English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

${\bf Purpose}$: Earlier work showed that IVIM-NET$_{orig}$, an unsupervisedphysics-informed deep neural network, was more accurate than otherstate-of-the-art intravoxel-incoherent motion (IVIM) fitting approaches to DWI.This study presents an improved version: IVIM-NET$_{optim}$, and characterizesits superior performance in pancreatic ductal adenocarcinoma (PDAC) patients.${\bf Method}$: In simulations (SNR=20), the accuracy, independence andconsistency of IVIM-NET were evaluated for combinations of hyperparameters (fitS0, constraints, network architecture, # hidden layers, dropout, batchnormalization, learning rate), by calculating the NRMSE, Spearman's $\rho$, andthe coefficient of variation (CV$_{NET}$), respectively. The best performingnetwork, IVIM-NET$_{optim}$ was compared to least squares (LS) and a Bayesianapproach at different SNRs. IVIM-NET$_{optim}$'s performance was evaluated in23 PDAC patients. 14 of the patients received no treatment between scansessions and 9 received chemoradiotherapy between sessions. Intersessionwithin-subject standard deviations (wSD) and treatment-induced changes wereassessed. ${\bf Results}$: In simulations, IVIM-NET$_{optim}$ outperformedIVIM-NET$_{orig}$ in accuracy (NRMSE(D)=0.18 vs 0.20; NMRSE(f)=0.22 vs 0.27;NMRSE(D*)=0.39 vs 0.39), independence ($\rho$(D*,f)=0.22 vs 0.74) andconsistency (CV$_{NET}$ (D)=0.01 vs 0.10; CV$_{NET}$ (f)=0.02 vs 0.05;CV$_{NET}$ (D*)=0.04 vs 0.11). IVIM-NET$_{optim}$ showed superior performanceto the LS and Bayesian approaches at SNRs<50. In vivo, IVIM-NET$_{optim}$sshowed significantly less noisy parameter maps with lower wSD for D and f thanthe alternatives. In the treated cohort, IVIM-NET$_{optim}$ detected the mostindividual patients with significant parameter changes compared to day-to-dayvariations. ${\bf Conclusion}$: IVIM-NET$_{optim}$ is recommended for IVIMfitting to DWI data.

Improved unsupervised physics-informed deep learning for intravoxel incoherent motion modeling and evaluation in pancreatic cancer patients