Using Machine Learning Classifiers to Assist Healthcare-Related Decisions: Classification of Electronic Patient Records

Surveillance Levels (SLs) are categories for medical patients (used in Brazil) that represent different types of medical recommendations. SLs are defined according to risk factors and the medical and developmental history of patients. Each SL is associated with specific educational and clinical measures. The objective of the present paper was to verify computer-aided, automatic assignment of SLs. The present paper proposes a computer-aided approach for automatic recommendation of SLs. The approach is based on the classification of information from patient electronic records. For this purpose, a software architecture composed of three layers was developed. The architecture is formed by a classification layer that includes a linguistic module and machine learning classification modules. The classification layer allows for the use of different classification methods, including the use of preprocessed, normalized language data drawn from the linguistic module. We report the verification and validation of the software architecture in a Brazilian pediatric healthcare institution. The results indicate that selection of attributes can have a great effect on the performance of the system. Nonetheless, our automatic recommendation of surveillance level can still benefit from improvements in processing procedures when the linguistic module is applied prior to classification. Results from our efforts can be applied to different types of medical systems. The results of systems supported by the framework presented in this paper may be used by healthcare and governmental institutions to improve healthcare services in terms of establishing preventive measures and alerting authorities about the possibility of an epidemic.

Changes in Scleral Architecture in Chronic Vogt-Koyanagi-Harada Disease.

PURPOSE: To describe scleral changes in chronic VKH. METHODS: Medical records of patients with chronic VKH were retrospectively reviewed. Change of scleral architecture was defined as progressive posterior bowing on OCT, axial length elongation, and/or increased myopia more than -1.0 D, not explicable by other etiologies. RESULTS: In total, 28 eyes (16 patients) with mean age of disease onset 32.5 ± 14.0 years were included in the study. Disease duration was 15.1 ± 10.2 years. Eight eyes (28.6%) showed progressive scleral architectural changes. Five eyes (18%) developed scleral changes on OCT, not seen on prior imaging (2-12 years earlier). One eye had posterior bowing on OCT with increased axial length, both eyes of a bilateral pseudophake developed increased myopia with increased axial length. Well-circumscribed chorioretinal atrophy within the arcade was associated with progressive scleral change. CONCLUSIONS: Progressive scleral change may develop as a late complication of VKH. The association with well-circumscribed chorioretinal atrophy suggests that chronic choroidal inflammation may be responsible.

Using new scanning electron microscopy (SEM) approaches to elucidate bacterial biofilm architecture

Healthcare-associated infections (HAI) are a major public health problem being Klebsiella pneumoniae and nontuberculous mycobacteria, both with high antibiotic resistance rates, among their etiological agent. Since biofilme assembly is pointed as one of the mechanisms involved in emergence of antibiotic resistance understanding bacteria organization within the biofilm and the identification of differences between planktonic and sessile forms of bacteria will be a step forward to fight HAI. In the present work we used SEM as a tool to characterize the internal structure of biofilm assembled on different surfaces. For SEM analysis, biofilms were allowed to form either on six-well cell culture plates, silicon or metallic disks placed inside the wells for different incubation periods at 37 °C. The biofilm assembled on the cell culture dish was for both secondary and backscattered electron analysis as described before. Biofilms assembled on silicon disks instead of being sectioned were prepared as metallographic samples, by grinding with grit SIC paper and polishing with diamond particles. Samples were cleaned (70% ethanol), dried with hot air, further coated and analysed. A preliminary study using FIB-SEM has been performed to access the ultrastructure of biofilms assembled on metallic surfaces. The results obtained showed that the same bacteria assembled biofilms with different ratios of biomass and extracellular matrix depending on the surface. SEM performed on thin sections of biofilms is a powerful tool to elucidate biofilm structure allowing the quantification of the major components. FIB-SEM is also a promising tool in this field.

Federated architecture for biomedical data integration

The last decades have been characterized by a continuous adoption of IT solutions in the healthcare sector, which resulted in the proliferation of tremendous amounts of data over heterogeneous systems. Distinct data types are currently generated, manipulated, and stored, in the several institutions where patients are treated. The data sharing and an integrated access to this information will allow extracting relevant knowledge that can lead to better diagnostics and treatments. This thesis proposes new integration models for gathering information and extracting knowledge from multiple and heterogeneous biomedical sources. The scenario complexity led us to split the integration problem according to the data type and to the usage specificity. The first contribution is a cloud-based architecture for exchanging medical imaging services. It offers a simplified registration mechanism for providers and services, promotes remote data access, and facilitates the integration of distributed data sources. Moreover, it is compliant with international standards, ensuring the platform interoperability with current medical imaging devices. The second proposal is a sensor-based architecture for integration of electronic health records. It follows a federated integration model and aims to provide a scalable solution to search and retrieve data from multiple information systems. The last contribution is an open architecture for gathering patient-level data from disperse and heterogeneous databases. All the proposed solutions were deployed and validated in real world use cases.

HEALTHCARE INFORMATION SYSTEMS INTEGRATION USING SOA

The growing pressure to increase the quality of health services, as well as reducing costs, has caused healthcare organizations to increase the use of Information and Communication Technologies (ICT) through the development and adoption of Healthcare Information Systems (HIS). However, the need for exchange of information between HIS and between organizations has also increased, resulting in the problem of interoperability. This problem is considered complex, but the use of Service Oriented Architecture (SOA) appears as a good way to address this issue. This paper presents a systematic review, performed in order to find out how and in which contexts SOA is being used to ensure the interoperability of HIS.

Communication Architecture Design for Real-Time Networked Control Systems

Networked control over data networks has received increasing attention in recent years. Among many problems in networked control systems (NCSs) is the need to reduce control latency and jitter and to deal with packet dropouts. This paper introduces our recent progress on a queuing communication architecture for real-time NCS applications, and simple strategies for dealing with packet dropouts. Case studies for a middle-scale process or multiple small-scale processes are presented for TCP/IP based real-time NCSs. Variations of network architecture design are modelled, simulated, and analysed for evaluation of control latency and jitter performance. It is shown that a simple bandwidth upgrade or adding hierarchy does not necessarily bring benefits for performance improvement of control latency and jitter. A co-design of network and control is necessary to maximise the real-time control performance of NCSs