Healthcare systems handle thousands of documents daily across various departments, requiring some effort during the digitalization processes. One strategy employed by medical staff is recording appointments for later transcription. However, this process is time-consuming and not practical for all scenarios. In this paper, we present a comprehensive methodology for converting medical audio recordings into structured documentation through multiple AI-based solutions. We propose and evaluate three distinct methods: a baseline two-stage pipeline using Mixtral 7B and Llama 70B models, a cyclic LLM approach leveraging self-improvement loops, and an embedding-based retrieval system utilizing BGE M3. Our experimental results show that the RBF kernel consistently outperformed linear kernels and logistic regression approaches across all metrics, maintaining high precision (0.87-0.94) and perfect recall.

The integration of heterogeneous clinical datasets represents a fundamental challenge in contemporary biomedical research, particularly when reconciling multi-language and multi-institution data sources. The challenge of this procedure lies in the effort required to map the original concepts with their standard definitions. Various automated mapping solutions can assist researchers in this process, but the complexity grows when handling multi-language datasets, resulting in substantial manual work for translation and mapping. In this paper, we proposed a novel framework for clinical concept harmonisation that leverages vector-based embeddings and semantic search methodologies to enhance interoperability in multi-cohort studies. The methodology incorporates comprehensive data profiling, ontology-driven concept alignment, and machine learning-based vector search within a unified architecture. We demonstrate the efficacy of this approach through practical application to Alzheimer's disease (AD) research datasets from distinct institutions with different languages, achieving effective cross-lingual concept mapping while maintaining compatibility with established standardisation frameworks.

Manual assembly processes remain a fundamental aspect of the manufacturing industry, primarily due to the dexterity and adaptability of human workers. However, the repetitive and physically demanding nature of these tasks highlights the need for an ergonomic and well-balanced workload, as poor ergonomics can also contribute to errors. This paper aims to automate the assembly line balancing process for a real-world case study. A new multi-objective Assembly Line Balancing Problem (ALBP) formulation is proposed, minimizing both workload variance and variance in workers’ energy expenditure across workstations while accounting for daily fluctuations in the number of operators. The proposed metabolic and time-sensitive assembly line balancing method integrates caloric considerations to reduce long-term risks of Work-Related Musculoskeletal Disorders (WMSDs). Using a graph-based approach, the framework ensures that precedence constraints are met while optimizing task assignments to minimize cycle time and metabolic cost per operator. To improve accuracy in estimating energy expenditure, this study employs the Methods-Time Measurement - Universal Analyzing System (MTM-UAS), which decomposes tasks into standardized motion elements. The methodology is validated through a real-world case study at Bosch Thermotechnology in Portugal, assigning tasks while minimizing the trade-offs between worker fatigue and production goals. The code to validate this study is publicly available at https://github.com/joaorafaelalmeida/line-balancing-algorithm.

Software engineering strategies have been studied over the last years, aiming to optimize the development of applications. Continuous Integration/Continuous Delivery (CI/CD) pipelines were a result of agile methodologies created to optimize those processes. While these pipelines can bring significant benefits to organizations, such as faster development and delivery, they also come with security risks. Such issues are sometimes ignored since the tools for automatically analyzing the security breaches in the application are focused on its artifacts (including source code, configurations, environments, and sandboxes), ignoring the CI/CD infrastructure. In this article, we explore the potential security flaws that may exist in CI/CD pipelines, along with the challenges and opportunities to investigate new solutions to mitigate such flaws. The article also proposes different models and strategies for protecting CI/CD pipelines.

The availability of large Electronic Health Records (EHR) databases has created new opportunities for clinical research. To improve interoperability of these databases a Common Data Model (CDM) can be used which enables standardized analytics. However, identifying the appropriate data sources for a specific study remains a challenge. The current strategy used for database discovery are based on catalogues that contain metadata which is often not rich enough for the task at hand. Additionally, sometimes this information is incorrect since it is inserted in such platforms manually by the data owners. In response to this challenge, we proposed the Concept Browser tool. The tool aims to streamline the process of efficiently exploring and selecting suitable OMOP CDM databases aligned with the study purpose. It was developed and validated in the EHDEN project, which currently contains information from more than 180 health databases across Europe and is now being used also in the DARWIN EU®initiative of the European Medicines Agency.

Integrating advanced data processing technologies into healthcare has shifted the medical studies paradigm. These evolve from data collection into management and analysis of Electronic Health Records (EHR) data. This change improved patient care and expanded the scope of clinical research through the secondary usage of existing data. Even though this problem was already solved in other initiatives, it raised new challenges, namely regarding cohort definition, data discovery, and evaluating the study feasibility. There are database catalogues to help in those tasks, but these fail in some cases due to insufficient information. Therefore, in this paper, we address this challenge by proposing a baseline method for information retrieval, including a synthetic dataset for further research. The information present in the dataset was generated from metadata extracted from real-world databases, which represents real problems that do not yet have a solution. The source code of this work is available at http://github.com/bioinformatics-ua/HealthDBFinder.

