Abstract: Background: Seasonal Influenza is considered to be a cyclic and ordered sequence of values, influenced by external factors that can be predicted and used to detect disease outbreaks and monitoring. In machine learning, the key challenges that limit these analyses are in model explainability and limitations associated with ecological bias. Aim: Determine the best environmental variable selection method to predict Seasonal Influenza in Norway, using an environmental medicine approach combined with machine learning techniques. Methods: This is a quasi-experimental study that compares three approaches (non-variable selection, isolate component, and multipollutant mixture), represented in five methods (univariable, bivariable, multivariable AME, multivariable PCA, multivariable LDA). Per method, the best co-variable combination will be performed, following the internal rules of each method. The best covariable combination is the result of three components: variable selection, validation data set and lag. The first one involves 13 environmental variables (temperature, relative humidity, specific humidity, air pressure, wind speed, precipitation, CO, NO, NO2, O3, PM10, PM2,5 and SO2); second one compares a test dataset compiled from the data from 2019, last year (2018) and a synthetic environmental (avg. 2013-2018) data set in the validation process; and the third one compares a combination of lag from 0 to 12. All the predictions are made using ARIMA algorithm. The evaluation is given in terms of MAE, MSE, RMSE, OR. The training set is from 2 Jun 2013 (week 22/2013) to 28 May 2018 (week 21/2018), and test set is from (week 22/2018) to (week 21/2019) with a window of predictions of 52 weeks. Results: The increment of dimensionality in the environmental variable selection introduce different noise levels and optimize the prediction. Considerations that impact the explainability, usability, ecological bias and performance will be described. Conclusion: The increment of dimensionality in the variable selection has a better impact on performance than using complex algorithms.

Abstract: Biological and physiological changes in response to stress are crucial to achieving adaption to environmental constraints. Typical stress responses activate the sympathetic nervous system and the hypothalamic-pituitary-adrenal cortical axis, which aim to maintain homeostasis in response to demands and influence various biological responses at neuroendocrine, cellular, and immune levels. In particular, there is compelling evidence showing the devastating effects of high-stress levels on immune responses and brain functionality. In that sense, data suggest that a sustained stress-physiological response predicts decreases in immunity and leads to allostatic load, the wear, and tear of biological systems. Similarly, in cancer research, chronic stress levels have been reported to impair the bodysurveillance systems and be related to cancer incidence, progression, and mortality. Despite the evidence demonstrating the associations between stress and cancer, there are limited data on whether lifetime stress exposure relates to clinical indicators of cancer aggressiveness. Considering the evidence showing the adverse effects of prolonged stress exposure on cancer disease, this thesis aimed to explore the relationship between lifetime stress exposure and the clinical characteristics of tumor-aggressiveness in a cohort of patients affected with triple-negative breast cancer (TNBC). Likewise, as research has shown that coping responses moderate the stress-physiological response, we aimed to investigate the role of coping responses in moderating the association between tumor aggressiveness and lifetime stress exposure. According to the scientific literature, high-stress levels would increase physiological arousal and allostatic load favoring a debilitated immune system. Thus, we hypothesized finding positive associations between lifetime stress exposure and tumor-aggressiveness. In the same line, we suspected that adaptative coping styles would be negatively related to tumoraggressiveness. The participants of this study were 29 women diagnosed with TNBC who were treated in the University Hospital for Gynecology Pius-Hospital (Oldenburg, Germany). Correlation and moderation regression analyses with the use of composite scores were conducted. Overall, we could not demonstrate significant associations between lifetime stress exposure and coping responses on the clinical indicators of tumor aggressiveness in this cohort of patients. Possible explanations for these results are discussed, and suggestions for future research are provided. This study was the first to investigate the effect of lifetime stress exposure and coping response on the aggressive tumor characteristics of TNBC. We provided a deep theoretical framework to understand the interplay between biological and psychological aspects of cancer disease, and we hope that our recommendations may serve as a base for upcoming studies.

