:10:28.428 --> 00:10:30.209
Talking about the future now,

236
00:10:30.489 --> 00:10:32.530
what are you currently working on?

237
00:10:33.150 --> 00:10:33.850
What are your plans?

238
00:10:34.370 --> 00:10:34.550
Yeah,

239
00:10:34.831 --> 00:10:38.932
I think we are really at the beginning to use AI for prostate biopsy.

240
00:10:38.972 --> 00:10:41.973
I think there are not a lot of institutes who are doing this.

241
00:10:42.074 --> 00:10:44.715
So maybe this is something,

242
00:10:45.055 --> 00:10:45.515
it's always,

243
00:10:45.815 --> 00:10:47.856
it will take time until everybody would say.

244
00:10:48.284 --> 00:10:48.884
Five years ago,

245
00:10:49.044 --> 00:10:52.166
everybody was doing a systematic transrectal biopsy.

246
00:10:52.446 --> 00:10:54.667
Then there was a shift to the transpirineal

247
00:10:55.187 --> 00:10:56.347
MRI fusion biopsy.

248
00:10:56.487 --> 00:11:03.170
And now we talk about the transpirineal AI-assisted MRI fusion biopsy in local anesthesia.

249
00:11:03.510 --> 00:11:09.953
I think there is a really fast development and it will take the next years to implement this in the daily work.

250
00:11:10.373 --> 00:11:10.633
Okay,

251
00:11:10.813 --> 00:11:13.955
so let's meet next year to have a quick update.

252
00:11:14.675 --> 00:11:15.415
I'm looking forward.

253
00:11:15.936 --> 00:11:16.756
Sounds exciting.

254
00:11:17.136 --> 00:11:18.137
Before leaving Dr.

255
00:11:18.197 --> 00:11:18.657
Carsten,

256
00:11:18.918 --> 00:11:21.440
I'd like to ask my guest a surprise question.

257
00:11:21.660 --> 00:11:21.840
Yes,

258
00:11:21.880 --> 00:11:22.120
I know.

259
00:11:22.941 --> 00:11:23.061
So,

260
00:11:23.662 --> 00:11:25.723
what is your song of the moment?

261
00:11:26.404 --> 00:11:27.545
The song of the moment?

262
00:11:28.566 --> 00:11:29.727
Really difficult question.

263
00:11:32.049 --> 00:11:33.070
I would say it is

264
00:11:33.490 --> 00:11:35.051
Leonard Skinner with Simple Man.

265
00:11:35.412 --> 00:11:35.592
Okay,

266
00:11:35.632 --> 00:11:35.892
great.

267
00:11:35.912 --> 00:11:36.032
So,

268
00:11:36.653 --> 00:11:37.113
perfect.

269
00:11:37.854 --> 00:11:41.136
Thanks again for your time and with precious and your highlight.

270
00:11:41.196 --> 00:11:45.240
I hope you enjoyed this moment with us and maybe you will come back soon.

271
00:11:45.660 --> 00:11:47.401
It was a great pleasure to talk with you.

272
00:11:47.561 --> 00:11:48.181
Thank you very much.

273
00:11:48.481 --> 00:11:48.861
Thank you.

274
00:11:53.403 --> 00:11:55.904
Huge thanks to our dedicated listeners.

275
00:11:56.365 --> 00:11:57.925
If you enjoyed the podcast,

276
00:11:58.105 --> 00:12:00.746
make sure to subscribe for more captivating content.

277
00:12:01.327 --> 00:12:03.227
Your support means the world.

278
00:12:04.928 --> 00:12:06.569
If you found this informative,

279
00:12:06.649 --> 00:12:07.369
a quick favor,

280
00:12:07.429 --> 00:12:07.749
please.

281
00:12:08.070 --> 00:12:09.370
Drop up a five-star review.

282
00:12:09.830 --> 00:12:13.292
Your feedback helps us grow and deliver quality insight.

283
00:12:14.192 --> 00:12:15.794
Have topics you'd love us to cover?

284
00:12:16.215 --> 00:12:19.599
Share your ideas in the comments or connect with us on social media.

285
00:12:20.220 --> 00:12:23.063
Your input guides the future of Pros That Talk.

286
00:12:23.544 --> 00:12:26.328
Thanks for being a crucial part of our community.

287
00:12:27.249 --> 00:12:28.871
For more urology insights,

288
00:12:29.211 --> 00:12:30.353
visit Kallis.com.

289
00:12:30.753 --> 00:12:32.856
Stay tuned and see you next time.