The advancement of metagenomics is closely tied to bioinformatic tools. These tools, which are essential for taxonomic classification and functional annotation, derive their reliability from the databases they use. However, the challenge arises when comparing these tools through literature, as their evaluations often occur within specific datasets. This practice makes it difficult to compare them directly with state-of-the-art tools, as it obscures how they might perform across a broader range of data. In this study, we evaluate the suitability of different databases for assessing the viability of metagenomic tools. By assessing these databases and comparing them to general databases, we aim to provide researchers with valuable insights into selecting the most appropriate database for their metagenomic studies, enhance the reliability and reproducibility of metagenomic technologies, and overcome challenges such as resolving low-abundance species, distinguishing closely related species, or handling environmental samples with high diversity.

Metagenomics is a rapidly expanding field that uses next-generation sequencing technology to analyze the genetic makeup of environmental samples. However, accurately identifying the organisms in a metagenomic sample can be complex, and traditional reference-based methods may need to be more effective in some instances. In this study, we present a novel approach for metagenomic identification, using data compressors as a feature for taxonomic classification. By evaluating a comprehensive set of compressors, including both general-purpose and genomic-specific, we demonstrate the effectiveness of this method in accurately identifying organisms in metagenomic samples. The results indicate that using features from multiple compressors can help identify taxonomy. An overall accuracy of 95% was achieved using this method using an imbalanced dataset with classes with limited samples. The study also showed that the correlation between compression and classification is insignificant, highlighting the need for a multi-faceted approach to metagenomic identification. This approach offers a significant advancement in the field of metagenomics, providing a reference-less method for taxonomic identification that is both effective and efficient while revealing insights into the statistical and algorithmic nature of genomic data. The code to validate this study is publicly available at https://github.com/ieeta-pt/xgTaxonomy.

The increasing of health data sources to support clinical practice is opening the path for its secondary use in biomedical research. This changes the research paradigm, from data generation to data management and analysis. Although the potential for secondary use of this data is vast, including the improvement of healthcare systems and the advancement of clinical research, data discovery is challenging. In order to maximize data reusability, the FAIR principles have been developed as a guiding framework for system development. Nevertheless, the discovery and reuse of biomedical data present two main challenges: i) data partners grappling with ethical and social concerns related to data discoverability; ii) clinical researchers struggling to find the best data sources for their research studies. In this paper, we present a platform that provides a set of tools, compliant with the FAIR principles, to help data custodians when sharing data about biomedical databases, while allowing researchers to search for and select databases that meet their specific research needs.

The eHealth sector in Portugal faces significant cybersecurity challenges, including increasing cyber threats and vulnerabilities, inadequate cybersecurity measures, and a short-age of skilled cybersecurity professionals. The importance of addressing these challenges has been emphasised by the European Union's Network and Information Systems (NIS) Directive, which aims to ensure a high common level of cybersecurity across the EU by requiring Member States to adopt national cybersecurity measures and cooperate on incident response. In response to these challenges, we propose in this work, a project capable of improving the cybersecurity posture of eHealth systems in Portugal. The project proposes a novel approach to establishing connections with various organisations, including the national CERT.PT and PANORAMA, for incident response and information sharing. The tasks involved in the project include risk assessment, penetration testing, skills gap analysis, training, incident notification and communication, and controls implementation. By implementing these tasks, the project has the potential to improve the incident response capabilities and overall cybersecurity posture of eHealth systems.

Genomics has profoundly influenced the field of medicine, with advancements in DNA sequencing contributing to personalized medicine and a more comprehensive understanding of various diseases' genomic underpinnings. Sharing genomic data is vital for progressing the field and devising novel approaches to decipher the genome. Nevertheless, the sensitive nature of this information necessitates robust security measures for protection during storage and transfer. In this paper, we introduce SecureFASTA, a novel tool for securely encrypting and decrypting FASTA files without requiring a shared secret while minimizing the number of keys exchanged between pairs. Our approach combines symmetric and asymmetric encryption techniques, utilizing the Advanced Encryption Standard (AES) cypher and Rivest-Shamir-Adleman (RSA) encryption. Additionally, we implement a checksum function using the Secure Hash Algorithm (SHA-256) to verify the integrity of transferred FASTA files. Our evaluation demonstrates that SecureFASTA is fast, reliable, and secure, surpassing existing tools in terms of security and user-friendliness. Consequently, it offers a valuable solution for securely sharing and leveraging sensitive genomic data, marking a significant breakthrough in genomics. The tool's source code is available at https://github.com/bioinformatics-ua/SecureFASTA.

