AI is transforming the practice of medicine in many specialties, and deep learning can help achieve better and earlier problem detection while reducing errors on diagnosis. This article demonstrates how a deep neural network (DNN) can be used to improve the precision and accuracy of temperature readings from a low-cost and low-accuracy sensor array. The array is composed of 32 temperature sensors, including 16 analog and 16 digital sensors. The DNN model is trained with a randomly-selected dataset using 640 vectors (80% of the data) and tested with 160 vectors (20%). The mean squared error as a loss function between the data and the model’s prediction is 1.47×10 on the training set and 1.22×10 on the test set, showing that this approach offers a new pathway towards significantly better datasets using ultra low-cost sensors.
Add to Favourites
To login click here