Inspiration
Many times doctors need to predict the state of the fetus inside the womb based on the diagnosis results they get when examining the mother. Doing this task continuously will make doctors' work tedious. So we thought why don't we import what the doctor is doing into the software?
What it does
Our plan is to create a web app that will allow doctors to understand whether the condition of the baby inside is Normal(which is ok), Suspect(something should be taken care of), or Pathological(something serious).
How we built it
So the technologies we used are Tensorflow, TensorflowJS, Keras, KerasTuner, and ReactJS.
Challenges we ran into
We were so confused to determine the number of nodes and layers we will need for our Tensorflow model. So we used a Hyperparameter optimization framework called Keras Tuner that helps us to determine how nodes and layers we should use in our model.
Accomplishments that we're proud of
We were successfully able to get good accuracy and were able to create a web app out of it.
What we learned
We learned how to implement Hyper Parameter Optimization framework which helps us to determine how many nodes and layers we need to implement.
What's next for Fetal State Predictor
If we train this model on even much bigger data set we can get a model of even much bigger accuracy and then we can implement this model for industry level( like in hospital).
Log in or sign up for Devpost to join the conversation.