An explainable machine learning approach for hospital emergency department visits forecasting using continuous training and multi-model regression.

BACKGROUND AND OBJECTIVE: In the last years, the Emergency Department (ED) has become an important source of admissions for hospitals. Since late 90s, the number of ED visits has been steadily increasing, and since Covid19 pandemic this trend has been much stronger. Accurate prediction of ED visits, even for moderate forecasting time-horizons, can definitively improve operational efficiency, quality of care, and patient outcomes in hospitals. METHODS: In this paper we propose two different interpretable approaches, based on Machine Learning algorithms, to accurately forecast hospital emergency visits. The proposed approaches involve a first step of data segmentation based on two different criteria, depending on the approach considered: first, a threshold-based strategy is adopted, where data is divided depending on the value of specific predictor variables. In a second approach, a cluster-based ensemble learning is proposed, in such a way that a clustering algorithm is applied to the training dataset, and ML models are then trained for each cluster. RESULTS: The two proposed methodologies have been evaluated in real data from two hospital ED visits datasets in Spain. We have shown that the proposed approaches are able to obtain accurate ED visits forecasting, in short-term and also long-term prediction time-horizons up to one week, improving the efficiency of alternative prediction methods for this problem. CONCLUSIONS: The proposed forecasting approaches have a strong emphasis on providing explainability to the problem. An analysis on which variables govern the problem and are pivotal for obtaining accurate predictions is finally carried out and included in the discussion of the paper.

External fixation in intra-articular fractures of the calcaneus: Is it a valid option as definitive treatment?

INTRODUCTION: The therapeutic management of calcaneal fractures is currently a major source of controversy in the literature. There is no consensus on the need to treat these injuries conservatively or surgically, nor on the criteria for deciding one option or the other. Although the gold standard has classically been the open approach and osteosynthesis, there are currently minimally invasive techniques that also report good results. Our objective is to present our results and experience with the MBA® Orthofix external fixator in a series of cases of calcaneal fractures. METHODS: We performed a retrospective observational study in our center, between the years 2019 and 2021, of Sanders types II-IV calcaneal fractures operated with MBA® Orthofix external fixator. We recorded a total of 38 patients, 42 fractures. We registered demographic information, intraoperative, postoperative, radiological and functional parameters, using the American Orthopedic Foot and Ankle Society (AOFAS), Manchester-Oxford Foot Questionnaire (MOXFQ), EQ-5D and VAS scales. RESULTS: A total of 26 men and 12 women were included, and the median age was 38 years. Mean follow-up was 24,4 months (6, 8-40, 1). The average time to surgery was 7 days and partial loading was started at 2.5 weeks after external fixation, which was removed at 9.2 weeks. The average Böhler angle correction was 7, 4°, Gissane - 12,2°, length 2 mm and calcaneal width was reduced by 5 mm. We recorded two superficial infections, one peroneal entrapment and three subtalar arthrodesis due to post-traumatic osteoarthritis. The AOFAS obtained was 79.1 + / - 15.7 points, MOXFQ 20.1 + / - 16.1 points, EQ-5D 0.84 + / - 0.2 and VAS 3.3 + / - 1.9. CONCLUSION: The external fixator is an excellent surgical alternative for complex articular fractures of the calcaneus, obtaining clinical and radiological results comparable to other osteosynthesis techniques and significantly reducing soft tissue complications.

Management of an iris stromal cyst by fine needle aspiration with absolute alcohol irrigation and argon laser photocoagulation to induce sclerosis.

PURPOSE: To describe the management of an iris stromal cyst by fine needle aspiration and irrigation with absolute alcohol to induce cyst sclerosis. METHODS: We present the case of a 45-year-old patient with no previous history referred for a 3-month history of an iris lesion in his right eye. Examination revealed a 2.1×3mm cyst with transparent walls at the right inferonasal iris root, and a primary acquired iris stromal cyst was diagnosed. Initially, observation was recommended, but at the 2-month follow-up, growth of the lesion was observed, so fine needle aspiration and cryotherapy were performed, with recurrence after 2 months. Subsequently, fine needle aspiration and irrigation with absolute alcohol for 2minutes to induce sclerosis of the cyst was carried out. Two weeks after surgery, argon laser was performed on the iris root and remaining walls of the cyst to induce adherence. RESULTS: After 1 year of the procedure, the patient remains asymptomatic, without associated complications. A depression and hyperpigmentation of the iris in the inferonasal sector persists, without recurrence of internal fluid or increase in cyst volume. Endothelial cell count and intraocular pressure monitoring have remained stable. CONCLUSION: Iris stromal cyst sclerosis by aspiration and irrigation with absolute alcohol was effective and avoided complications associated with resection of the lesion.

