International Journal of Electronic Healthcare (7 papers in press)
Network Analysis of Infant Brain Functional Connectivity using Spectral EEG.
by Hemang Shrivastava, Hare Ram Sah, Manoj Ramaiya, Jayesh Gangrade
Abstract: Currently MRI is the primary method used to calculate brain connectivity in sleeping infants. In this study, we analysed functional connectivity in infants through event-related potentials (ERP) calculated using electroencephalogram (EEG). Time locked EEG is cheaper than MRI and provides temporally precise brain response. Each electrode of the EEG acquisition cap used for recording the EEG signals served as a node in the derived connectivity network. Finally, we have calculated connectivity network measures like characteristic path length (CPL), global efficiency (GE), and average clustering coefficient (ACC) for each infant. Both term and pre-term infants resting state connectivity was characterised by a dense cluster between central and centro-parietal locations. However, the tactile network in term infants displayed lateralisation to the centro-parietal locations overlaying the somatosensory cortex, while the pre-terms tactile network was dense between frontal and central locations. Overall, the tactile network CPL decreased, while GE and ACC increased with higher gestational age at birth, indicating increased efficiency. This is the first study demonstrating differences between tactile functional connectivity in term and pre-term infants using EEG. We suggest that network measures like CPL, GE and ACC may be used as biomarkers of tactile-dependent functional outcomes in early childhood.
Keywords: functional connectivity; electroencephalography; EEG; network analysis; infants brain; resting state connectivity; event-related potentials; ERP; brain-computer interface; BCI.
Balanced Scorecard Adoption in Healthcare
by Renato Lopes Da Costa, Leandro Ferreira Pereira, Álvaro Dias, Carlos Jeronimo, Rui Gonçalves
Abstract: The health sector is gaining more and more space in the market. The offer of health services, of different specialties, has been increasing exponentially in the last decades. However, professionals in the field have little knowledge of management and many do not know how to use this knowledge in favour of their establishment. The lack of adequate management in health services can lead to not only financial losses, but the delivery of low-quality services can have serious consequences for the health of the client/patient. The objective of the present study was to understand the potential of balanced scorecard adoption in health institutions. The research was based on secondary data of hundreds of articles about the adoption of this standard is health sector. The results showed that the balanced scorecard is an effective method to apply in public and private health sector.
Keywords: health service; balanced scorecard; BSC; health management; standard adoption.
MiPrescription as the information infrastructure backbone in the Colombian healthcare system
by Sandra Agudelo, Polyxeni Vassilakopoulou, Margunn Aanestad
Abstract: mplementing national public healthcare information infrastructures are often challenging, this is even more so in the global south. In this paper we offer an analysis of a successful development and introduction of a national healthcare information system. The MIPRES was developed by the Colombian Ministry of Health as a web application for the prescription of high-cost technologies (drugs and devices), but over time grew to become the backbone of a renewed digital healthcare information infrastructure. Through a qualitative empirical study, informed by information infrastructure theory, we document the emergence and evolution of MIPRES and analyse both the contextual conditions and strategic choices that led to its success.
Keywords: infrastructure; healthcare; digital healthcare systems; electronic prescribing; health informatics; health information technology; information systems; health systems.
Adoption and Implementation of Electronic Healthcare Management System
by Olayemi Olawumi, Sunday Adewale Olaleye, Frank Adusei-Mensah, Adedayo Olawuni, Richard O. Agjei
Abstract: Electronic healthcare management system (EHMS) is seen to have a positive impact on healthcare, but its implementation and adoption are still very low; also, research results on its influence are limited. To ascertain the cause for this gap, a study was conducted to identify gaps in research and knowledge regarding EHMS adoption. This study sought to quantify this lack of research by identifying the current state of EHMS globally and determines how research on implementation, influence, and adoption of EHMS has evolved; two databases were searched for literatures in EHMS, and a bibliometric analysis was performed to understand the nature of research and publication trends in EHMS. We found a relatively small number of literatures that focused on EHMS and a declining state of publication. This study highlights the need to develop a strong evidence base research to support the influence, adoption and effective implementation of EHMS in healthcare institutions.
