Transformers for extracting breast cancer information from Spanish clinical narratives.

The wide adoption of electronic health records (EHRs) offers immense potential as a source of support for clinical research. However, previous studies focused on extracting only a limited set of medical concepts to support information extraction in the cancer domain for the Spanish language. Building on the success of deep learning for processing natural language texts, this paper proposes a transformer-based approach to extract named entities from breast cancer clinical notes written in Spanish and compares several language models. To facilitate this approach, a schema for annotating clinical notes with breast cancer concepts is presented, and a corpus for breast cancer is developed. Results indicate that both BERT-based and RoBERTa-based language models demonstrate competitive performance in clinical Named Entity Recognition (NER). Specifically, BETO and multilingual BERT achieve F-scores of 93.71% and 94.63%, respectively. Additionally, RoBERTa Biomedical attains an F-score of 95.01%, while RoBERTa BNE achieves an F-score of 94.54%. The findings suggest that transformers can feasibly extract information in the clinical domain in the Spanish language, with the use of models trained on biomedical texts contributing to enhanced results. The proposed approach takes advantage of transfer learning techniques by fine-tuning language models to automatically represent text features and avoiding the time-consuming feature engineering process.

Machine Learning-Assisted Recurrence Prediction for Patients With Early-Stage Non-Small-Cell Lung Cancer.

PURPOSE: Stratifying patients with cancer according to risk of relapse can personalize their care. In this work, we provide an answer to the following research question: How to use machine learning to estimate probability of relapse in patients with early-stage non-small-cell lung cancer (NSCLC)? MATERIALS AND METHODS: For predicting relapse in 1,387 patients with early-stage (I-II) NSCLC from the Spanish Lung Cancer Group data (average age 65.7 years, female 24.8%, male 75.2%), we train tabular and graph machine learning models. We generate automatic explanations for the predictions of such models. For models trained on tabular data, we adopt SHapley Additive exPlanations local explanations to gauge how each patient feature contributes to the predicted outcome. We explain graph machine learning predictions with an example-based method that highlights influential past patients. RESULTS: Machine learning models trained on tabular data exhibit a 76% accuracy for the random forest model at predicting relapse evaluated with a 10-fold cross-validation (the model was trained 10 times with different independent sets of patients in test, train, and validation sets, and the reported metrics are averaged over these 10 test sets). Graph machine learning reaches 68% accuracy over a held-out test set of 200 patients, calibrated on a held-out set of 100 patients. CONCLUSION: Our results show that machine learning models trained on tabular and graph data can enable objective, personalized, and reproducible prediction of relapse and, therefore, disease outcome in patients with early-stage NSCLC. With further prospective and multisite validation, and additional radiological and molecular data, this prognostic model could potentially serve as a predictive decision support tool for deciding the use of adjuvant treatments in early-stage lung cancer.

An Artificial Intelligence-Based Tool for Data Analysis and Prognosis in Cancer Patients: Results from the Clarify Study.

BACKGROUND: Artificial intelligence (AI) has contributed substantially in recent years to the resolution of different biomedical problems, including cancer. However, AI tools with significant and widespread impact in oncology remain scarce. The goal of this study is to present an AI-based solution tool for cancer patients data analysis that assists clinicians in identifying the clinical factors associated with poor prognosis, relapse and survival, and to develop a prognostic model that stratifies patients by risk. MATERIALS AND METHODS: We used clinical data from 5275 patients diagnosed with non-small cell lung cancer, breast cancer, and non-Hodgkin lymphoma at Hospital Universitario Puerta de Hierro-Majadahonda. Accessible clinical parameters measured with a wearable device and quality of life questionnaires data were also collected. RESULTS: Using an AI-tool, data from 5275 cancer patients were analyzed, integrating clinical data, questionnaires data, and data collected from wearable devices. Descriptive analyses were performed in order to explore the patients' characteristics, survival probabilities were calculated, and a prognostic model identified low and high-risk profile patients. CONCLUSION: Overall, the reconstruction of the population's risk profile for the cancer-specific predictive model was achieved and proved useful in clinical practice using artificial intelligence. It has potential application in clinical settings to improve risk stratification, early detection, and surveillance management of cancer patients.

Negation and uncertainty detection in clinical texts written in Spanish: a deep learning-based approach.