Abstract: Despite some epidemiological evidence on the protective effects of polyphenol intake on epithelial ovarian cancer (EOC) risk from case-control studies, the evidence is scarce from prospective studies and non-existent for several polyphenol classes. Therefore, we aimed to investigate the associations between the intake of total, classes and subclasses of polyphenols and EOC risk in a large prospective study. The study was conducted in the European Prospective Investigation into Cancer and Nutrition (EPIC) cohort, which included 309,129 adult women recruited mostly from the general population. Polyphenol intake was assessed through validated country-specific dietary questionnaires and the Phenol-Explorer database. During a mean follow-up of 14 years, 1469 first incident EOC cases (including 806 serous, 129 endometrioid, 102 mucinous, and 67 clear cell tumours) were identified. In multivariable-adjusted Cox regression models, the hazard ratio in the highest quartile of total polyphenol intake compared with the lowest quartile (HRQ4vsQ1) was 1.14 (95% CI 0.94–1.39; p-trend = 0.11). Similarly, the intake of most classes and subclasses of polyphenols were not related to either overall EOC risk or any EOC subtype. A borderline statistically significant positive association was observed between phenolic acid intake (HRQ4vsQ1 = 1.20, 95% CI 1.01–1.43; p-trend = 0.02) and EOC risk, especially for the serous subtype and in women with obesity, although these associations did not exceed the Bonferroni correction threshold. The current results do not support any association between polyphenol intake and EOC in our large European prospective study. Results regarding phenolic acid intake need further investigation.

Abstract: Introduction: The occlusion of coronary grafts used during myocardial revascularization surgery is a relatively frequent event. It has been described that between 7–15% are affected before 24 hours, and up to 30% per year being mainly related to technical errors. Pulsed Doppler offers the necessary requirements for an adequate intraoperative evaluation of grafts. The main objective of this study is to demonstrate it using the Fluxvisum prototype. Material and method: Longitudinal study carried out between March 2012 and June 2016 in surgically revascularized patients in which the elaborated coronary grafts were evaluated intraoperatively using pulsed Doppler. Results: 110 patients were studied, predominantly men older than 60 years, with a high prevalence of HT, DM and previous AMI, and diagnosis on admission to NSTEMI. The average measurements were: Qmed:31.7ml/min, PI:3.2, DFI:70.2. The Qmed was the flowmeter variable that was most statistically associated with the appearance of major cardiac events. In the logistic regression, the quality of LAD revascularization stood out as the main predictor of MACE (p=0.050, OR:2.599, 95%CI:0.972-6.947), death (p=0.004, OR:13.947, 95%CI:2,311-84,163), AMI (p=0.000, OR:11.331, 95%CI:2.995-42.867) and perioperative acute ischemia (p=0.049, OR:2.864, 95%CI:0.997-8.222). Conclusions: Pulsed Doppler flowmetry proved to be a tool with high prognostic power as a predictor of postoperative comorbidity and major cardiac events in surgically revascularized patients. Resumen: Introducción: La oclusión de los injertos coronarios utilizados durante la cirugía de revascularización miocárdica es un hecho relativamente frecuente. Se ha descrito que entre el 7–15% se afectan antes de las 24 horas, y hasta un 30% al año relacionándose principalmente con errores técnicos. El Doppler pulsado ofrece los requerimientos necesarios para una adecuada evaluación intraoperatoria de los injertos. El objetivo principal de este estudio es demostrarlo utilizando el prototipo Fluxvisum. Material y método: Estudio longitudinal realizado entre marzo de 2012 y junio del 2016 en pacientes revascularizados quirúrgicamente en los que los injertos coronarios elaborados fueron evaluados intraoperatoriamente utilizando Doppler pulsado. Resultados: Se estudiaron 110 enfermos predominantemente hombres mayores de 60 años, con una alta prevalencia de hipertensión arterial (HTA), diabetes mellitus (DM) e infarto agudo de miocardio (IMA) previo, y diagnostico al ingreso de SCASEST. Las mediciones promedio fueron: Qmed:31,7ml/min, IP:3,2, DFI:70,2. El Qmed fue la variable flujométrica que más se asoció estadísticamente con la aparición de eventos cardiacos mayores. En la regresión logística, la calidad de la revascularización de la DA destacó como el principal factor predictor de MACE (p=0,050, OR:2,599, CI95%:0,972–6,947), muerte (p=0,004, OR:13,947, CI95%:2,311–84,163), IMA (p=0,000, OR:11,331, CI95%:2,995–42,867) e isquemia aguda perioperatoria (p=0,049, OR:2,864, CI95%:0,997–8,222). Conclusiones: La flujometría con Doppler pulsado demostró ser una herramienta que tiene un alto poder pronóstico como predictor de comorbilidad posoperatoria y eventos cardiacos mayores en pacientes revascularizados quirúrgicamente.

