Introduction to Deep Learning in Clinical Neuroscience (accepted)
Kapitel i bok, 2021
Important aspects are data pre-processing, volumetric segmentation, and specific task-performing DL methods, such as CNNs and AEs. Additionally, GAN-expansion and domain mapping are useful DL techniques for generating artificial data and combining several smaller datasets.
We present results of DL-based segmentation and accuracy in predicting glioma subtypes based on MRI features. Dice scores range from 0.77-0.89. In mixed glioma cohorts, IDH mutation can be predicted with a sensitivity of 0.98 and specificity of 0.97. Results in test cohorts have shown improvements of 5-7% in accuracy, following GAN-expansion of data and domain mapping of smaller datasets.
The provided DL examples are promising, although not yet in clinical practice. DL has demonstrated usefulness in data augmentation and for overcoming data variability. DL methods should be further studied, developed and validated for broader clinical use. Ultimately, DL models can serve as effective decision support systems, and are especially well-suited for time-consuming, detail-focused, and data-ample tasks.
prognosis
outcome
glioma
Deep learning
prediction
Författare
Eddie de Dios
Sahlgrenska universitetssjukhuset
Muhaddisa Barat Ali
Chalmers, Elektroteknik, Signalbehandling och medicinsk teknik, Signalbehandling
Irene Yu-Hua Gu
Chalmers, Elektroteknik
Tomas Gomez Vecchio
Sahlgrenska universitetssjukhuset
Göteborgs universitet
Chenjie Ge
Chalmers, Elektroteknik, Signalbehandling och medicinsk teknik, Signalbehandling
Asgeir Store Jakola
Göteborgs universitet
Sahlgrenska universitetssjukhuset
Machine Learning in Clinical Neuroscience: Foundations and Applications (to appear, in Springer Nature book)
Ämneskategorier
Annan data- och informationsvetenskap
Klinisk medicin
Bioinformatik (beräkningsbiologi)
Datavetenskap (datalogi)
Styrkeområden
Hälsa och teknik