Detecting negation and uncertainty is crucial for medical text mining applications; otherwise, extracted information can be incorrectly identified as real or factual events. Although several approaches have been proposed to detect negation and uncertainty in clinical texts, most efforts have focused on the English language. Most proposals developed for Spanish have focused mainly on negation detection and do not deal with uncertainty. In this paper, we propose a deep learning-based approach for both negation and uncertainty detection in clinical texts written in Spanish. The proposed approach explores two deep learning methods to achieve this goal: (i) Bidirectional Long-Short Term Memory with a Conditional Random Field layer (BiLSTM-CRF) and (ii) Bidirectional Encoder Representation for Transformers (BERT). The approach was evaluated using NUBES and IULA, two public corpora for the Spanish language. The results obtained showed an F-score of 92% and 80% in the scope recognition task for negation and uncertainty, respectively. We also present the results of a validation process conducted using a real-life annotated dataset from clinical notes belonging to cancer patients. The proposed approach shows the feasibility of deep learning-based methods to detect negation and uncertainty in Spanish clinical texts. Experiments also highlighted that this approach improves performance in the scope recognition task compared to other proposals in the biomedical domain.

Influenza and Measles-MMR: two case study of the trend and impact of vaccine-related Twitter posts in Spanish during 2015-2018.

Social media, and in particularly Twitter, can be a resource of enormous value to retrieve information about the opinion of general populaton to vaccines. The increasing popularity of this social media has allowed to use its content to have a clear picture of their users on this topic. In this paper, we perform a study about vaccine-related messages published in Spanish during 2015-2018. More specifically, the paper has focused on two specific diseases: influenza and measles (and MMR as its vaccine). By also including an analysis about the sentiment expressed on the published tweets, we have been able to identify the type of messages that are published on Twitter with respect these two pathologies and their vaccines. Results showed that in contrary on popular opinions, most of the messages published are non-negative. On the other hand, the analysis showed that some messages attracted a huge attention and provoked peaks in the number of published tweets, explaining some changes in the observed trends.

Unravelling data for rapid evidence-based response to COVID-19: a summary of the unCoVer protocol.

INTRODUCTION: unCoVer-Unravelling data for rapid evidence-based response to COVID-19-is a Horizon 2020-funded network of 29 partners from 18 countries capable of collecting and using real-world data (RWD) derived from the response and provision of care to patients with COVID-19 by health systems across Europe and elsewhere. unCoVer aims to exploit the full potential of this information to rapidly address clinical and epidemiological research questions arising from the evolving pandemic. METHODS AND ANALYSIS: From the onset of the COVID-19 pandemic, partners are gathering RWD from electronic health records currently including information from over 22 000 hospitalised patients with COVID-19, and national surveillance and screening data, and registries with over 1 900 000 COVID-19 cases across Europe, with continuous updates. These heterogeneous datasets will be described, harmonised and integrated into a multi-user data repository operated through Opal-DataSHIELD, an interoperable open-source server application. Federated data analyses, without sharing or disclosing any individual-level data, will be performed with the objective to reveal patients' baseline characteristics, biomarkers, determinants of COVID-19 prognosis, safety and effectiveness of treatments, and potential strategies against COVID-19, as well as epidemiological patterns. These analyses will complement evidence from efficacy/safety clinical trials, where vulnerable, more complex/heterogeneous populations and those most at risk of severe COVID-19 are often excluded. ETHICS AND DISSEMINATION: After strict ethical considerations, databases will be available through a federated data analysis platform that allows processing of available COVID-19 RWD without disclosing identification information to analysts and limiting output to data aggregates. Dissemination of unCoVer's activities will be related to the access and use of dissimilar RWD, as well as the results generated by the pooled analyses. Dissemination will include training and educational activities, scientific publications and conference communications.

Predictors of health-related quality of life in musculoskeletal disease patients: a longitudinal analysis.