Evaluation of palliative therapy, alone or in combination with toceranib phosphate, in dogs diagnosed with metastatic or recurrent beta-cell neoplasia.

AIMS: To compare survival in dogs with recurrent or metastatic insulinomas that were treated with palliative therapy, alone or in combination with toceranib phosphate and to assess tolerability of the combined therapy in dogs. MATERIALS AND METHODS: Dogs diagnosed with insulinoma were retrospectively identified in the records of the Veterinary Teaching Hospital Complutense (Madrid, Spain). Diagnosis of insulinoma was based on clinical signs of hypoglycaemia, concentrations in serum of glucose <3.3 mmol/L and insulin >10 µIU/mL and presence of a pancreatic mass on diagnostic imaging. Dogs were treated surgically or medically, according to clinical stage established by imaging techniques, and monitored with blood and urine analyses monthly and abdominal ultrasonography every 3 months until death. Dogs that presented with metastatic disease at diagnosis or with recurrent hypoglycaemia after surgery were treated, according to the owner's decision, with one of two treatment protocols: palliative therapy alone (control group, n=7: diet, prednisone, famotidine or omeprazole, ±octreotide) or palliative therapy in combination with toceranib (treatment group, n=5; median dose of toceranib 2.52 mg/kg). Overall survival time (OST) and adverse events were compared between the two treatment groups. RESULTS: The OST was longer in the treatment group (median 399, min 125, max 476 days) compared to the control group (median 67, min 23, max 387 days; p=0.042). Dogs in the treatment group had a higher incidence of grade 1-2 gastrointestinal toxicity (diarrhoea) than dogs in the control group (p=0.010). In all cases, gastrointestinal toxicity was solved by temporarily discontinuing toceranib. CONCLUSIONS AND CLINICAL RELEVANCE: The use of toceranib combined with palliative treatment in dogs with suspect metastatic or recurrent insulinomas increased survival time and was adequate tolerated.

[Evaluation of the change in pneumococcal vaccination in 2018 in adults in the Community of Madrid]. / Evaluación del cambio de pauta de vacunación frente al neumococo de 2018 en la población adulta en la Comunidad de Madrid.

BACKGROUND: In the Community of Madrid since 2005 vaccination against pneumococcus in adults from 60 years of age has been carried out with pneumococcal polysaccharide vaccine (PPV23). In January 2018, the guideline changed in favor of pneumococcal conjugate vaccine (PCV13). The objective of this study was to analyze the data of pneumococcal vaccination and to evaluate the implementation of the new guideline of vaccination against pneumococcus in adults in the Community of Madrid. METHODS: Cross-sectional study to estimate vaccine coverage by a retrospective review of the record of vaccination history of people ≥60 years resident in the Community of Madrid. RESULTS: In the Community of Madrid until 2018, 83% of the population were vaccinated with PPV23, 6% with PCV13 and 11% with both vaccines. 96.5% came from Primary Care records. The doses administered of PCV13 surpassed those of PPV23 in ≥60 years in all age groups. 78,660 people ≥60 years were vaccinated with PCV13 (19.5% without risk factors, 67.3% with chronic diseases, 2.5% of the high-risk group and 10.7% belonging to both groups). CONCLUSIONS: In 2018, following Community of Madrid recommendations, 6,639 people were not properly vaccinated with PCV13 and 72,021 (91.6%) were properly vaccinated. The greatest confusion occurred in people ≥60 years without risk factors who had a previous PPV23 and did not require another vaccine but received a PCV13 without complying with the recommended schedule.

Multi-task learning for the prediction of wind power ramp events with deep neural networks.