BACKGROUND: Secondary use of health data is a valuable source of knowledge that boosts observational studies, leading to important discoveries in the medical and biomedical sciences. The fundamental guiding principle for performing a successful observational study is the research question and the approach in advance of executing a study. However, in multi-centre studies, finding suitable datasets to support the study is challenging, time-consuming, and sometimes impossible without a deep understanding of each dataset. METHODS: We propose a strategy for retrieving biomedical datasets of interest that were semantically annotated, using an interface built by applying a methodology for transforming natural language questions into formal language queries. The advantages of creating biomedical semantic data are enhanced by using natural language interfaces to issue complex queries without manipulating a logical query language. RESULTS: Our methodology was validated using Alzheimer's disease datasets published in a European platform for sharing and reusing biomedical data. We converted data to semantic information format using biomedical ontologies in everyday use in the biomedical community and published it as a FAIR endpoint. We have considered natural language questions of three types: single-concept questions, questions with exclusion criteria, and multi-concept questions. Finally, we analysed the performance of the question-answering module we used and its limitations. The source code is publicly available at https://bioinformatics-ua.github.io/BioKBQA/. CONCLUSION: We propose a strategy for using information extracted from biomedical data and transformed into a semantic format using open biomedical ontologies. Our method uses natural language to formulate questions to be answered by this semantic data without the direct use of formal query languages.

One of the main goals of clinical studies consists of identifying diseases' causes and improving the efficacy of medical treatments. Sometimes, the reduced number of participants is a limiting factor for these studies, leading researchers to organise multi-centre studies. However, sharing health data raises certain concerns regarding patients' privacy, namely related to the robustness of anonymisation procedures. Although these techniques remove personal identifiers from registries, some studies have shown that anonymisation procedures can sometimes be reverted using specific patients' characteristics. In this paper, we propose a secure architecture to explore distributed databases without compromising the patient's privacy. The proposed architecture is based on interoperable repositories supported by a common data model.

Many medical studies have been conducted aiming for better understanding of the causes of diseases and to assist in treatments and protective factors. In some cases, these studies do not produce impactful findings due to the small number of participants. Some initiatives already invested efforts in conducting multicentre studies, which raises other technical challenges due to the heterogeneity of datasets. The analysis of such data sources implies dealing with different data structures, terminologies, concepts, languages, and most importantly, the knowledge behind the data. In this paper, we present a methodology to centralise different datasets into the tranSMART application, using a harmonising strategy based on standard data schema. This methodology can help researchers to generate evidence from a wider variety of data sources. This proposal was validated using Alzheimer's Disease cohorts from several countries, combining at the end 6,669 subjects and 172 clinical concepts. The harmonised datasets can provide multi-cohort queries and analysis. The software package is available, under the MIT license, at https://github.com/bioinformatics-ua/tranSMART-migrator.

Today, medical studies enable a deeper understanding of health conditions, diseases and treatments, helping to improve medical care services. In observational studies, an adequate selection of datasets is important, to ensure the study's success and the quality of the results obtained. During the feasibility study phase, inclusion and exclusion criteria are defined, together with specific database characteristics to construct the cohort. However, it is not easy to compare database characteristics and their evolution over time during this selection. Data comparisons can be made using the data properties and aggregations, but the inclusion of temporal information becomes more complex due to the continuous evolution of concepts over time. In this paper, we propose two visualisation methods aiming for a better description of data evolution in clinical registers using biomedical standard vocabularies.

Advances in DNA sequencing technologies have led to an unprecedented growth of sequenced data. However, when sequencing de-novo genomes, one of the biggest challenges is the classification of DNA sequences that do not match with any biological sequence from the literature. The use of reference-free methods to identify these organisms supported by compressors is one strategy for taxonomic identification. However, with the high number of compressors available, and the computational resources required to operate them, there is a problem in selecting the best compressors for classification with limited computational resources. In this paper, we present a two-step pipeline to analyze nine compressors, to understand which ones could be the best candidates for taxonomic identification. We use 500 randomly selected sequences from five taxonomic groups to conduct this analysis. The results show that besides being an excellent repre-sentative feature, depending on the compressor, the Normalized Compression (NC) reflects different aspects concerning the nature of a given sequence and its complexity. Furthermore, we show that neither the compression capability of a compressor nor the compressibility of the file correlates with classification accuracy. The code used in this work is publicly available at https://github.com/bioinformatics-ua/COMPACT.