INTRODUCTION: Rheumatic and musculoskeletal diseases (RMDs) have a significant impact on patients' health-related quality of life (HRQoL) exacerbating disability, reducing independence and work capacity, among others. Predictors' identification affecting HRQoL could help to place efforts that minimize the deleterious impact of these conditions on patients' wellbeing. This study evaluates the influence of demographic and clinical predictors on the HRQoL of a cohort of RMD patients, measured using the Rosser classification index (RCI). METHODS: We included patients attending the Hospital Clínico San Carlos (HCSC) rheumatology outpatient clinic from 1 April 2007 to 30 November 2017. The primary outcome was the HRQoL assessed in each of the patient's visits using the RCI. Demographic and clinical variables extracted from a departmental electronic health record (EHR) were used as predictors: RMD diagnoses, treatments, comorbidities, and averaged HRQoL values from previous periods (for this last variable, values were imputed if no information was available). Association between predictors and HRQoL was analyzed using penalized generalized estimating equations (PGEEs). To account for imputation bias, the PGEE model was repeated excluding averaged HRQoL predictors, and common predictors were considered. DISCUSSION: A total of 18,187 outpatients with 95,960 visits were included. From 410 initial predictors, 19 were independently associated with patients' HRQoL in both PGEE models. Chronic kidney disease (CKD), an episode of prescription of third level analgesics, monoarthritis, and fibromyalgia diagnoses were associated with worse HRQoL. Conversely, the prescription in the previous visit of acid-lowering medication, colchicine, and third level analgesics was associated with better HRQoL. CONCLUSION: We have identified several diagnoses, treatments, and comorbidities independently associated with HRQoL in a cohort of outpatients attending a rheumatology clinic.

To burn-out or not to burn-out: a cross-sectional study in healthcare professionals in Spain during COVID-19 pandemic.

OBJECTIVE: To assess the prevalence of burn-out syndrome in healthcare workers working on the front line (FL) in Spain during COVID-19. DESIGN: Cross-sectional, online survey-based study. SETTINGS: Sampling was performed between 21st April and 3rd May 2020. The survey collected demographic data and questions regarding participants' working position since pandemic outbreak. PARTICIPANTS: Spanish healthcare workers working on the FL or usual ward were eligible. A total of 674 healthcare professionals answered the survey. MAIN OUTCOMES AND MEASURES: Burn-out syndrome was assessed by the Maslach Burnout Inventory-Medical Personnel. RESULTS: Of the 643 eligible responding participants, 408 (63.5%) were physicians, 172 (26.8%) were nurses and 63 (9.8%) other technical occupations. 377 (58.6%) worked on the FL. Most participants were women (472 (73.4%)), aged 31-40 years (163 (25.3%)) and worked in tertiary hospitals (>600 beds) (260 (40.4%)). Prevalence of burn-out syndrome was 43.4% (95% CI 39.5% to 47.2%), higher in COVID-19 FL workers (49.6%, p<0.001) than in non- COVID-19 FL workers (34.6%, p<0.001). Women felt more burn-out (60.8%, p=0.016), were more afraid of self-infection (61.9%, p=0.021) and of their performance and quality of care provided to the patients (75.8%, p=0.015) than men. More burn-out were those between 20 and 30 years old (65.2%, p=0.026) and those with more than 15 years of experience (53.7%, p=0.035).Multivariable logistic regression analysis revealed that, working on COVID-19 FL (OR 1.93; 95% CI 1.37 to 2.71, p<0.001), being a woman (OR 1.56; 95% CI 1.06 to 2.29, p=0.022), being under 30 years old (OR 1.75; 95% CI 1.06 to 2.89, p=0.028) and being a physician (OR 1.64; 95% CI 1.11 to 2.41, p=0.011) were associated with high risk of burn-out syndrome. CONCLUSIONS: This survey study of healthcare professionals reported high rates of burn-out syndrome. Interventions to promote mental well-being in healthcare workers exposed to COVID-19 need to be immediately implemented.

Normal tissue content impact on the GBM molecular classification.

Molecular classification of glioblastoma has enabled a deeper understanding of the disease. The four-subtype model (including Proneural, Classical, Mesenchymal and Neural) has been replaced by a model that discards the Neural subtype, found to be associated with samples with a high content of normal tissue. These samples can be misclassified preventing biological and clinical insights into the different tumor subtypes from coming to light. In this work, we present a model that tackles both the molecular classification of samples and discrimination of those with a high content of normal cells. We performed a transcriptomic in silico analysis on glioblastoma (GBM) samples (n = 810) and tested different criteria to optimize the number of genes needed for molecular classification. We used gene expression of normal brain samples (n = 555) to design an additional gene signature to detect samples with a high normal tissue content. Microdissection samples of different structures within GBM (n = 122) have been used to validate the final model. Finally, the model was tested in a cohort of 43 patients and confirmed by histology. Based on the expression of 20 genes, our model is able to discriminate samples with a high content of normal tissue and to classify the remaining ones. We have shown that taking into consideration normal cells can prevent errors in the classification and the subsequent misinterpretation of the results. Moreover, considering only samples with a low content of normal cells, we found an association between the complexity of the samples and survival for the three molecular subtypes.

