Department of Electrical Engineering, École de technologie supérieure, Montréal, H3C 1K3, Canada.
Sci Rep. 2023 Jul 11;13(1):11237. doi: 10.1038/s41598-023-38290-8.
In the upcoming years, artificial intelligence is going to transform the practice of medicine in most of its specialties. Deep learning can help achieve better and earlier problem detection, while reducing errors on diagnosis. By feeding a deep neural network (DNN) with the data from a low-cost and low-accuracy sensor array, we demonstrate that it becomes possible to significantly improve the measurements' precision and accuracy. The data collection is done with an array composed of 32 temperature sensors, including 16 analog and 16 digital sensors. All sensors have accuracies between [Formula: see text]. 800 vectors are extracted, covering a range from to 30 to [Formula: see text]. In order to improve the temperature readings, we use machine learning to perform a linear regression analysis through a DNN. In an attempt to minimize the model's complexity in order to eventually run inferences locally, the network with the best results involves only three layers using the hyperbolic tangent activation function and the Adam Stochastic Gradient Descent optimizer. The model is trained with a randomly-selected dataset using 640 vectors (80% of the data) and tested with 160 vectors (20%). Using the mean squared error as a loss function between the data and the model's prediction, we achieve a loss of only 1.47x10[Formula: see text] on the training set and 1.22x10[Formula: see text] on the test set. As such, we believe this appealing approach offers a new pathway towards significantly better datasets using readily-available ultra low-cost sensors.
在未来几年，人工智能将改变大多数医学专业的实践。深度学习可以帮助实现更好和更早的问题检测，同时减少诊断错误。通过将深度神经网络 (DNN) 与低成本、低精度传感器阵列的数据进行融合，我们证明了显著提高测量精度和准确性是可能的。数据采集是通过一个由 32 个温度传感器组成的阵列完成的，其中包括 16 个模拟传感器和 16 个数字传感器。所有传感器的精度都在 [公式：见正文] 之间。提取了 800 个向量，涵盖了从 30 到 [公式：见正文] 的范围。为了提高温度读数，我们使用机器学习通过 DNN 执行线性回归分析。为了使模型的复杂度最小化，以便最终在本地进行推断，使用双曲正切激活函数和 Adam 随机梯度下降优化器的最佳结果网络只涉及三个层。使用随机选择的数据集中的 640 个向量（数据的 80%）进行训练，并使用 160 个向量（20%）进行测试。使用均方误差作为数据和模型预测之间的损失函数，我们在训练集上的损失仅为 1.47x10[公式：见正文]，在测试集上的损失为 1.22x10[公式：见正文]。因此，我们相信这种有吸引力的方法为使用现成的超低成本传感器获得更好的数据集提供了一条新途径。
