Advancing towards Modeling 2.0 with neural-ODE via Wolfram Language

Staff Picks
, , , , , , , ,
1

SUPPLEMENTARY WOLFRAM MATERIALS for the ARTICLE:

Lu, J., Bender, B., Jin, J.Y. et al. (2021).

Deep learning prediction of patient response time course from early data via neural-pharmacokinetic/pharmacodynamic modelling.

NATURE Machine Intelligence 3, 696–704.

Deep learning prediction of patient response time course from early data via neural-pharmacokinetic/pharmacodynamic modelling | Nature Machine Intelligence

Full article PDF

NeuralPKPD_simulation_in_action.gif

Video

NeuralPKPD_prediction_update_in_action.gif

Video

Open in Wolfram Cloud ↗
16 Likes
2

news-congrads-kkluyshnik-02-04-19.jpg – you have earned Featured Contributor Badge 2015-11-02_10-07-11.png
Your exceptional post has been selected for our editorial column Staff Picks http://wolfr.am/StaffPicks and Your Profile is now distinguished by a Featured Contributor Badge and is displayed on the Featured Contributor Board. Thank you!

1 Like
3

Hi,
Thanks for the great work. I am following your nb and I have error messages at the following commands. I appreciate your help. I am guessing these are input data.

n = 10;
trainedPK = NetTrain[combinedPKNet,<|"PKPDData" -> Table[trainPKPDData, n],
   "Target" -> Table[inputPKObs, n],"Weight" -> Table[inputWeight, n] ,"InitialState" -> Table[initState, n],"Dose" -> Table[inputDosing, n]|>, LossFunction -> "Loss", ValidationSet -> None, MaxTrainingRounds -> 10,  Method -> {"ADAM", "LearningRate" -> 10^-4,  "GradientClipping" -> 1,  "L2Regularization" -> 0.1} ] 

NetTrain::invindim3: Data provided to port "Dose" should be a non-empty list of n*1 matrices, but was inputDosing.

Sangdon Lee

1 Like