The DICOM standard has been widely adopted for the exchange and management of biomedical images. Its hierarchical structure allows representing data and metadata of medical imaging studies. However, other patient data not directly related with the study, such as prescriptions and treatments, are stored in independent Electronic Health Record (EHR) systems. With the increasing production of medical imaging studies, repositories responsible for storing DICOM images started to contain massive amounts of data. Therefore, retrieving a subset of images based on similar criteria as the ones used in EHR systems, is a complex task for a medical researcher. In this paper, we propose an architecture to define cohorts over medical imaging data sets. This proposal uses a DICOM archive to index and to retrieve images, while the studies' selection is performed through a web application, ATLAS, which is normally used on observational studies upon EHR data. The presented architecture was validated using a public data set with synthetic EHR data.

The increasing interest in finding drugs and treatments for rare diseases led to the creation of research studies and clinical trials which data and results have been stored in multiple, heterogeneous databases. The lack of data harmonisation, combined with the need to improve current medical knowledge, has encouraged the research community to create computational solutions to aggregate this information. Although such platforms were created in the same area, orphan diseases, they were normally developed for different purposes, increasing the task complexity for end-users when needing to search for gene-to-phenotype information (e.g. genes, mutations, symptoms, etc.). Aiming to help answer these questions, we conducted a comprehensive analysis of the existent platforms designed to retrieve and visualise information about genetic rare diseases. In this analysis, we found several platforms from which we identified 7 candidates based on a set of inclusion and exclusion criteria. Through this analysis we were able to assess each system's characteristics and identify the most appropriate for distinct use cases and audiences, namely medical researchers, bioinformaticians and patients and relatives.

Cohort studies have been playing a key role in helping our understanding of diseases, health conditions, and treatments. These cohorts are often composed of a small number of subjects, especially in rare diseases studies, which reduces the statistical power of the results. One solution that can strengthen the scientific findings is to combine distinct studies and perform then multi-cohort analysis. However, even studies conducted for the same purpose in distinct research groups can have different scopes and medical observations, which preclude across-cohort exploration. In this paper, we propose a recommendation system to automatically discover cohorts of interest. This methodology uses context-based retrieval techniques combined with collaborative filtering to find relevant cohorts and scientific literature about a specific clinical investigation. The system was validated in a community focused on the study of Alzheimer's diseases, which includes 62 cohorts.

The exploration of multiple cohorts allows researchers to answer new research questions using more substantial clinical data. However, this is only possible if the cohorts are interoperable, which implies the migration of the original cohort into a common data schema. The problem of this procedure is the effort necessary to map the original concepts into their standard definitions. While several automatic mapping solutions can help in this task, its complexity increases when dealing with multi-language cohorts, leading to a significant manual effort in translating and mapping. In this paper we propose a system that combines text mining with language detection techniques, aiming to optimise these migration pipelines. This system was designed to be integrated into already existing migration workflows, without the need of adapting them. The system was validated using Alzheimer's diseases cohorts, but it is enough general to be applied in other use cases.

Clinical decision support systems (CDSS) are currently essential tools to guide medical diagnostics and patients' treatments, and they are specially important for the better care management of chronic diseases, such as cancer and diabetes. These systems help to decide on the best treatment solution, namely in centres where there is a shortage of medical experts. CDSS tools are often integrated into the Electronic Health Record (EHR) to facilitate the reuse of patient data. However, many times, creating new and intuitive protocols that are disease-specific is still a challenge. In this paper we present an open source solution (GenericCDSS) that can be used to streamline the development of autonomous CDSS, avoiding the dependency on third-party tools to manage patient data and clinical protocols. The software tool provides a modern user interface, supporting multi-platforms such as mobile and desktop devices. GenericCDSS is publicly available at https://github.com/bioinformatics-ua/GenericCDSS, under a GNU GPL license.

Hyperglycemia is a health condition characterized by abnormally high blood glucose, typically caused by a deficient usage, or lack, of insulin. Due to the metabolic derangements of this clinical condition, its regular monitoring, as well the administration of the most effective treatment, are major concerns for healthcare institutions. In this paper, we present a computational solution to build diabetes management protocols, which helps health professionals providing an adequate treatment for each hyperglycemic inpatient.

Medical imaging laboratories are supported by information and communication systems commonly denominated as PACS, that encompasses technology for acquisition, archive, distribution and visualization of digital images in network. Concerning the data and workflow management, traditional solutions used in production provide a limited set of services usually configured at system installation. As result, healthcare institutions are not able to fully explore their infrastructure or adapt it to new operational requirements, either for clinical or research procedures. This article proposes a framework for services orchestration and workflow management in distributed medical imaging environments. It was designed for end-user usage and is accessible through a Web portal that allows to document, repeat and allocate procedures and tasks to correct resources, either from information systems or human interventions. It provides an abstraction layer for integration with distinct data sources through standard services, allows the creation of new services through orchestration of existent ones and the scheduling of tasks. Moreover, it includes a logging and alert mechanism integrated with email service. The solution was validated through the specification of two use cases that were deployed in production environment.