BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//CERN//INDICO//EN
BEGIN:VEVENT
SUMMARY:Quantum Extreme Reservoir Computing for brain tumour MRI classific
 ation
DTSTART;VALUE=DATE-TIME:20260908T135000Z
DTEND;VALUE=DATE-TIME:20260908T142000Z
DTSTAMP;VALUE=DATE-TIME:20260922T191401Z
UID:indico-contribution-2026@events.ncbj.gov.pl
DESCRIPTION:Speakers: Chaimae El Bouazizi (Okinawa Institute of Science an
 d Technology)\nMedical imaging provides a demanding test for quantum machi
 ne learning because clinically relevant images are much larger and structu
 rally richer than standard benchmarks (MNIST). In this work\, we study qua
 ntum extreme reservoir computing (QERC) for binary brain-tumour classifica
 tion using the Br35H MRI dataset\, with images resized to 150 × 150 pixel
 s. Since such inputs cannot be encoded directly into a small quantum reser
 voir\, we compare principal component analysis with two autoencoder-based 
 compression schemes.\n\nThe results show that reconstruction quality alone
  is not a reliable guide to downstream quantum classification. An autoenco
 der can reproduce the images well while mapping them into a narrow region 
 of latent space\, thereby supplying the quantum encoder with insufficientl
 y varied inputs. We quantify this concentration using the participation ra
 tio and find that broader latent distributions generally produce better an
 d more stable classification. Batch normalization improves performance by 
 spreading the latent variables more evenly before quantum encoding.\n\nThe
 se results show that the compression method is not simply a preprocessing 
 choice. For this type of quantum image-classification pipeline\, it must b
 oth retain useful information and generate latent variables that can be en
 coded effectively by the quantum reservoir.\n\nhttps://events.ncbj.gov.pl/
 event/468/contributions/2026/
LOCATION:
URL:https://events.ncbj.gov.pl/event/468/contributions/2026/
END:VEVENT
END:VCALENDAR