PREMIERE Seminar Imperial College London 2023 

In this presentation, we delve into the dynamic synergy between Machine Learning and physics-driven approaches, addressing critical challenges in engineering and healthcare. From fracture mechanics and predictive maintenance in industry to personalized medical diagnostics and cancer therapy, we explore how these innovative techniques are reshaping the landscape of problem-solving. Join me on this journey as we unlock the immense potential of data-driven science to advance our understanding and transform industries.
Table of Contents:
00:03 -
01:47 - Three ideas
02:30 - Legato-team Luxembourg
03:00 -
03:50 - « Nous sommes si éloignés de connaître tous les agens [orthographe de l'époque] de la nature, et leurs divers modes d'action qu'il ne serait pas philosophique de nier les phénomènes, uniquement parce qu'ils sont inexplicables dans l'état actuel de nos connaissances. Seulement, nous devons les examiner avec une attention d'autant plus scrupuleuse, qu'il paraît plus difficile de les admettre ; et c'est ici que le calcul des probabilités devient indispensable, pour déterminer jusqu'à quel point il faut multiplier les observations ou les expériences, afin d'obtenir en faveur des agens qu'elles indiquent, une probabilité supérieure aux raisons que l'on peut avoir d'ailleurs, de ne pas les admettre. »
- Pierre-Simon de Laplace
03:52 - "We are so far from knowing all the agents of nature and their various modes of action that it would not be philosophical to deny phenomena solely because they are inexplicable in the current state of our knowledge. Instead, we must examine them with even greater scrutiny, the more difficult they appear to admit. This is where the calculation of probabilities becomes essential, to determine to what extent we should multiply observations or experiments in order to obtain, in favor of the agents they indicate, a probability greater than the reasons we may otherwise have for not admitting them."
- Pierre-Simon de Laplace
05:55 -
06:17 - Why is this important?
06:47 - Why is this important?
07:19 - Why is it important?
08:07 - Why is it important?
08:08 -
09:13 -
09:41 - Quantify the quality of the simulation
12:14 - Quantify the quality of the simulation
12:38 - Quantify the quality of the simulation
13:18 - First idea
14:01 -
15:14 -
15:22 -
16:09 -
16:37 -
16:44 -
17:08 -
17:11 -
17:14 -
17:24 -
17:24 -
17:39 -
17:43 -
17:52 -
18:00 -
18:03 -
18:05 -
18:17 -
19:04 -
19:08 -
19:30 -
19:50 -
20:07 -
20:24 -
20:27 -
20:55 -
21:08 -
21:12 -
21:17 -
21:22 -
22:42 -
22:54 -
22:55 - A baby is a scientist in the crib (Gopnik)
25:53 - The bayesian brain
26:13 -
27:35 -
27:38 -
27:47 -
28:04 -
28:12 -
28:15 -
28:18 -
28:25 -
28:44 -
29:10 -
29:12 - The bayesian brain and the bayesian computer
29:13 - Second idea
29:35 -
30:50 -
32:07 -
35:37 -
35:38 - Interlude
35:42 -
37:39 -
37:43 - inCERT: Computer assisted surgery with confidence
43:13 -
43:17 -
43:38 - How important is astrocyte morphology for neural metabolism?
43:39 -
43:40 -
43:40 -
43:44 - How important is astrocyte morphology for neural metabolism?
44:03 - Metabolism and calcium signalling cross-talk in spatially resolved cellular domains
44:03 - Introduction I Astrocytes
44:32 - Introduction I Astrocytes as Metabolic Mediator
44:57 - Introduction I Reactive Astrocytes
45:25 - Model I Metabolic Model
46:26 - Model I Metabolism and Calcium Model
46:28 - Results I Geometries and  dynamics impact on metabolites
46:30 - Results I Cytosolic  waves and geometrical impact
46:30 -
46:36 - HELP SURGEONS NAVIGATE
46:46 - European Investment Bank Conference 2019
48:01 - Breast conserving surgical pipeline
49:21 - Breast conserving surgical pipeline
50:17 - Inverse deformation: motivation
50:19 - Preoperative surgical drawing: outcome
50:21 - A new breast model pipeline
50:23 - A new breast model pipeline
50:30 - A new breast model pipeline
50:38 - A new breast model pipeline
50:47 - A new breast model pipeline
51:23 - A new breast model: results
51:54 - A new breast model: results
51:55 - SOniCS: SOFA and FEniCS
53:34 - SOniCS: intuitive material model implementation
53:35 - SOniCS: intuitive material model implementation
53:36 - SOniCS: intuitive material model implementation
53:37 - SOniCS: intuitive material model implementation
53:38 - SOniCS: intuitive material model implementation
53:39 - SOniCS: intuitive material model implementation
53:39 - SOniCS: intuitive material model implementation
53:40 - SOniCS: results
53:58 - Model selection and error control in biomechanics
54:57 - HELP SURGEONS NAVIGATE
55:03 -
55:05 - ADONIS: Archaeological Digital Forensics
55:15 - CASE STUDY: Tell Nebi Yunus, Mosul, Iraq
55:34 - CASE STUDY: Tell Nebi Yunus, Mosul, Iraq
55:39 - CASE STUDY: Tell Nebi Yunus, Mosul, Iraq
59:20 - SCAN TO SIM
59:58 - SCAN TO SIM

Хобби