Keywords: e-health; electronic healthcare management system; EHMS; electronic health record; EHRs; bibliometrics; literature review; citation and co-citation analysis.
Basic Information requirements for designing COVID-19 disease registration system
by Sorayya Rezayi, Marjan Ghazisaeidi, Shahrzad Amirazodi, Soheila Saeedi
Abstract: Collecting data related to COVID-19 disease can significantly impact the management of this pandemic. One of the tools that can be helpful in controlling this disease is developing registry. This study aimed to determine the technical and data requirements of the registry for COVID-19. Databases with keywords related to 'COVID-19' and 'registry' were searched. The existing registries in COVID-19 in Iran and other countries were examined. To finalise the data elements, focus group sessions were held with experts. Thirteen main data classes were identified: demographic information, critical dates, type of sample, patient condition, signs and symptoms, history of smoking and consuming narcotics, history of drugs consumption, imaging finding, differential diagnoses of a severe respiratory disease syndrome, underlying conditions, complications of the disease, respiratory support and discharge outcomes. Determining data requirements is the first step in developing a registry that can create a standard infrastructure for consistent data collection.
Keywords: registry; data requirement; COVID-19; focus group; minimum dataset; MDS.
A New Transparent Cloud-Based Model for Sharing Medical Images with Data Compression and Proactive Resource Elasticity
by Thiago Lopes, Igor De Nardin, Cristiano André Costa, Rodrigo Da Rosa
Abstract: Telemedicine and remote diagnosis are yet challenging because nowadays we do not have scalable and standard solutions for sharing DICOM images across different hospitals. DiCloud is a model to enable DICOM image sharing among different hospitals, enabling patients to access their exams effortlessly. We provide an interoperability module where current DICOM applications do not need to be changed, allowing users to save and retrieve data to/from both the cloud and local PACS. Besides this novelty, DiCloud also brings the following state-of-the-art contributions: 1) ARIMA-based proactive cloud elasticity by anticipating resource reorganisation before under- or overprovisioning situations take place; 2) compression engine for sending and receiving messages at both client and cloud parts transparently. The results demonstrate that it is possible to keep the QoS (0% request errors) and reduce cloud costs. In particular, proactive elasticity enabled fewer connected VMs for more extended periods, saving costs related to cloud usage.
Keywords: DiCloud; DICOM; sharing; performance; high-performance computing; elasticity; scalability.
An Effective Learning Rate Scheduler for Stochastic Gradient Descent Based Deep Learning Model in Healthcare Diagnosis System
by Sathyabama K, K. Saruladha
Abstract: This study develops an effective stochastic gradient descent (SGD) and time with exponential decay (TED)-based learning rate scheduler called SGD-TED model for deep learning-based healthcare diagnosis. The presented SGD-TED model involves pre-processing, classification, SGD-based parameter tuning and TED-based learning rate scheduling. Once the data is pre-processed, three DL models namely recurrent neural network (RNN), long short-term memory (LSTM) and gated recurrent unit (GRU) are used for diagnosis. Then, the hyperparameter tuning takes place by SGD and TED is applied to schedule the learning rate proficiently. The application of SGD-TED approach in the DL models considerably helps to increase the classification performance. The effectiveness of the SGD-TED model is assessed on three benchmark medical dataset and the experimental outcome ensured that the SGD-TED-LSTM model has resulted to a higher accuracy of 98.59%, 93.68% and 95.20% on the applied diabetes, EEG Eye State and sleep stage dataset.
Keywords: deep learning; stochastic gradient descent; SGD; learning rate scheduler; long short-term memory; LSTM; healthcare; time with exponential decay; TED; recurrent neural network.