Service composition for biomedical applications

A exigente inovação na área das aplicações biomédicas tem guiado a evolução das tecnologias de informação nas últimas décadas. Os desafios associados a uma gestão, integração, análise e interpretação eficientes dos dados provenientes das mais modernas tecnologias de hardware e software requerem um esforço concertado. Desde hardware para sequenciação de genes a registos electrónicos de paciente, passando por pesquisa de fármacos, a possibilidade de explorar com precisão os dados destes ambientes é vital para a compreensão da saúde humana. Esta tese engloba a discussão e o desenvolvimento de melhores estratégias informáticas para ultrapassar estes desafios, principalmente no contexto da composição de serviços, incluindo técnicas flexíveis de integração de dados, como warehousing ou federação, e técnicas avançadas de interoperabilidade, como serviços web ou LinkedData. A composição de serviços é apresentada como um ideal genérico, direcionado para a integração de dados e para a interoperabilidade de software. Relativamente a esta última, esta investigação debruçou-se sobre o campo da farmacovigilância, no contexto do projeto Europeu EU-ADR. As contribuições para este projeto, um novo standard de interoperabilidade e um motor de execução de workflows, sustentam a sucesso da EU-ADR Web Platform, uma plataforma para realizar estudos avançados de farmacovigilância. No contexto do projeto Europeu GEN2PHEN, esta investigação visou ultrapassar os desafios associados à integração de dados distribuídos e heterogéneos no campo do varíoma humano. Foi criada uma nova solução, WAVe - Web Analyses of the Variome, que fornece uma coleção rica de dados de variação genética através de uma interface Web inovadora e de uma API avançada. O desenvolvimento destas estratégias evidenciou duas oportunidades claras na área de software biomédico: melhorar o processo de implementação de software através do recurso a técnicas de desenvolvimento rápidas e aperfeiçoar a qualidade e disponibilidade dos dados através da adopção do paradigma de web semântica. A plataforma COEUS atravessa as fronteiras de integração e interoperabilidade, fornecendo metodologias para a aquisição e tradução flexíveis de dados, bem como uma camada de serviços interoperáveis para explorar semanticamente os dados agregados. Combinando as técnicas de desenvolvimento rápidas com a riqueza da perspectiva "Semantic Web in a box", a plataforma COEUS é uma aproximação pioneira, permitindo o desenvolvimento da próxima geração de aplicações biomédicas.

Mining biomedical information from scientific literature

The rapid evolution and proliferation of a world-wide computerized network, the Internet, resulted in an overwhelming and constantly growing amount of publicly available data and information, a fact that was also verified in biomedicine. However, the lack of structure of textual data inhibits its direct processing by computational solutions. Information extraction is the task of text mining that intends to automatically collect information from unstructured text data sources. The goal of the work described in this thesis was to build innovative solutions for biomedical information extraction from scientific literature, through the development of simple software artifacts for developers and biocurators, delivering more accurate, usable and faster results. We started by tackling named entity recognition - a crucial initial task - with the development of Gimli, a machine-learning-based solution that follows an incremental approach to optimize extracted linguistic characteristics for each concept type. Afterwards, Totum was built to harmonize concept names provided by heterogeneous systems, delivering a robust solution with improved performance results. Such approach takes advantage of heterogenous corpora to deliver cross-corpus harmonization that is not constrained to specific characteristics. Since previous solutions do not provide links to knowledge bases, Neji was built to streamline the development of complex and custom solutions for biomedical concept name recognition and normalization. This was achieved through a modular and flexible framework focused on speed and performance, integrating a large amount of processing modules optimized for the biomedical domain. To offer on-demand heterogenous biomedical concept identification, we developed BeCAS, a web application, service and widget. We also tackled relation mining by developing TrigNER, a machine-learning-based solution for biomedical event trigger recognition, which applies an automatic algorithm to obtain the best linguistic features and model parameters for each event type. Finally, in order to assist biocurators, Egas was developed to support rapid, interactive and real-time collaborative curation of biomedical documents, through manual and automatic in-line annotation of concepts and relations. Overall, the research work presented in this thesis contributed to a more accurate update of current biomedical knowledge bases, towards improved hypothesis generation and knowledge discovery.

