No description available.

No description available.

Electronic Health Record (EHR) creates a huge number of databases which are being updated dynamically. Major goal of interoperability in healthcare is to facilitate the seamless exchange of healthcare related data and an environment to supports interoperability and secure transfer of data. The health care organisations face difficulties in exchanging patient’s health care information and laboratory reports etc. due to a lack of semantic interoperability. Hence, there is a need of semantic web technologies for addressing healthcare interoperability problems by enabling various healthcare standards from various healthcare entities (doctors, clinics, hospitals etc.) to exchange data and its semantics which can be understood by both machines and humans. Thus, a framework with a similarity analyser has been proposed in the thesis that dealt with semantic interoperability. While dealing with semantic interoperability, another consideration was the use of word embedding and ontology for knowledge discovery. In medical domain, the main challenge for medical information extraction system is to find the required information by considering explicit and implicit clinical context with high degree of precision and accuracy. For semantic similarity of medical text at different levels (conceptual, sentence and document level), different methods and techniques have been widely presented, but I made sure that the semantic content of a text that is presented includes the correct meaning of words and sentences. A comparative analysis of approaches included ontology followed by word embedding or vice-versa have been applied to explore the methodology to define which approach gives better results for gaining higher semantic similarity. Selecting the Kidney Cancer dataset as a use case, I concluded that both approaches work better in different circumstances. However, the approach in which ontology is followed by word embedding to enrich data first has shown better results. Apart from enriching the EHR, extracting relevant information is also challenging. To solve this challenge, the concept of analogy has been applied to explain similarities between two different contents as analogies play a significant role in understanding new concepts. The concept of analogy helps healthcare professionals to communicate with patients effectively and help them understand their disease and treatment. So, I utilised analogies in this thesis to support the extraction of relevant information from the medical text. Since accessing EHR has been challenging, tweets text is used as an alternative for EHR as social media has appeared as a relevant data source in recent years. An algorithm has been proposed to analyse medical tweets based on analogous words. The results have been used to validate the proposed methods. Two experts from medical domain have given their views on the proposed methods in comparison with the similar method named as SemDeep. The quantitative and qualitative results have shown that the proposed analogy-based method bring diversity and are helpful in analysing the specific disease or in text classification.

Rates of immune-mediated diseases (IMDs) have rapidly increased. Although the exact etiology has not yet been fully elucidated, disruptions to the microbiome has been proposed as a potential mechanism. We conducted a retrospective, longitudinal, birth cohort study utilizing electronic health records (EHR) to investigate the association between early life antibiotic exposure and the risk of developing juvenile idiopathic arthritis (JIA), pediatric psoriasis, or type 1 diabetes. Incident rate ratios (IRR) were estimated using modified Poisson regression models and adjusted for significant confounders. Children exposed to two or more antibiotics prior to 12 months of age had a 69% increased risk of developing JIA (1.69 IRR, 95% CI [1.04-2.73]), which rose to 97% when exposed prior to 6 months (1.97 IRR, 95% CI [1.11-3.49]). Children exposed to a penicillin antibiotic had a 62% increase in risk for psoriasis (1.62 IRR, 95% CI [1.06-2.49]), which rose slightly to 64% when exposure occurred between 6 and 12 months of age [(1.64 IRR, 95% CI [1.04-2.59]). We found a moderate to strong association between early antibiotic exposure and risk for JIA and psoriasis when exposure was examined by age, frequency, and type of antibiotic, but not for type 1 diabetes. Potential interactions effects between infection and antibiotics with an increased susceptibility to early life infections among children with an IMD was also observed. Overall, children exposed to antibiotics at an early age have an increased probability of developing an IMD after 12 months of age. However, alternative explanations for this association should be considered. / Public Health