Abstract: BACKGROUND: Attention-deficit/hyperactivity disorder (ADHD) is a neurodevelopmental disorder associated with significant morbidity and mortality that may affect over 5% of children and approximately 2.8% of adults worldwide. Pharmacological and behavioral therapies for ADHD exist, but critical symptoms such as dysexecutive deficits remain unaffected. In a randomized, sham-controlled, double-blind, crossover mechanistic study, we assessed the cognitive and physiological effects of transcranial direct current stimulation (tDCS) in 40 adult patients with ADHD in order to identify diagnostic (cross-sectional) and treatment biomarkers (targets). METHODS: Patients performed three experimental sessions in which they received 30 minutes of 2 mA anodal tDCS targeting the left dorsolateral prefrontal cortex, 30 minutes of 2 mA anodal tDCS targeting the right dorsolateral prefrontal cortex, and 30 minutes of sham. Before and after each session, half the patients completed the Eriksen flanker task and the other half completed the stop signal task while we assessed behavior (reaction time, accuracy) and neurophysiology (event-related potentials). RESULTS: Anodal tDCS to the left dorsolateral prefrontal cortex modulated cognitive (reaction time) and physiological (P300 amplitude) measures in the Eriksen flanker task in a state-dependent manner, but no effects were found in the stop signal reaction time of the stop signal task. CONCLUSIONS: These findings show procognitive effects in ADHD associated with the modulation of event-related potential signatures of cognitive control, linking target engagement with cognitive benefit, proving the value of eventrelated potentials as cross-sectional biomarkers of executive performance, and mechanistically supporting the statedependent nature of tDCS. We interpret these results as an improvement in cognitive control but not action cancellation, supporting the existence of different impulsivity constructs with overlapping but distinct anatomical substrates, and highlighting the implications for the development of individualized therapeutics.

Abstract: Cornelia de Lange Syndrome (CdLS) is an autosomal dominant (NIPBL, SMC3 and RAD21) or X-linked (SMC1A and HDAC8) disorder, characterized by distinctive craniofacial appearance, growth retardation, intellectual disability and limb anomalies. Its clinical presentation can be extremely variable. Here we report on a girl with classical CdLS phenotype, growth retardation, intellectual disability and gastroesophageal reflux disease, who is mosaic for the c.6647A>C mutation in the NIPBL gene. While this mutation was undetected by standard Sanger sequencing, it was identified by subsequent exome and panel sequencing approaches. Allele quantification by pyrosequencing showed the presence of the mutation in about 10%, 23.5% and 46.5% of DNA samples from peripheral blood leukocytes, buccal epithelial cells and fibroblasts, respectively. NIPBL mutations in a mosaic state are a frequent cause of CdLS, but clinical diagnosis of these patients can be challenging. In our case, the panel enriched sequencing as well as the exome sequencing clearly demonstrated to be very sensitive tools for mosaic mutation detection.

