Session D1-R1

Medical, Biomedical, and Health Informatics

Conference
7:00 PM — 9:00 PM HKT
Local
Mar 1 Mon, 5:00 AM — 7:00 AM CST

A Motor Rehabilitation's Motion Range Assessment with Low-cost Virtual Reality Serious Game

Douglas Farias and Rafael de Souza (Universidade Tecnológica Federal do Paraná, Brazil); Paulo Nardi (UTFPR, Brazil); Eduardo F Damasceno (Federal University of Technology - Parana - UTFPR, Brazil)

1
This article shows a goniometric evaluation with two softwares developed for rehabilitation of Spasmodic Torti- collis (ST) with two distinct technologies (Virtual Reality and Augmented Reality) in a low cost of production . Our study was conducted with 12 (twelve) people. Objective: To verify, by functional criteria (kinesiological) and computational criteria (interaction and involvement), which was the best technique in the approach for ST treatment. Method: A cross-sectional study method was applied, analyzing a removal of the subjects' moment compared to control groups at the same moment. Result: The two technologies, Virtual Reality and Augmented Reality are presented as an instrument that adds a playful object to therapy and favors the active participation of the patient during treatment. However, Virtual Reality tends to have greater fidelity for goniometric data generation.

Detection of cardiomegaly in chest radiography using semi-supervised training

Clement Bernardo Marques, Jessica dos Santos de Oliveira, Maria Fernanda Wanderley, Priscilla Wagner and Walter Martins Filho (NeuralMed, Brazil)

2
A detection model of cardiomegaly condition on chest xRays has been developed using a new technique of semi-supervised learning called RemixMatch. This technique uses Augmentation anchoring which consists in putting many strongly augmented versions of one input and fit the outputs of the model with the predictions of a weakly augmented version of the same input. The idea is to enforce the consistency of the model using unlabeled data. This technique produced very promising results. Indeed, we were able to reproduce approximately the same results (compared with the model trained with 100% of the labeled dataset) using only 30% of the labeled dataset.

Beneficial Effects of Cerebellar Low Frequency Repetitive Transcranial Magnetic Stimulation on the Patient with Meige's syndrome

Xue Shi and Xiaolin Su (Shenzhen People's Hospital, China); Yi Guo (Jinan University, China)

1
Meige's syndrome, one type of the segmental cranial dystonia, is characterized by blepharospasm and oromandibular dystonia and can be involved with involuntary movement of lower facial muscles, mouth, pharyngeal or cervical muscles. The exact pathophysiology is still obscure. Clinically, there are no curative drugs and the therapeutic effects of botulinum toxin injection is limited. Repetitive transcranial magnetic stimulation (rTMS), one of the non-invasive techniques of brain stimulation, can be able to induce lasting changes of cortical excitability and remodeling of brain networks. In this study, we first reported the beneficial and long-lasting effect of cerebellar low-frequency rTMS on the patient with Meige's syndrome, which seems to offer the alternative therapeutic method for the disease.

Ordered Microporous Zeolite Film for Memristive Electronics

Bin Li (Qingdao University of Science and Technology, China)

1
Microporous zeolite materials, a kind of microporous crystalline aluminosilicates, have been widely applied in the fields of adsorption, separation, and catalysis. Here, zeolite materials are demonstrated to construct memristor to realize the memristive behavior. The set voltage of device can be tuned through adjusting the composition of zeolite. Meanwhile, OFF/ON ratio are increased when heteroatom irons are introduced into the framework of zeolite.

Flexible NH3 gas sensor based on porous nanosheet-assembled ZnFe2O4/polyaniline yolk-shell microspheres

Yi Zeng (Jilin University, China)

1
Porous nanosheet-assembled ZnFe2O4/polyaniline yolk-shell spheres are synthesized using a combination of solvothermal-thermal treatment and in situ chemical oxidation polymerization method. The as-prepared ZnFe2O4 microspheres are composed of homogeneous primary ZnFe2O4 nanosheets with uniform size and high porosity. Because of their large specific surface area and easily penetrable structures, the as-synthesized ZnFe2O4/polyaniline yolk-shell microspheres exhibit high sensitivity to ammonia and fast response-recovery capability at room temperature.

Session D1-R2

E-Health Services and Applications

Conference
7:00 PM — 9:00 PM HKT
Local
Mar 1 Mon, 5:00 AM — 7:00 AM CST

The fight against COVID-19 in the age of Connected Medical Sensors and Computational Grids

Philippe Gavillet (CERN, Brazil); Bernard Marechal (IF/UFRJ, Brazil); Diego Carvalho (Federal Centre for Engineering Studies and Technological Education - CEFET/RJ, Brazil)

1
The dramatic progress in the medical sensor performance and connectivity opens new healthcare perspectives such as home diagnostic of patients, prompt detection of epidemic emergence and in such a case, detailed study of disease properties, statistical monitoring of the evolution from the regional to international levels while providing most relevant field data for validation of epidemiological forecasts.

