Disease Evolution Predictor Deep Neural Networks

Multiple recent studies have shown that WMH on a patient may decrease (i.e., shrink/regress), stay the same (i.e., stable), or increase (i.e., grow/progress) over a period of time. In this project, we refer to theses changes as evolution of WMH and propose deep neural network models to predict and delineate the evolution of WMH automatically. In this porject, we primarily used irregularity map (IM) and generative adversarial network (GAN) for our proposed Disease Evolution Predictor (DEP) model.

Code repository:

• DEP-GAN using Irregularity Map (IM)

Published publications:

• Rachmadi, M. F., Valdés-Hernández, M. D. C., Makin, S., Wardlaw, J. M., & Komura, T. (2019). Predicting the Evolution of White Matter Hyperintensities in Brain MRI using Generative Adversarial Networks and Irregularity Map. In: Shen D. et al. (eds) Medical Image Computing and Computer Assisted Intervention – MICCAI 2019. MICCAI 2019. Lecture Notes in Computer Science, vol 11766. Springer, Cham. doi: doi.org/10.1007/978-3-030-32248-9_17
• Rachmadi, M. F., Valdés-Hernández, M. D. C., & Komura, T. (2018, September). Transfer Learning for Task Adaptation of Brain Lesion Assessment and Prediction of Brain Abnormalities Progression/Regression using Irregularity Age Map in Brain MRI. In International Workshop on PRedictive Intelligence In MEdicine (pp. 85-93). Springer, Cham. doi: 10.1007/978-3-030-00320-3_11

Pre-print:

• Rachmadi, M. F., Valdés-Hernández, M. D. C., Makin, S., Wardlaw, J. M., & Komura, T. (2019). Automatic Spatial Estimation of White Matter Hyperintensities Evolution in Brain MRI using Disease Evolution Predictor Deep Neural Networks. bioRxiv, 738641. pre-print doi: doi.org/10.1101/738641

I have also shot-listed relevant papers in MICCAI 2019 dealing with similar problem on the following blog post.