by Beijing Zhongke Journal Publishing Co.
When an image of a patient or a human organ is virtually reproduced, a digital twin-based system is developed using important data collected from different biomedical sensors and these ubiquitous sensors that can be studied. Similar to a smartwatch, real-time information about a patient’s or human’s blood pressure, pulse rate, body temperature, sleep patterns and general physical activities can be obtained. A virtual model can be created during clinic or hospital visits using laboratory test data, and diagnostic imaging investigations can be conducted. Additionally, genetic data can be encoded as digital twins. When all of this data is combined into a single virtual model, all the details of the patient’s medical history are available to aid in decision making. Credit: Beijing Zhongke Journal Publishing Co. Ltd.
Digital twins are virtual representations of devices and processes that capture physical properties of the environment and operational algorithms/techniques in the context of medical devices and technologies. Digital twins can enable healthcare organizations to determine ways to improve medical processes, improve patient experience, reduce operating expenses and extend the value of care.
During the current COVID-19 pandemic, various medical devices, such as X-ray and CT scan machines and methods, are constantly being used to collect and analyze medical images. When collecting and processing a large volume of data in the form of images, machines and processes sometimes suffer from system failures, creating critical problems for hospitals and patients.
To remedy this, a study published in Virtual reality and smart hardware introduces an intelligent healthcare system based on a digital twin embedded with medical devices to collect information regarding the current health status, configuration and maintenance history of the device/machine/system. Additionally, medical images i.e. X-rays are analyzed using a deep learning model to detect COVID-19 infection.
The designed system is based on the Cascading Recurrent Convolutional Neural Network (RCNN) architecture. In this architecture, the detector stages are deeper and sequentially more selective against small and nearby false positives. This architecture is a multi-stage extension of the RCNN model and trained sequentially using the output of one stage to train the other. At each step, the bounding boxes are adjusted to locate an appropriate value of the closest false positives when learning the different steps. In this way, the arrangement of the detectors is adjusted to increase the intersection on the union, overcoming the problem of overfitting. This study trains the model using X-ray images because the model was previously trained on another dataset.
The developed system achieves good accuracy during the detection phase of COVID-19. The experimental results reveal the efficiency of the detection architecture, which gives an average average accuracy rate of 0.94.
Imran Ahmed et al, Integration of Digital Twins and Deep Learning for Medical Image Analysis in the Age of COVID-19, Virtual reality and smart hardware (2022). DOI: 10.1016/j.vrih.2022.03.002
Provided by Beijing Zhongke Journal Publishing Co.
Quote: Integrating digital twins and deep learning for medical image analysis in the era of COVID-19 (2 November 2022) retrieved 2 November 2022 from https://medicalxpress.com/news/2022-11-digital-twins-deep-medical -image.html
This document is subject to copyright. Except for fair use for purposes of private study or research, no part may be reproduced without written permission. The content is provided for information only.