Classification of fNIRS Data Under Uncertainty: A Bayesian Neural Network Approach

Talha Siddique and Md Shaad Mahmud (University of New Hampshire, USA)

2
Functional Near-Infrared Spectroscopy (fNIRS) is a non-invasive form of Brain-Computer Interface (BCI). It is used for the imaging of brain hemodynamics and has gained popularity due to the certain pros it poses over other similar technologies. The overall functionalities encompass the capture, processing and classification of brain signals. Since hemodynamic responses are contaminated by physiological noises, several methods have been implemented in the past literature to classify the responses in focus from the unwanted ones. However, the methods, thus far does not take into consideration the uncertainty in the data or model parameters. In this paper, we use a Bayesian Neural Network (BNN) to carry out a binary classification on an open-access dataset, consisting of unilateral finger tapping (left- and right-hand finger tapping). A BNN uses Bayesian statistics to assign a probability distribution to the network weights instead of a point estimate. In this way, it takes data and model uncertainty into consideration while carrying out the classification. We used Variational Inference (VI) to train our model. Our model produced an overall classification accuracy of 86.44% over 30 volunteers. We illustrated how the evidence lower bound (ELBO) function of the model converges over iterations. We further illustrated the uncertainty that is inherent during the sampling of the posterior distribution of the weights. We also generated a ROC curve for our BNN classifier using test data from a single volunteer and our model has an AUC score of 0.855.

A Self-supported Flexible Electrode Based on Graphene Modified Carbon Cloth for Glucose Detection

Qing Liu (Shenzhen Institutes of Advanced Technology,Chinese Academy of Sciences, China); Xixi Ji and Li Wang (Harbin Institute of Technology Shenzhen, China); Huanhuan Feng (Harbin Institute of Technology, Shenzhen, China)

1
Exploring flexible electrode materials with good robustness and stability is essential in developing sensor devices for biomedical applications. Here we fabricate a new type flexible electrode based on graphene modified carbon cloth. Graphene nano-sheets (GNSs) were in situ grown vertically on the surface of carbon cloth (CC) via thermal chemical vapor deposition (CVD). The GNSs/CC composite was characterized by Brunauer-Emmett-Teller (BET), scanning electron microscopy (SEM), Raman spectrometry and cyclic voltammetry (CV). Decent enzyme loading capacity of GNSs/CC was confirmed by doping of glucose oxidase (GOD), which transformed this flexible composite material into a biological electrochemical sensor. Detection of glucose yielded remarkable sensitivity of 85 μA mM −1 cm−2 in the linear range of 0.01~ 0.5 mM.

Wearable Thermoelectric Generators Based on Liquid Metal

Dongting Jiang, Zhengfang Qian and Renheng Wang (Shenzhen University, China)

1
The exploration of renewable resources has been carried out for centuries. Among all the candidates, thermoelectric (TE) materials have attracted great attention, as they are environment-friendly, easy to fabricate, and have permanent lifetime. Furthermore, when flexibility is introduced to TE materials, wearable thermoelectric generators (WTEGs) can be integrated. WTEGs are promising devices for energy applications, as they can be attached to human skins and provide electricity based on the temperature difference between the human body and the environment. Therefore, WTEGs show great potentials in the fields of space exploration, sensors, human care, and bio-health, etc. For a desirable WTEG, it is challenging to gain a high stretchability and TE efficiency simultaneously. To address the issue, in this work, liquid metals (LMs) have been used as the interconnects to enhance the device flexibility and minimize the contact resistance [8]. LMs are in the state of liquid at room temperature, hence, they exhibit remarkable flexibility. Besides, their metallic behavior leads to high electrical conductivity. Therefore, LMs can potentially lead to the realization of WTEGs with high flexibility and TE performance

Smartphone Hypertension Detector: Monitoring Sleep Behavior to Detect Possible Hypertensive Population

Junqiao Fan (The Hong Kong Polytechnic University, China); Shihao Xu (Lanzhou University, China); Xiping Hu (Chinese Academy of Sciences, China)

1
Hypertension detection is an important global health challenge, in which there exists a large population of hypertensive people, yet a great portion of them remain undiagnosed. Recently, studies have indicated a strong correlation between early obstructive sleep apnea and the development of hypertension. Based on the relationship, we propose the smartphone hypertension detector which can detect hypertensive users by automatically monitoring user's overnight sleep behavior. The portable mobile application can serve as a premonitory detection method complementary to traditional medical detection methods. Experiments of smartphone hypertension detector have demonstrated its desired functionality and practicality for real-world deployment.

Made with in Toronto · Privacy Policy · © 2022 Duetone Corp.