<p dir="ltr">This research explores adversarial attacks on Graph Convolutional Transformer (GCT) models that utilize Electronic Health Record (EHR) data. As deep learning models become increasingly integral to healthcare, securing their robustness against adversarial threats is critical. This research assesses the susceptibility of GCT models to specific adversarial attacks, namely the Fast Gradient Sign Method (FGSM) and the Jacobian-based Saliency Map Attack (JSMA). It examines their effect on the model’s prediction of mortality and readmission. Through experiments conducted with the MIMIC-III and eICU datasets, the study finds that although the GCT model exhibits superior performance in processing EHR data under normal conditions, its accuracy drops when subjected to adversarial conditions—from an accuracy of 86% with test data to about 57% and an area under the curve (AUC) from 0.86 to 0.51. These findings averaged across both datasets and attack methods, underscore the urgent need for effective adversarial defense mechanisms in AI systems used in healthcare. This thesis contributes to the field by identifying vulnerabilities and suggesting various strategies to enhance the resilience of GCT models against adversarial manipulations.</p>

<p dir="ltr">Within Electronic Health Record (EHR) Systems, physicians face extensive documentation, leading to alarming mental burnout. The disproportionate focus on data entry over direct patient care underscores a critical concern. Integration of Natural Language Processing (NLP) powered EHR systems offers relief by reducing time and effort in record maintenance.</p><p dir="ltr">Our research introduces the Automated Electronic Health Record System, which not only transcribes dialogues but also employs advanced clinical text classification. With an accuracy exceeding 98.97%, it saves over 90% of time compared to manual entry, as validated on MIMIC III and MIMIC IV datasets.</p><p dir="ltr">In addition to our system's advancements, we explore integration of Diagnostic Decision Support System (DDSS) leveraging Large Language Models (LLMs) and transformers, aiming to refine healthcare documentation and improve clinical decision-making. We explore the advantages, like enhanced accuracy and contextual understanding, as well as the challenges, including computational demands and biases, of using various LLMs.</p>

No / This study aimed to evaluate the use of a shared electronic primary health care record (EHR) to assist with medicines reconciliation in the hospital from admission to discharge. Methods: This is a prospective cross-sectional, comparison evaluation for 2 phases, in a short-term elderly admissions ward in the United Kingdom. In phase 1, full reconciliation of the medication history was attempted, using conventional methods, before accessing the EHR, and then the EHR was used to verify the reconciliation. In phase 2, the EHR was the initial method of retrieving the medication history-validated by conventional methods. Results: Where reconciliation was led by conventional methods, and before any access to the EHR was attempted, 28 (28%) of hospital prescriptions were found to contain errors. Of 99 prescriptions subsequently checked using the EHR, only 50 (50%) matched the EHR. Of the remainder, 25% of prescriptions contained errors when verified by the EHR. However, 26% of patients had an incorrect list of current medications on the EHR. Using the EHR as the primary method of reconciliation, 33 (32%) of 102 prescriptions matched the EHR. Of those that did not match, 39 (38%) of prescriptions were found to contain errors. Furthermore, 37 (36%) of patients had an incorrect list of current medications on the EHR. The most common error type on the discharge prescription was drug omission; and on the EHR, wrong drug. Common potentially serious errors were related to unidentified allergies and adverse drug reactions. Conclusions: The EHR can reduce medication errors. However, the EHR should be seen as one of a range of information sources for reconciliation; the primary source being the patient or their carer. Both primary care and hospital clinicians should have read-and-write access to the EHR to reduce errors at care transitions. We recommend further evaluation studies.

This paper seeks to break down how a large scale Electronic Health Records system could improve quality of care and reduce monetary waste in the healthcare system. The paper further explores issues regarding regulations to data exchange and data interoperability. Due to the massive size of healthcare data, the exponential increase in the speed of data generation through innovative technologies, and the complexity of healthcare data types, the widespread of a large-scale EHR system has hit barriers. Much of the data available is unstructured or contained within a singular healthcare provider’s systems. To fully utilize all the data available, methods for making data interoperable and regulations for data exchange to protect and support patients must be made. Through angles addressing data exchange and interoperability, we seek to break down the constraints and issues that EHR systems still face and gain an understanding of the regulatory landscape.