In Machine Learning, the most common way to address a given problem is to optimize an error measure by training a single model to solve the desired task. However, sometimes it is possible to exploit latent information from other related tasks to improve the performance of the main one, resulting in a learning paradigm known as Multi-Task Learning (MTL). In this context, the high computational capacity of deep neural networks (DNN) can be combined with the improved generalization performance of MTL, by designing independent output layers for every task and including a shared representation for them. In this paper we exploit this theoretical framework on a problem related to Wind Power Ramps Events (WPREs) prediction in wind farms. Wind energy is one of the fastest growing industries in the world, with potential global spreading and deep penetration in developed and developing countries. One of the main issues with the majority of renewable energy resources is their intrinsic intermittency, which makes it difficult to increase the penetration of these technologies into the energetic mix. In this case, we focus on the specific problem of WPREs prediction, which deeply affect the wind speed and power prediction, and they are also related to different turbines damages. Specifically, we exploit the fact that WPREs are spatially-related events, in such a way that predicting the occurrence of WPREs in different wind farms can be taken as related tasks, even when the wind farms are far away from each other. We propose a DNN-MTL architecture, receiving inputs from all the wind farms at the same time to predict WPREs simultaneously in each of the farms locations. The architecture includes some shared layers to learn a common representation for the information from all the wind farms, and it also includes some specification layers, which refine the representation to match the specific characteristics of each location. Finally we modified the Adam optimization algorithm for dealing with imbalanced data, adding costs which are updated dynamically depending on the worst classified class. We compare the proposal against a baseline approach based on building three different independent models (one for each wind farm considered), and against a state-of-the-art reservoir computing approach. The DNN-MTL proposal achieves very good performance in WPREs prediction, obtaining a good balance for all the classes included in the problem (negative ramp, no ramp and positive ramp).

Validación de un nuevo modelo de simulación inorgánica para la anastomosis uretrovesical laparoscópica / Validation of a new artificial model for simulated training of a laparoscopic vesicourethral anastomosis

Objetivo: El objetivo de este estudio es demostrar la validez de un modelo inorgánico de bajo coste para el aprendizaje y entrenamiento de la anastomosis uretrovesical laparoscópica. Materiales y métodos: En este estudio participaron alumnos que asistieron a alguna de las ediciones de los cursos monográficos sobre prostatectomía radical laparoscópica (PRL) celebrados durante el periodo de 2015 a 2017. Estos participantes se dividieron en 2 grupos de acuerdo con su experiencia previa en cirugía laparoscópica (CL). Las tareas que realizaron sobre el simulador inorgánico fueron la resección de la próstata, "tarea 1" y la anastomosis uretrovesical, "tarea 2". Una vez realizados estos ejercicios, los participantes del estudio completaron un cuestionario anónimo donde se recogieron sus datos demográficos y su nivel de experiencia en CL. Además, los asistentes realizaron una valoración de la capacidad didáctica del órgano sintético empleado, evaluando su utilidad como herramienta para la formación específica de PRL. Para demostrar la validación aparente y de contenidos los participantes mostraron su opinión acerca de la textura, la consistencia, la morfología y la similitud del órgano con el paciente real. La valoración se realizó según una escala de Likert de 5 puntos. Resultados: Los alumnos se distribuyeron en 2 grupos: 10 expertos (grupo E) y 12 noveles (grupo N). La única diferencia significativa entre las puntuación de noveles y de expertos fue respecto a la inclusión de esta herramienta en los programas de formación (grupo E = 5 puntos frente al grupo N = 4,4 ± 0,59, p = 0,024). Los expertos calificaron todas las cuestiones con mayores puntaciones que los noveles. En cuanto a la valoración general del modelo inorgánico, los participantes noveles dieron una calificación media de 8,00±0,91 puntos sobre 10, siendo superada por la valoración de los participantes del grupo de expertos, que dieron una puntuación media de 9,4 ± 0,51. Conclusión: Este modelo inorgánico ha demostrado poseer validez aparente, de contenidos y constructiva, además de ser una herramienta didáctica ideal para el aprendizaje y el entrenamiento de la resección prostática y de la anastomosis uretrovesical laparoscópica

