BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//TU/e ME//Events//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
BEGIN:VTIMEZONE
TZID:Europe/Amsterdam
BEGIN:STANDARD
DTSTART:19701025T030000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:19700329T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:event-15@tuemeche.nl
DTSTAMP:20261008T003556Z
DTSTART;TZID=Europe/Amsterdam:20260903T110000
DTEND;TZID=Europe/Amsterdam:20260903T123000
SUMMARY:Learning-Based Calibration\, Estimation and Stage Control for Ele
 ctron Microscopy
DESCRIPTION:Speaker: Jilles van Hulst\nHost: Duarte Guerreiro Tomé Antun
 es\n\nHow much can you learn from a handful of noisy images? For electron
  microscopes\, the answer decides how sharp their images get and how stea
 dily they can hold the object under study. An electron microscope can pho
 tograph individual atoms\, an ability indispensable for new medicines\, b
 etter batteries\, and the chips in every phone. In the COVID-19 pandemic\
 , electron microscopes revealed the structure of the spike protein within
  weeks and accelerated vaccine development. Yet these instruments cannot 
 focus themselves. Keeping them sharp is the daily work of scarce\, highly
  trained operators. The PhD research of Jilles van Hulst\, at the Eindhov
 en University of Technology with microscope manufacturer Thermo Fisher Sc
 ientific\, teaches the machines to do this themselves.\n\nAtomic resoluti
 on\nNo microscope can show details smaller than the wavelength it uses\, 
 and visible light is thousands of times wider than an atom. Electrons off
 er a way out: a fast electron behaves as a quantum wave far smaller than 
 an atom. If a light wave were stretched to the length of a football field
 \, the electron wave would span just a few sheets of paper. Electrons are
  also electrically charged\, so magnetic fields can act as lenses for the
 m. However\, a theorem from 1936 proves that these lenses always distort 
 the image. Correctors have existed since 1998\, but they add dozens of se
 ttings that drift and need constant retuning.\n\nLearning from a handful 
 of images\nRetuning these settings is hard\, and anyone who has focused a
 n old camera by hand knows why. Looking at a single blurry photo\, you ca
 nnot tell which way to turn the focus ring. Only once you turn it do you 
 see where you were\, and perfect focus is precisely the point where turni
 ng either way makes it worse. An electron microscope poses this puzzle no
 t for one focus ring but for dozens of settings at once\, and every image
  costs time and can damage the specimen. Van Hulst therefore trained arti
 ficial intelligence (AI) to get the most out of each image. The AI learns
  from a simulator\, which can generate unlimited practice images\, but a 
 simulation is never exactly the real machine. So the full method also lea
 rns that difference during operation\, from the few real images it collec
 ts. The symmetry of the blur around perfect focus is what pins that point
  down uniquely. The result is a microscope that calibrates itself in abou
 t twenty seconds\, twice as accurately as the best automated methods.\n\n
 Standing still at the scale of atoms\nSeeing atoms also demands stillness
 . For a three-dimensional image\, the specimen is tilted step by step whi
 le dozens of pictures are taken\, and the point under study must stay in 
 view. The stage carrying the specimen can move across millimeters\, yet m
 ust hold the imaged point to within nanometers. The actuators that move t
 he stage are imperfect\, and no sensor measures where the specimen actual
 ly is. Van Hulst used the images that the microscope already records as t
 he missing sensor. They reveal exactly how far the specimen has shifted\,
  and a learning controller removes errors it has seen before. On an opera
 tional microscope this reached nanometer accuracy\, up to ten times bette
 r than the existing approach.\n\nMore knowledge from less data\nThe commo
 n thread is a lesson from statistics: what you can learn from data depend
 s on what you already know. Van Hulst therefore developed general mathema
 tical methods that build known structure\, such as physical laws\, symmet
 ries\, and models of motion\, into the learning itself\, so that fewer me
 asurements suffice. Together\, these results bring the self-driving micro
 scope closer. It tunes itself\, holds its specimen steady\, and collects 
 data on its own\, so that researchers can focus on the discoveries.\n\nMo
 re info: https://tuenl.sharepoint.com/sites/intranet-mechanical-engineeri
 ng/_layouts/15/Event.aspx?ListGuid=9bfaaae6-070c-4371-810d-a43d7ee02bf2&I
 temId=241
LOCATION:Atlas 0.710
URL:https://tuemeche.nl/peoplepages/event.php?id=15
CATEGORIES:PhD Defense
END:VEVENT
END:VCALENDAR