No abstract

In this paper, a novel method for classifying electrocardiogram signals in mobile devices is proposed, which classifies different arrhythmias according to the Association for the Advancement of Medical Instrumentation standard EC57. A convolutional neural network has been constructed, trained and validated with the MIT-BIH Arrhythmia Dataset, which has 5 different classes: normal beat, supraventricular premature beat, premature ventricular contraction, fusion of ventricular and normal beat, unclassifiable beat. Once trained and validated, the model is subjected to a post-training quantization stage using the TensorFlow Lite conversion method. The obtained results were satisfactory, before and after the quantization, the convolutional neural network obtained an accuracy of 98.5%. With the quantization technique it was possible to obtain a significant reduction in model size, thus enabling the development of the mobile application, this reduction was approximately 90% compared to the original model size.

This paper presents an approach for the classification of child chest X-ray images into two classes: pneumonia and normal. We employ Convolutional Neural Networks, from pre-trained networks together with a quantization process, using the platform TensorFlow Lite method. This reduces the processing requirement and computational cost. Results have shown accuracy up to 95.4% and 94.2% for MobileNetV1 and MobileNetV2, respectively. The resulting mobile app also presents a simple and intuitive user interface.

No abstract