4D Deep Learning for Multiple Sclerosis Lesion Activity Segmentation - - PowerPoint PPT Presentation

▶

Dec 19, 2023 177 likes •249 views

Nils Gessert 1 , Marcel Bengs 1 , Julia Krger 2 , Roland Opfer 2 , Ann-Christin Ostwaldt 2 , Praveena Manogaran 3 , Sven Schippling 3 , Alexander Schlaefer 1 Institute of Medical Technology and Intelligent Systems 4D Deep Learning for Multiple

SLIDE 1

Institute of Medical Technology and Intelligent Systems

Nils Gessert1, Marcel Bengs1, Julia Krüger2, Roland Opfer2, Ann-Christin Ostwaldt2, Praveena Manogaran3, Sven Schippling3, Alexander Schlaefer1

4D Deep Learning for Multiple Sclerosis Lesion Activity Segmentation

1Institute of Medical Technology and Intelligent Systems, Hamburg University of Technology 2jung diagnostics GmbH 3Department of Neurology, University Hospital Zurich and University of Zurich

SLIDE 2

July 2020 Slide 2

Lesion Activity Segmentation

Baseline MRI Scan Follow-up MRI Scan

Old Lesion Enlarged Lesion

Lesion Activity

Enlargement

SLIDE 3

July 2020 Slide 3

Two-Path 3D Encoder-Decoder

Baseline Follow-Up Follow-Up & Prediction

Krüger, Julia, et al. "Fully automated longitudinal segmentation of new or enlarging Multiple Scleroses (MS) lesions using 3D convolution neural networks."

History

SLIDE 4

July 2020 Slide 4

Encoder-convGRU-Decoder

Dataset:

44 MS cases, three time points each
FLAIR image volumes, varying size

SLIDE 5

July 2020 Slide 5

Results and Discussion

56 57 58 59 60 61 62 63 64 65 Dice Enc-Dec T=2 Enc-Dec T=3 Enc-cGRU-Dec T=2 Enc-cGRU-Dec T=3 5 10 15 20 25 30 35 40 LFPR Enc-Dec T=2 Enc-Dec T=3 Enc-cGRU-Dec T=2 Enc-cGRU-Dec T=3 79,5 80 80,5 81 81,5 82 82,5 83 83,5 84 84,5 LTPR Enc-Dec T=2 Enc-Dec T=3 Enc-cGRU-Dec T=2 Enc-cGRU-Dec T=3