둜그인

The autoPET3 Challenge -- Automated Lesion Segmentation in Whole-Body PET/CT - Multitracer Multicenter Generalization

μž‘μ„±μž
  • Haebom
μΉ΄ν…Œκ³ λ¦¬
λΉ„μ–΄ 있음

μ €μž

Jakob Dexl, Katharina Jeblick, Andreas Mittermeier, Balthasar Schachtner, Anna Theresa Stuber, Johanna Topalis, Maximilian Rokuss, Fabian Isensee, Klaus H. Maier-Hein, Hamza Kalisch, Jens Kleesiek, Constantin M. Seibold, Hussain Alasmawi, Lap Yan Lennon Chan, Yixuan Yuan, Alexander Jaus, Rainer Stiefelhagen, Pauline Ornela Megne Choudja, Konstantin Nikolaou, Christian La Fougere, Sergios Gatidis, Matthias P. Fabritius, Maurice Heimer, Gizem Abaci, Lalith Kumar Shiyam Sundar, Rudolf A. Werner, Jens Ricke, Clemens C. Cyran, Thomas Kustner, Michael Ingrisch

πŸ’‘ κ°œμš”

λ³Έ μ—°κ΅¬λŠ” μ „μ‹  PET/CT μ˜μƒμ—μ„œ 병변 μžλ™ λΆ„ν•  μ„±λŠ₯을 ν‰κ°€ν•˜κΈ° μœ„ν•΄ autoPET3 μ±Œλ¦°μ§€λ₯Ό μ„€κ³„ν•˜κ³  κ²°κ³Όλ₯Ό λ³΄κ³ ν•©λ‹ˆλ‹€. μ±Œλ¦°μ§€λŠ” λ‹€μ–‘ν•œ μΆ”μ μž 및 μ„Όν„° 쑰합에 λŒ€ν•œ μΌλ°˜ν™” μ„±λŠ₯을 μ€‘μ μ μœΌλ‘œ λ‹€λ£¨μ—ˆμœΌλ©°, 1,014개의 [18F]-FDG PET/CT 및 597개의 [18F]/[68Ga]-PSMA PET/CT 연ꡬλ₯Ό ν¬ν•¨ν•˜λŠ” λŒ€κ·œλͺ¨ 데이터셋을 ν™œμš©ν–ˆμŠ΅λ‹ˆλ‹€. μ΅œμƒμœ„ μ•Œκ³ λ¦¬μ¦˜μ€ λŒ€μ‘°κ΅° λŒ€λΉ„ DSCλ₯Ό 8% ν–₯μƒμ‹œν‚€κ³  μ˜€μŒμ„± λΆ€ν”Όλ₯Ό 5mL κ°μ†Œμ‹œν‚€λŠ” μ„±κ³Όλ₯Ό λ³΄μ˜€μŠ΅λ‹ˆλ‹€.

πŸ”‘ μ‹œμ‚¬μ  및 ν•œκ³„

β€’
ν›ˆλ ¨ 데이터 λ²”μœ„ λ‚΄μ—μ„œμ˜ PET/CT 병변 뢄할은 이미 νŒλ… μˆ˜μ€€μ— κ·Όμ ‘ν•œ μ„±λŠ₯을 λ‹¬μ„±ν–ˆμŠ΅λ‹ˆλ‹€.
β€’
보지 λͺ»ν•œ μΆ”μ μž-μ„Όν„° 쑰합에 λŒ€ν•œ ꡬ성 μΌλ°˜ν™”λŠ” μ—¬μ „νžˆ μ–΄λ €μš΄ 문제이며, 주둜 체적 κ³ΌλŒ€ μΆ”μ •μœΌλ‘œ 인해 λ°œμƒν•©λ‹ˆλ‹€.
β€’
μ•Œκ³ λ¦¬μ¦˜ μ„ νƒλ³΄λ‹€λŠ” λ°μ΄ν„°μ˜ μ΄μ§ˆμ„±κ³Ό μΌ€μ΄μŠ€μ˜ λ‚œμ΄λ„κ°€ μ„±λŠ₯ νŽΈμ°¨μ— 더 큰 영ν–₯을 λ―ΈμΉ©λ‹ˆλ‹€.
β€’
보지 λͺ»ν•œ 데이터에 λŒ€ν•œ 체적 κ³ΌλŒ€ 좔정을 μ€„μ΄λŠ” 것이 ν–₯ν›„ μ—°κ΅¬μ˜ μ€‘μš”ν•œ κ³Όμ œμž…λ‹ˆλ‹€.
πŸ‘
Slashpage둜 μ œμž‘λ¨