Uppsatsen tar upp blockchain-teknologin och undersöker om den kan användas vid hantering av patientjournaler i ett svenskt patientjournalsystem. Genom att kombinera en genomförbarhetsstudie och SWOT-analys har tre aspekter; tekniska, legala och organisatoriska undersökts med hjälp av perspektiven från SWOT. De tekniska, legala och organisatoriska aspekterna berör viktiga faktorer och frågor som bör uppfyllas för ett projekt ska anses genomförbart. Uppsatsen är en explorativ fallstudie med dokumentanalys som metod. Primärt har akademisk litteratur samt lagar och författningar undersökts och analyserats. Med hjälp av informationen från insamlade dokument analyseras och diskuteras innehållet utifrån teknisk, legal genomförbarhet och organisatorisk genomförbarhet. Den tekniska samt legala aspekten antyder att blockchain i dess nuvarande form inte är genomförbart för att användas som stöd i svenskt patientjournalsystem. Emellertid har en hög genomförbarhet identifierats inom den organisatoriska aspekten. / This paper examines whether blockchain-technology can be used to assist an EHR system (electronic health records) in Sweden. By combining a feasibility study and SWOT-analysis, three major aspects; technical, legal and organizational, have been examined with help from the perspectives in SWOT. The aspects cover key factors and questions which ought to be fulfilled for a project to be considered feasible. An exploratory case study has been conducted combined with the method of document analysis. The documents have primarily consisted of academic literature and law acts and constitutions such as GDPR. The feasibility aspects have been analysed and discussed with the help of found literature. The technical and legal aspects suggest that the use of blockchain in its current shape and form is not feasible in assisting an EHR system. However, within the organizational aspect, a high level of feasibility has been concluded

The purpose of this master thesis was to investigate if using both X-ray images and patient's register data could increase the performance of a neural network in discrimination of two types of fractures in the thigh bone, called atypical femoral fractures (AFF) and normal femoral fractures (NFF). We also examined and evaluated how the fusion of the two data types could be done and how different types of fusion affect the performance. Finally, we evaluated how the number of variables in the register data affect a network's performance. Our image dataset consisted of 1,442 unique images from 580 patients (16.85% of the images were labelled AFF corresponding to 15.86% of the patients). Since the dataset is very imbalanced, sensitivity is a prioritized evaluation metric. The register data network was evaluated using five different versions of register data parameters: two (age and sex), seven (binary and non-binary) and 44 (binary and non-binary). Having only age and sex as input resulted in a classifier predicting all samples to class 0 (NFF), for all tested network architectures. Using a certain network structure (celled register data model 2), in combination with the seven non-binary parameters outperforms using both two and 44 (both binary and non-binary) parameters regarding mean AUC and sensitivity. Highest mean accuracy is obtained by using 44 non-binary parameters. The seven register data parameters have a known connection to AFF and includes age and sex. The network with X-ray images as input uses a transfer learning approach with a pre-trained ResNet50-base. This model performed better than all the register data models, regarding all considered evaluation metrics.        Three fusion architectures were implemented and evaluated: probability fusion (PF), feature fusion (FF) and learned feature fusion (LFF). PF concatenates the prediction provided from the two separate baseline models. The combined vector is fed into a shallow neural network, which are the only trainable part in this architecture. FF fuses a feature vector provided from the image baseline model, with the raw register data parameters. Prior to the concatenation both vectors were normalized and the fused vector is then fed into a shallow trainable network. The final architecture, LFF, does not have completely frozen baseline models but instead learns two separate feature vectors. These feature vectors are then concatenated and fed into a shallow neural network to obtain a final prediction. The three fusion architectures were evaluated twice: using seven non-binary register data parameters, or only age and sex. When evaluated patient-wise, all three fusion architectures using the seven non-binary parameters obtain higher mean AUC and sensitivity than the single modality baseline models. All fusion architectures with only age and sex as register data parameters results in higher mean sensitivity than the baseline models. Overall, probability fusion with the seven non-binary parameters results in the highest mean AUC and sensitivity, and learned feature fusion with the seven non-binary parameters results in the highest mean accuracy.