Low-Cost Non-Wearable Fall Detection System Implemented on a Single Board Computer for People in Need of Care.

This work aims at proposing an affordable, non-wearable system to detect falls of people in need of care. The proposal uses artificial vision based on deep learning techniques implemented on a Raspberry Pi4 4GB RAM with a High-Definition IR-CUT camera. The CNN architecture classifies detected people into five classes: fallen, crouching, sitting, standing, and lying down. When a fall is detected, the system sends an alert notification to mobile devices through the Telegram instant messaging platform. The system was evaluated considering real daily indoor activities under different conditions: outfit, lightning, and distance from camera. Results show a good trade-off between performance and cost of the system. Obtained performance metrics are: precision of 96.4%, specificity of 96.6%, accuracy of 94.8%, and sensitivity of 93.1%. Regarding privacy concerns, even though this system uses a camera, the video is not recorded or monitored by anyone, and pictures are only sent in case of fall detection. This work can contribute to reducing the fatal consequences of falls in people in need of care by providing them with prompt attention. Such a low-cost solution would be desirable, particularly in developing countries with limited or no medical alert systems and few resources.

Um anjo para ELA: arquitetura baseada em visão computacional aplicada ao monitoramento on-line de parâmetros respiratórios de pacientes com Esclerose Lateral Amiotrófica (ELA) em ambiente hospitalar / An angel for ALS: architecture based on computer vision applied for on-line monitoring respiratory parameters of patients with Amyotrophic Lateral Sclerosis (ALS) in hospital environment

INTRODUÇÃO: A Esclerose Lateral Amiotrófica (ELA) é uma doença neurodegenerativa, caracterizada por uma progressiva e fatal perda de neurônios motores do córtex cerebral, tronco encefálico e medula espinhal, mas que mantém preservada a atividade intelectual e cognitiva do paciente. Pacientes acometidos por essa doença irão invariavelmente necessitar do auxílio de ventiladores mecânicos. MÉTODOS: Foi utilizado um conjunto de hardware e software para realizar o monitoramento dos parâmetros respiratórios dos pacientes em leitos hospitalares como forma de auxiliar à equipe de saúde. O monitoramento desses parâmetros deu-se por meio de uma webcam, que capturava os valores exibidos na tela do ventilador mecânico, e do emprego de técnicas de visão computacional e Optical Character Recognition (OCR). Neste sentido, o sistema foi testado sob três condições de luminosidade diferentes para verificar a eficácia do mesmo. RESULTADOS: O sistema apresentou uma média geral de acertos de 94.90%. Além disso, quando a interferência luminosa foi mínima, o sistema obteve uma média geral de acertos de 97,76%. CONCLUSÃO: A adoção de um sistema computacional baseado em visão computacional para auxílio da equipe de saúde no monitoramento hospitalar de pacientes com ELA mostrou-se satisfatória. No entanto, a pesquisa mostrou que a adoção de um sistema com maior imunidade à interferências luminosas externas tende a apresentar melhores resultados.

INTRODUCTION: Amyotrophic Lateral Sclerosis (ALS) is a neurodegenerative disease characterized by a progressive and fatal loss of motor neurons in the cerebral cortex, brainstem and spinal cord. In spite of that, the patient's intellectual and cognitive activity remains preserved. Patients affected by this disease will invariably need the help of mechanical ventilators. METHODS: A set of hardware and software was used to perform the monitoring of respiratory parameters of patients in hospital beds as a means of assisting the healthcare team. The monitoring of these parameters was performed by a webcam that captured the values displayed on the screen of the ventilator, and the employment of computer vision techniques and Optical Character Recognition (OCR). In this sense, the system was tested under three different lighting conditions to verify its effectiveness. RESULTS:The system presented an overall average of 94.90% of correct answers. Furthermore, when the luminous interference was minimum, it achieved an overall average of success of 97.76%. CONCLUSION: The adoption of a computational system based on computer vision to aid the healthcare team in hospital monitoring of patients with ALS was satisfactory. However, the research has shown that the adoption of a system with greater immunity to external light interference tends to achieve better results.