Objective: The aim of this study is to prove the effectiveness of a low cost, artificial model for training of a laparoscopic urethrovesical anastomosis. Materials and methods: This study included urologists who attended specialised courses on laparoscopic radical prostatectomy (LRP) held during the period 2015 to 2017. They were divided into 2 groups according to their previous experience in laparoscopic surgery. The tasks performed on the artificial simulator were prostate resection, "task 1", and urethrovesical anastomosis, "task 2". Once these exercises were completed, the study participants filled in an anonymous questionnaire regarding their demographic data and experience level in laparoscopic surgery (LS). In addition, they gave their opinions about the didactic capacity of the artificial organ and evaluated its usefulness as a tool for LRP training. To demonstrate face and content validity, the participants judged the texture, consistency, morphology and evaluated its similarity to the real organ. The assessment was made with a five-point Likert scale. Results: The students were divided into 2 groups: 10 experts (Group E) and 12 novices (Group N). The only significant difference between the scores of novices and experts was regarding the inclusion of this tool in the training programs (Group E = 5 points versus group N = 4.4 ± 0.59, P = .024). The experts' group rated all the items with higher scores than the novices’ one. Regarding the general assessment of the simulation model, the novice participants gave an average score of 8.00 ± 0.91 points out of 10, while the experts’ group granted higher scores of 9.4 ± 0,51. Conclusion: This artificial model has shown to have an elevated face, content and construct validity, as well being an optimal didactic tool for training in the techniques of prostate resection and laparoscopic urethrovesical anastomosis

Validation of a new artificial model for simulated training of a laparoscopic vesicourethral anastomosis. / Validación de un nuevo modelo de simulación inorgánica para la anastomosis uretrovesical laparoscópica.

OBJECTIVE: The aim of this study is to prove the effectiveness of a low cost, artificial model for training of a laparoscopic urethrovesical anastomosis. MATERIALS AND METHODS: This study included urologists who attended specialised courses on laparoscopic radical prostatectomy (LRP) held during the period 2015 to 2017. They were divided into 2 groups according to their previous experience in laparoscopic surgery. The tasks performed on the artificial simulator were prostate resection, "task 1", and urethrovesical anastomosis, "task 2". Once these exercises were completed, the study participants filled in an anonymous questionnaire regarding their demographic data and experience level in laparoscopic surgery (LS). In addition, they gave their opinions about the didactic capacity of the artificial organ and evaluated its usefulness as a tool for LRP training. To demonstrate face and content validity, the participants judged the texture, consistency, morphology and evaluated its similarity to the real organ. The assessment was made with a five-point Likert scale. RESULTS: The students were divided into 2groups: 10 experts (Group E) and 12 novices (Group N). The only significant difference between the scores of novices and experts was regarding the inclusion of this tool in the training programs (Group E=5 points versus group N=4.4±0.59, P=.024). The experts' group rated all the items with higher scores than the novices' one. Regarding the general assessment of the simulation model, the novice participants gave an average score of 8.00±0.91 points out of 10, while the experts' group granted higher scores of 9.4±0,51. CONCLUSION: This artificial model has shown to have an elevated face, content and construct validity, as well being an optimal didactic tool for training in the techniques of prostate resection and laparoscopic urethrovesical anastomosis.

Real-time incidence of travel-related symptoms through a smartphone-based app remote monitoring system: a pilot study.

Trip Doctor®, a Smartphone-based app monitoring system, was developed to detect infections among travelers in real-time. For testing, 106 participants were recruited (62.2% male, mean age 36 years (SD = 11)). Majority of trips were for tourism and main destinations were in South East Asia. Mean travel duration was 14 days (SD = 10). Diarrhea was the most frequently reported symptom (15.5%). The system demonstrated adequate usability and is ready to be used on a larger scale.

Effect of kaolin silver complex on the control of populations of Brettanomyces and acetic acid bacteria in wine.

In this work, the effects of kaolin silver complex (KAgC) have been evaluated to replace the use of SO2 for the control of spoilage microorganisms in the winemaking process. The results showed that KAgC at a dose of 1 g/L provided effective control against the development of B. bruxellensis and acetic acid bacteria. In wines artificially contaminated with an initial population of B. bruxellensis at 104 CFU/mL, a concentration proven to produce off flavors in wine, only residual populations of the contaminating yeast remained after 24 days of contact with the additive. Populations of acetic bacteria inoculated into wine at concentrations of 102 and 104 CFU/mL were reduced to negligible levels after 72 h of treatment with KAgC. The antimicrobial effect of KAgC against B. bruxellensis and acetic bacteria was also demonstrated in a wine naturally contaminated by these microorganisms, decreasing their population in a similar way to a chitosan treatment. Related to this effect, wines with KAgC showed lower concentrations of acetic acid and 4-ethyl phenol than wines without KAgC. The silver concentration from KAgC that remained in the finished wines was below the legal limits. These results demonstrated the effectiveness of KAgC to reduce spoilage microorganisms in winemaking.