Reconstructing the patient's natural history from electronic health records.

The automatic extraction of a patient's natural history from Electronic Health Records (EHRs) is a critical step towards building intelligent systems that can reason about clinical variables and support decision making. Although EHRs contain a large amount of valuable information about the patient's medical care, this information can only be fully understood when analyzed in a temporal context. Any intelligent system should then be able to extract medical concepts, date expressions, temporal relations and the temporal ordering of medical events from the free texts of EHRs; yet, this task is hard to tackle, due to the domain specific nature of EHRs, writing quality and lack of structure of these texts, and more generally the presence of redundant information. In this paper, we introduce a new Natural Language Processing (NLP) framework, capable of extracting the aforementioned elements from EHRs written in Spanish using rule-based methods. We focus on building medical timelines, which include disease diagnosis and its progression over time. By using a large dataset of EHRs comprising information about patients suffering from lung cancer, we show that our framework has an adequate level of performance by correctly building the timeline for 843 patients from a pool of 989 patients, achieving a precision of 0.852.

Sex is a strong prognostic factor in stage IV non-small-cell lung cancer patients and should be considered in survival rate estimation.

BACKGROUND: Biological differences between the sexes have a major impact on disease and treatment outcome. In this paper, we evaluate the prognostic value of sex in stage IV non-small-cell lung cancer (NSCLC) in the context of routine clinical data, and compare this information with other external datasets. METHODS: Clinical data from stage IV NSCLC patients from Hospital Puerta de Hierro (HPH) were retrieved from electronic health records using big data analytics (N = 397). In addition, data from the Spanish Lung Cancer Group (GECP) Tumor Registry (N = 1382) and from a published study available from the cBioPortal (MSK) (N = 601) were analyzed. Survival curves were estimated using the Kaplan-Meier method. A Cox proportional hazards regression model was used to assess the prognostic value of sex. A meta-analysis to compare the outcome for males and females in terms of overall survival (OS) and progression free survival (PFS) was performed. RESULTS: The median OS time was 12 months for males and 19 months for females (overall HR = 0.77; 95% CI: 0.68-0.87; P < 0.001). Similarly, females with stage IV NSCLC harboring an EGFR-sensitizing mutation lived significantly longer than males (median OS: males, 19 months; females, 32 months) with a lower risk of death compared with males (overall HR = 0.75; 95% CI: 0.67-0.84). In addition, female patients benefited more from EGFR inhibitors in terms of PFS and OS (overall HR = 0.45; 95% CI: 0.32-0.64, and HR = 0.62; 95% CI: 0.48-0.80, respectively). Median PFS was 21 months in females and 12 months in males (P < 0.001). CONCLUSIONS: Using routine clinical data we confirmed the previous finding that among stage IV NSCLC patients, females had a significantly better prognosis than males. The effect size of the sex was notable, highlighting the fact that survival rates are usually estimated and patients are generally managed without considering the sexes separately, which may lead to suboptimal results.

Importance of medical data preprocessing in predictive modeling and risk factor discovery for the frailty syndrome.