Abstract: Cornelia de Lange syndrome (CdLS) is a dominant inherited congenital developmental disorder characterized by facial dysmorphism, growth and cognitive impairment, limb malformations and multiple organ involvement. Mutations in five genes, encoding subunits of the cohesin complex (SMC1A, SMC3, RAD21) and its regulators (NIPBL, HDAC8), are responsible for ~ 70% of CdLS cases. Cohesin complex participates in chromosome segregation, DNA repair mechanisms, gene expression and chromosome conformation. Turner syndrome (TS) affects about one in 2000 live born females and results from complete or partial absence of one of the X chromosomes, frequently accompanied by cell-line mosaicism. Here, we describe a patient with CdLS due to a mutation in the NIPBL gene (c.1445_1448delGAGA, p.(Arg482Asnfs*20)) and mosaic TS (mos 45,X/46,XX karyotype). The patient showed multiple clinical features related to CdLS: craniofacial dysmorphism, pre- and post-natal growth delay, minor musculoskeletal anomalies, congenital heart defects and hirsutism. She also presented severe neurological involvement, including slight hypertonia, learning disabilities, verbal and motor development delay, intellectual impairment and (autistic-like features, aggression, self-injurious behaviour). In addition, the proband was clinically diagnosed with TS because of two typical recognizable features: the peripheral lymphedema and the webbed neck. Molecular characterization showed that the NIPBL mutation was present in the two tissues analyzed from different embryonic origins (peripheral blood lymphocytes from mesoderm and oral mucosa epithelial cells from ectoderm). However, FISH analyses revealed that the percentage of cells with monosomy X was low and tissue-specific. These findings indicate that, ontogenically, the NIPBL mutation may have appeared before the mosaic monosomy X. The coexistence in a patient of two rare disorders raises the issue of whether there is indeed a cause-effect association. In addition, we are forced to investigate the correlation between the genotype (a frameshift NIPBL mutation) and karyotype (mosaic X chromosome aneuploidy), with the patient’s phenotype. Therefore, we compare the clinical signs of each organ system described in our case to the typical clinical features of Cornelia de Lange and Turner syndromes. We conclude that our case seems to show a predominant CdLS phenotype, although additional TS manifestations might appear in adolescence. Besides, the significant neurological involvement reinforces the idea that the brain is the organ most sensitive to cohesin disruption.

Abstract: Objective: To investigate opportunities for task shifting to decongest an outpatient neurology clinic in Zambia by describing current patient flow through the clinic and potential nodes for intervention using process mapping. Background: Zambia has a population of approximately 18 million people with 4 full-time adult neurologists, as of 2018, who all practice at the University Teaching Hospital (UTH), the main tertiary care center in the country. As a result of this provider-to-patient ratio, the outpatient neurology clinic is overcrowded and overbooked. Task-shifting programs have shown to improve efficiency, access and quality of care through the use of less specialized healthcare workers in low- and middle-income countries (LMIC). Methods: We evaluated patient flow in the UTH neurology outpatient clinic through the development and analysis of a process map. The characteristics of the clinic population between 2014 and 2018 were retrospectively reviewed from the clinic register. Between July and August 2018, we prospectively collected appointment lag times and time each patient spent waiting at various points in the clinic process. We conducted interviews with clinic staff and neurologists to generate a detailed process map of current pathways to care within the clinic. We then devised task- shifting strategies to help reduce patient wait times based on the overview of clinic process mapping and patient demographics. Results: From 2014 to 2018, there were 4701 outpatients seen in the neurology clinic. The most common neurologi- cal diagnoses were epilepsy (39.2%), headache (21.5%) and cerebrovascular disease (16.7%). During prospective data collection, patients waited an average of 57.8 (SD 73.4) days to be seen by a neurologist. The average wait time from arrival in the clinic to departure was 4.0 (SD 2.5) h. The process map and interviews with clinic staff revealed long waiting times due to a paucity of providers. Nurses and clerks represent an influential stakeholder group, but are not actively involved in any activity to reduce wait times. A large proportion of follow-up patients were stable and seen solely to obtain medication refills. Conclusions: Epilepsy, headache, and stroke make up the largest percentage of outpatient neurological illness in Zambia. Targeting stable patients in these diagnostic categories for a task-shifting intervention may lead to substan- tially decreased patient wait times. Potential interventions include shifting clinical follow-ups and medication refills to less specialized healthcare workers.