What to expect and when: benznidazole toxicity in chronic Chagas' disease treatment.

Background: Benznidazole is one of the two most effective antiparasitic drugs for Chagas' disease treatment. However, knowledge about its toxicity profile is mostly based on post-marketing observational studies. Objectives: Our study combines data from two prospective clinical trials designed to assess the safety of the drug newly produced by ELEA Laboratories (Abarax®). Methods: Eligible participants were selected using a consecutive sampling strategy in the CINEBENZ and BIOMARCHA studies between 2013 and 2016 (EUDRACT 2011-002900-34 and 2012-002645-38, respectively, and clinicaltrials.gov NCT01755403 and NCT01755377, respectively). Enrolled subjects received treatment with 5 mg/kg/day benznidazole orally in two divided doses for 8 weeks and were followed up fortnightly. Results: We observed 305 adverse reactions in 85 of 99 participants (85.9%). Each patient had a median of three adverse reactions, 89.5% were mild and the median duration was 12 days. Most adverse reactions appeared in the first month of treatment except arthritis and peripheral neuropathy. Twenty-six patients did not complete treatment: 2 were withdrawn, 1 for ectopic pregnancy and 1 for epilepsy relapse due to cysticercosis; 2 were lost to follow-up; and 22 were owing to adverse reactions, two of them severe. We observed some unexpected adverse reactions that have not been described previously, such as psychiatric symptoms, erectile dysfunction, menstrual cycle alterations and lung infiltration. Conclusions: There is a very high frequency of adverse reactions to benznidazole. Most adverse reactions are mild, but the treatment burden is significant and unexpected reactions are not rare. Severe reactions are uncommon, but they can be life-threatening. Further studies are necessary to optimize treatment.

[MIII2MII3] n+ trigonal bipyramidal cages based on diamagnetic and paramagnetic metalloligands.

A family of five [MIII2MII3] n+ trigonal bipyramidal cages (MIII = Fe, Cr and Al; MII = Co, Zn and Pd; n = 0 for 1-3 and n = 6 for 4-5) of formulae [Fe2Co3L6Cl6] (1), [Fe2Zn3L6Br6] (2), [Cr2Zn3L6Br6] (3), [Cr2Pd3L6(dppp)3](OTf)6 (4) and [Al2Pd3L6(dppp)3](OTf)6 (5) (where HL is 1-(4-pyridyl)butane-1,3-dione and dppp is 1,3-bis(diphenylphosphino)propane) are reported. Neutral cages 1-3 were synthesised using the tritopic [MIIIL3] metalloligand in combination with the salts CoIICl2 and ZnIIBr2, which both act as tetrahedral linkers. The assembly of the cis-protected [PdII(dppp)(OTf)2] with [MIIIL3] afforded the anionic cages 4-5 of general formula [MIII2PdII3](OTf)6. The metallic skeleton of all cages describes a trigonal bipyramid with the MIII ions occupying the two axial sites and the MII ions sitting in the three equatorial positions. Direct current (DC) magnetic susceptibility, magnetisation and heat capacity measurements on 1 reveal weak antiferromagnetic exchange between the FeIII and CoII ions. EPR spectroscopy demonstrates that the distortion imposed on the {MO6} coordination sphere of [MIIIL3] by complexation in the {MIII2MII3} supramolecules results in a small, but measurable, increase of the zero field splitting at MIII. Complete active space self-consistent field (CASSCF) calculations on the three unique CoII sites of 1 suggest DCo ≈ -14 cm-1 and E/D ≈ 0.1, consistent with the magnetothermal and spectroscopic data.

RUNX1 amplification in AML with myelodysplasia-related changes and ring 21 chromosomes.