BACKGROUND: Increasing life expectancy results in more elderly people struggling with age related diseases and functional conditions. This poses huge challenges towards establishing new approaches for maintaining health at a higher age. An important aspect for age related deterioration of the general patient condition is frailty. The frailty syndrome is associated with a high risk for falls, hospitalization, disability, and finally increased mortality. Using predictive data mining enables the discovery of potential risk factors and can be used as clinical decision support system, which provides the medical doctor with information on the probable clinical patient outcome. This enables the professional to react promptly and to avert likely adverse events in advance. METHODS: Medical data of 474 study participants containing 284 health related parameters, including questionnaire answers, blood parameters and vital parameters from the Toledo Study for Healthy Aging (TSHA) was used. Binary classification models were built in order to distinguish between frail and non-frail study subjects. RESULTS: Using the available TSHA data and the discovered potential predictors, it was possible to design, develop and evaluate a variety of different predictive models for the frailty syndrome. The best performing model was the support vector machine (SVM, 78.31%). Moreover, a methodology was developed, making it possible to explore and to use incomplete medical data and further identify potential predictors and enable interpretability. CONCLUSIONS: This work demonstrates that it is feasible to use incomplete, imbalanced medical data for the development of a predictive model for the frailty syndrome. Moreover, potential predictive factors have been discovered, which were clinically approved by the clinicians. Future work will improve prediction accuracy, especially with regard to separating the group of frail patients into frail and pre-frail ones and analyze the differences among them.

Understanding diseases as increased heterogeneity: a complex network computational framework.

Owing to the complexity of the human body, most diseases present a high interpersonal variability in the way they manifest, i.e. in their phenotype, which has important clinical repercussions-for instance, the difficulty in defining objective diagnostic rules. Here we explore the hypothesis that signs and symptoms used to define a disease should be understood in terms of the dispersion (as opposed to the average) of physical observables. To that end, we propose a computational framework, based on complex networks theory, to map groups of subjects to a network structure, based on their pairwise phenotypical similarity. We demonstrate that the resulting structure can be used to improve the performance of classification algorithms, especially in the case of a limited number of instances, with both synthetic and real datasets. Beyond providing an alternative conceptual understanding of diseases, the proposed framework could be of special relevance in the growing field of personalized, or N-to-1, medicine.

PET-CT image fusion using random forest and à-trous wavelet transform.

New image fusion rules for multimodal medical images are proposed in this work. Image fusion rules are defined by random forest learning algorithm and a translation-invariant à-trous wavelet transform (AWT). The proposed method is threefold. First, source images are decomposed into approximation and detail coefficients using AWT. Second, random forest is used to choose pixels from the approximation and detail coefficients for forming the approximation and detail coefficients of the fused image. Lastly, inverse AWT is applied to reconstruct fused image. All experiments have been performed on 198 slices of both computed tomography and positron emission tomography images of a patient. A traditional fusion method based on Mallat wavelet transform has also been implemented on these slices. A new image fusion performance measure along with 4 existing measures has been presented, which helps to compare the performance of 2 pixel level fusion methods. The experimental results clearly indicate that the proposed method outperforms the traditional method in terms of visual and quantitative qualities and the new measure is meaningful.

Acronym Disambiguation in Spanish Electronic Health Narratives Using Machine Learning Techniques.

Electronic Health Records (EHRs) are now being massively used in hospitals what has motivated current developments of new methods to process clinical narratives (unstructured data) making it possible to perform context-based searches. Current approaches to process the unstructured texts in EHRs are based in applying text mining or natural language processing (NLP) techniques over the data. In particular Named Entity Recognition (NER) is of paramount importance to retrieve specific biomedical concepts from the text providing the semantic type of the concept retrieved. However, it is very common that clinical notes contain lots of acronyms that cannot be identified by NER processes and even if they are identified, an acronym may correspond to several meanings, so disambiguation of the found term is needed. In this work we provide an approach to perform acronym disambiguation in Spanish EHR using machine learning techniques.

A multicenter study of the early detection of synaptic dysfunction in Mild Cognitive Impairment using Magnetoencephalography-derived functional connectivity.

Synaptic disruption is an early pathological sign of the neurodegeneration of Dementia of the Alzheimer's type (DAT). The changes in network synchronization are evident in patients with Mild Cognitive Impairment (MCI) at the group level, but there are very few Magnetoencephalography (MEG) studies regarding discrimination at the individual level. In an international multicenter study, we used MEG and functional connectivity metrics to discriminate MCI from normal aging at the individual person level. A labeled sample of features (links) that distinguished MCI patients from controls in a training dataset was used to classify MCI subjects in two testing datasets from four other MEG centers. We identified a pattern of neuronal hypersynchronization in MCI, in which the features that best discriminated MCI were fronto-parietal and interhemispheric links. The hypersynchronization pattern found in the MCI patients was stable across the five different centers, and may be considered an early sign of synaptic disruption and a possible preclinical biomarker for MCI/DAT.