Abstract: Objective: To investigate opportunities for task shifting to decongest an outpatient neurology clinic in Zambia by describing current patient flow through the clinic and potential nodes for intervention using process mapping. Background: Zambia has a population of approximately 18 million people with 4 full-time adult neurologists, as of 2018, who all practice at the University Teaching Hospital (UTH), the main tertiary care center in the country. As a result of this provider-to-patient ratio, the outpatient neurology clinic is overcrowded and overbooked. Task-shifting programs have shown to improve efficiency, access and quality of care through the use of less specialized healthcare workers in low- and middle-income countries (LMIC). Methods: We evaluated patient flow in the UTH neurology outpatient clinic through the development and analysis of a process map. The characteristics of the clinic population between 2014 and 2018 were retrospectively reviewed from the clinic register. Between July and August 2018, we prospectively collected appointment lag times and time each patient spent waiting at various points in the clinic process. We conducted interviews with clinic staff and neurologists to generate a detailed process map of current pathways to care within the clinic. We then devised task- shifting strategies to help reduce patient wait times based on the overview of clinic process mapping and patient demographics. Results: From 2014 to 2018, there were 4701 outpatients seen in the neurology clinic. The most common neurologi- cal diagnoses were epilepsy (39.2%), headache (21.5%) and cerebrovascular disease (16.7%). During prospective data collection, patients waited an average of 57.8 (SD 73.4) days to be seen by a neurologist. The average wait time from arrival in the clinic to departure was 4.0 (SD 2.5) h. The process map and interviews with clinic staff revealed long waiting times due to a paucity of providers. Nurses and clerks represent an influential stakeholder group, but are not actively involved in any activity to reduce wait times. A large proportion of follow-up patients were stable and seen solely to obtain medication refills. Conclusions: Epilepsy, headache, and stroke make up the largest percentage of outpatient neurological illness in Zambia. Targeting stable patients in these diagnostic categories for a task-shifting intervention may lead to substan- tially decreased patient wait times. Potential interventions include shifting clinical follow-ups and medication refills to less specialized healthcare workers.

Abstract: Functional Electrical Stimulation (FES) and robotic gait orthoses are rehabilitation technologies that help patients who have lost their normal gait function due to conditions such as spinal cord injury or stroke. FES promotes active muscle contractions that facilitate the rehabilitation process, while the Lokomat (an automated gait orthosis developed by Hocoma in collaboration with Balgrist University Hospital) provides passive limb movements that help to restore and increase mobility. The goal of this project was to combine both technologies in order to improve the beneficial effects that they provide individually. For that purpose, a control algorithm was developed for real-time Linux using an Iterative Learning Control approach in parallel with feedback control. The system was implemented on the ankle and knee joints. The ankle was controlled using a desired angle based on medical literature, while the knee was controlled using a reference force trajectory measured on subjects inside the Lokomat under high-effort conditions. As an alternative to force control on the knee, angle control was implemented for experiments in the lab. The controllers were tested on healthy subjects with and without the inclusion of voluntary movement. Based on these tests, we show the advantages and disadvantages of the control scheme and the physical setup, and give ideas for further improvement. Resumen: La estimulación eléctrica funcional (FES) y las ortesis de marcha robótica son tecnologías de rehabilitación que ayudan a los pacientes que han perdido su función de marcha normal debido a afecciones como una lesión de la médula espinal o un accidente cerebrovascular. FES promueve contracciones musculares activas que facilitan el proceso de rehabilitación, mientras que Lokomat (una órtesis de marcha automatizada desarrollada por Hocoma en colaboración con el Hospital Universitario Balgrist) proporciona movimientos pasivos de las extremidades que ayudan a restaurar y aumentar la movilidad. El objetivo de este proyecto fue combinar ambas tecnologías para mejorar los efectos beneficiosos que proporcionan individualmente. Para ello, se desarrolló un algoritmo de control para Linux en tiempo real utilizando un enfoque de control por aprendizaje iterativo en paralelo con el control de retroalimentación. El sistema se implementó en las articulaciones del tobillo y la rodilla. El tobillo se controló usando un ángulo deseado basado en la literatura médica, mientras que la rodilla se controló usando una trayectoria de fuerza de referencia medida en sujetos dentro del Lokomat bajo condiciones de alto esfuerzo. Como alternativa al control de fuerza en la rodilla, se implementó el control de ángulo para experimentos en el laboratorio. Los controladores se probaron en sujetos sanos con y sin la inclusión de movimiento voluntario. Con base en estas pruebas, mostramos las ventajas y desventajas del esquema de control y la configuración física, y brindamos ideas para mejoras adicionales.