Ring 21 is an unstable structural abnormality of chromosome 21 that can lead to RUNX1 gene amplification. We present a unique case with a carrier patient of a constitutional ring chromosome 21 (partial monosomy and trisomy 21) with dysmorphic features and congenital malformations phenotype, who developed acute myeloid leukaemia with myelodysplasia-related changes and two ring 21 chromosomes with RUNX1 amplification. The patient's constitutional ring 21 chromosome showed alterations in tumour suppressor genes, and oncogenes, but not in RUNX1. RUNX1 gene expression at acute myeloid leukaemia diagnosis, showed no upregulation, so other genes may also be the genetic amplification targets in this patient. Copyright © 2016 John Wiley & Sons, Ltd.

ß-Cell replacement as a treatment for type 1 diabetes: an overview of possible cell sources and current axes of research.

To efficiently treat type 1 diabetes, exogenous insulin injections currently represent the main approach to counter chronic hyperglycaemia. Unfortunately, such a therapeutic approach does not allow for perfectly maintained glucose homeostasis and, in time, cardiovascular complications may arise. Therefore, seeking alternative/improved treatments has become a major health concern as an increasing proportion of type 2 diabetes patients also require insulin supplementation. Towards this goal, numerous laboratories have focused their research on ß-cell replacement therapies. Herein, we will review the current state of this research area and describe the cell sources that could potentially be used to replenish the depleted ß-cell mass in diabetic patients.

NUP98-HOXA9 bearing therapy-related myeloid neoplasm involves myeloid-committed cell and induces HOXA5, EVI1, FLT3, and MEIS1 expression.

INTRODUCTION: Chromosomal rearrangements involving NUP98 gene have been associated with human leukemias such as de novo AML, therapy-related AML (t-AML), myelodysplastic syndrome (MDS), and chronic myeloid leukemia (CML). Genetic fusion NUP98-HOXA9, caused by t(7;11)(p15;p15), is a recurrent cytogenetic alteration in de novo acute myeloid leukemia (AML) usually found in young Asian patients and its description in therapy-related myeloid neoplasms (t-MN) is rare. Only one Asian case with molecular demonstration of the NUP98-HOXA9 fusion has been reported in therapy-related leukemia. NUP98-HOXA9 leukemogenic mechanism is derived from the transcription factor activity of the chimeric protein, which enhances the expression of genes related to cellular differentiation arrest and proliferation. PATIENTS AND METHODS: We studied a Caucasian woman with a therapy-related acute myeloid leukemia after Ewing's sarcoma. Molecular demonstration of the genetic fusion NUP98-HOXA9 was performed by RT-PCR, and gene expression was analyzed by real-time PCR, including four AML patients with MLL rearrangements for comparative analysis. Cytologic and flow cytometric analysis was also carried out. RESULTS: After cytologic and flow cytometric analysis diagnostics was therapy-related myeloid neoplasm (t-MN). The major component of blasts in the acute leukemia was with neutrophilic differentiation, but 13% erythroid lineage blasts were also found. Cytogenetic and FISH analysis revealed t(7;11)(p15;p15) and NUP98-HOXA9 fusion gene was demonstrated. Gene expression analysis showed upregulation of EVI1 and MEIS1 in the index patient, both of them previously related to a worst outcome. CONCLUSION: In this work, we include a detailed molecular, clinical, cytological, and cytometric study of the second t-AML bearing NUP98-HOXA9 genetic fusion.

Ética de la virtud, reencontrando el sentido de la clínica / Virtue ethics, rediscovering the meaning of the clinic

No disponible

Incidence of the C104R TACI mutation in patients with primary antibody deficiency

No disponible

The coral reefs optimization algorithm: a novel metaheuristic for efficiently solving optimization problems.

This paper presents a novel bioinspired algorithm to tackle complex optimization problems: the coral reefs optimization (CRO) algorithm. The CRO algorithm artificially simulates a coral reef, where different corals (namely, solutions to the optimization problem considered) grow and reproduce in coral colonies, fighting by choking out other corals for space in the reef. This fight for space, along with the specific characteristics of the corals' reproduction, produces a robust metaheuristic algorithm shown to be powerful for solving hard optimization problems. In this research the CRO algorithm is tested in several continuous and discrete benchmark problems, as well as in practical application scenarios (i.e., optimum mobile network deployment and off-shore wind farm design). The obtained results confirm the excellent performance of the proposed algorithm and open line of research for further application of the algorithm to real-world problems.