This work presents an analysis of the efficiency and effectiveness of a Video-Based Pain Monitoring System running on a Raspberry selected because it is a cheap device that can be easily carried around. The objective of the evaluated system is to allow the assessment of pain based on two characteristics: Heart Rate (HR) and facial expressions detected through the Facial Action Coding System (FACS). To measure HR an Eulerian Video Magnification (EVM) based method was implemented. EVM is one of the main current approaches to measure HR by Remote PhotoPlethysmoGraphy. FACS was used to detect pain intensity with the Prkachin and Solomon Pain Intensity (PSPI) equation which is one of the most used approaches to detect pain intensity based on facial features. To identify the PSPI value we trained a Regression Neural Network (RNN) with the UNBC-McMaster database. The experimental results demonstrate the strengths and limitations of the evaluated system.

This work presents an approach to the automatic detection of Butterfly Malar Rash (BMR) in images. BMR is a Lupus symptom characterized by a reddish facial rash that appears symmetrically in the cheeks and the back of the nose. The proposed approach is based on Transfer Learning, a popular approach in Deep Learning that consists in the use of pre-trained models as the starting point for computer vision and natural language processing tasks. To perform the experiments, a database was created with images manually collected from the Instagram social network, searching for images with #butterflyrash. We evaluated the proposed approach with eight Convolutional Neural Networks (CNN) architecture. The experimental results are good results, with a precision of up to 0.957.

The objective of this work is to present a model of neural network for the detection and segmentation of human's body parts (HPNet). We offer a multiplatform real-time solution that can be run on ordinary computers, as well as on mobile devices and embedded systems. Our proposal is characterized by presenting a compact solution, and by investigating a part of object detection still little explored. One of the striking features presented is the ability to recognize parts of the human body even in uncontrolled environments, due to the use of a random subset of Google's public database (Open Images Dataset) that contains images with objects in the most varied sizes, positions, lighting and occlusion conditions. At first, we offer a solution only for the detection and segmentation of the common parts of the human body, but we intend to expand its capabilities to detect other more specific parts and regions. The main purpose of our model is its use to solve specific problems that require the detection and segmentation of human's body parts, for example, in user authentication.