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.

Hybrid materials for biomedical applications

The increased longevity of humans and the demand for a better quality of life have led to a continuous search for new implant materials. Scientific development coupled with a growing multidisciplinarity between materials science and life sciences has given rise to new approaches such as regenerative medicine and tissue engineering. The search for a material with mechanical properties close to those of human bone produced a new family of hybrid materials that take advantage of the synergy between inorganic silica (SiO4) domains, based on sol-gel bioactive glass compositions, and organic polydimethylsiloxane, PDMS ((CH3)2.SiO2)n, domains. Several studies have shown that hybrid materials based on the system PDMS-SiO2 constitute a promising group of biomaterials with several potential applications from bone tissue regeneration to brain tissue recovery, passing by bioactive coatings and drug delivery systems. The objective of the present work was to prepare hybrid materials for biomedical applications based on the PDMS-SiO2 system and to achieve a better understanding of the relationship among the sol-gel processing conditions, the chemical structures, the microstructure and the macroscopic properties. For that, different characterization techniques were used: Fourier transform infrared spectrometry, liquid and solid state nuclear magnetic resonance techniques, X-ray diffraction, small-angle X-ray scattering, smallangle neutron scattering, surface area analysis by Brunauer–Emmett–Teller method, scanning electron microscopy and transmission electron microscopy. Surface roughness and wettability were analyzed by 3D optical profilometry and by contact angle measurements respectively. Bioactivity was evaluated in vitro by immersion of the materials in Kokubos’s simulated body fluid and posterior surface analysis by different techniques as well as supernatant liquid analysis by inductively coupled plasma spectroscopy. Biocompatibility was assessed using MG63 osteoblastic cells. PDMS-SiO2-CaO materials were first prepared using nitrate as a calcium source. To avoid the presence of nitrate residues in the final product due to its potential toxicity, a heat-treatment step (above 400 °C) is required. In order to enhance the thermal stability of the materials subjected to high temperatures titanium was added to the hybrid system, and a material containing calcium, with no traces of nitrate and the preservation of a significant amount of methyl groups was successfully obtained. The difficulty in eliminating all nitrates from bulk PDMS-SiO2-CaO samples obtained by sol-gel synthesis and subsequent heat-treatment created a new goal which was the search for alternative sources of calcium. New calcium sources were evaluated in order to substitute the nitrate and calcium acetate was chosen due to its good solubility in water. Preparation solgel protocols were tested and homogeneous monolithic samples were obtained. Besides their ability to improve the bioactivity, titanium and zirconium influence the structural and microstructural features of the SiO2-TiO2 and SiO2-ZrO2 binary systems, and also of the PDMS-TiO2 and PDMS-ZrO2 systems. Detailed studies with different sol-gel conditions allowed the understanding of the roles of titanium and zirconium as additives in the PDMS-SiO2 system. It was concluded that titanium and zirconium influence the kinetics of the sol-gel process due to their different alkoxide reactivity leading to hybrid xerogels with dissimilar characteristics and morphologies. Titanium isopropoxide, less reactive than zirconium propoxide, was chosen as source of titanium, used as an additive to the system PDMS-SiO2-CaO. Two different sol-gel preparation routes were followed, using the same base composition and calcium acetate as calcium source. Different microstructures with high hydrophobicit were obtained and both proved to be biocompatible after tested with MG63 osteoblastic cells. Finally, the role of strontium (typically known in bioglasses to promote bone formation and reduce bone resorption) was studied in the PDMS-SiO2-CaOTiO2 hybrid system. A biocompatible material, tested with MG63 osteoblastic cells, was obtained with the ability to release strontium within the values reported as suitable for bone tissue regeneration.