Intelligent Therapy Assistant (ITA) for cognitive rehabilitation in patients with acquired brain injury.

BACKGROUND: This paper presents the design, development and first evaluation of an algorithm, named Intelligent Therapy Assistant (ITA), which automatically selects, configures and schedules rehabilitation tasks for patients with cognitive impairments after an episode of Acquired Brain Injury. The ITA is integrated in "Guttmann, Neuro Personal Trainer" (GNPT), a cognitive tele-rehabilitation platform that provides neuropsychological services. METHODS: The ITA selects those tasks that are more suitable for the specific needs of each patient, considering previous experiences, and improving the personalization of the treatment. The system applies data mining techniques to cluster the patients according their cognitive impairment profile. Then, the algorithm rates every rehabilitation task, based on its cognitive structure and the clinical impact of executions done by similar patients. Finally, it configures the most suitable degree of difficulty, depending on the impairment of the patient and his/her evolution during the treatment. RESULTS: The ITA has been evaluated during 18 months by 582 patients. In order to evaluate the effectiveness of the ITA, a comparison between the traditional manual planning procedure and the one presented in this paper has been done, taking into account: a) the selected tasks assigned to rehabilitation sessions; b) the difficulty level configured for the sessions; c) and the improvement of their cognitive capacities after completing treatment. CONCLUSIONS: The obtained results reveal that the rehabilitation treatment proposed by the ITA is as effective as the one performed manually by therapists, arising as a new powerful support tool for therapists. The obtained results make us conclude that the proposal done by the ITA is very close to the one done by therapists, so it is suitable for real treatments.

Mining recurring concepts in a dynamic feature space.

Most data stream classification techniques assume that the underlying feature space is static. However, in real-world applications the set of features and their relevance to the target concept may change over time. In addition, when the underlying concepts reappear, reusing previously learnt models can enhance the learning process in terms of accuracy and processing time at the expense of manageable memory consumption. In this paper, we propose mining recurring concepts in a dynamic feature space (MReC-DFS), a data stream classification system to address the challenges of learning recurring concepts in a dynamic feature space while simultaneously reducing the memory cost associated with storing past models. MReC-DFS is able to detect and adapt to concept changes using the performance of the learning process and contextual information. To handle recurring concepts, stored models are combined in a dynamically weighted ensemble. Incremental feature selection is performed to reduce the combined feature space. This contribution allows MReC-DFS to store only the features most relevant to the learnt concepts, which in turn increases the memory efficiency of the technique. In addition, an incremental feature selection method is proposed that dynamically determines the threshold between relevant and irrelevant features. Experimental results demonstrating the high accuracy of MReC-DFS compared with state-of-the-art techniques on a variety of real datasets are presented. The results also show the superior memory efficiency of MReC-DFS.

Parenclitic networks: uncovering new functions in biological data.

We introduce a novel method to represent time independent, scalar data sets as complex networks. We apply our method to investigate gene expression in the response to osmotic stress of Arabidopsis thaliana. In the proposed network representation, the most important genes for the plant response turn out to be the nodes with highest centrality in appropriately reconstructed networks. We also performed a target experiment, in which the predicted genes were artificially induced one by one, and the growth of the corresponding phenotypes compared to that of the wild-type. The joint application of the network reconstruction method and of the in vivo experiments allowed identifying 15 previously unknown key genes, and provided models of their mutual relationships. This novel representation extends the use of graph theory to data sets hitherto considered outside of the realm of its application, vastly simplifying the characterization of their underlying structure.

Feature selection in the reconstruction of complex network representations of spectral data.

Complex networks have been extensively used in the last decade to characterize and analyze complex systems, and they have been recently proposed as a novel instrument for the analysis of spectra extracted from biological samples. Yet, the high number of measurements composing spectra, and the consequent high computational cost, make a direct network analysis unfeasible. We here present a comparative analysis of three customary feature selection algorithms, including the binning of spectral data and the use of information theory metrics. Such algorithms are compared by assessing the score obtained in a classification task, where healthy subjects and people suffering from different types of cancers should be discriminated. Results indicate that a feature selection strategy based on Mutual Information outperforms the more classical data binning, while allowing a reduction of the dimensionality of the data set in two orders of magnitude.