Inteligencia artificial en enfermedades infecciosas / Artificial intelligence in infectious diseases

Estamos asistiendo a una verdadera revolución tecnológi-ca en el campo de la salud. Los procesos basados en la aplicación de la inteligencia artificial (IA) y el aprendizaje automático (AA) están llegando progresivamente a todas las áreas disciplinares, y su aplicación en el campo de las enfermedades infecciosas es ya vertiginoso, acelerado por la pandemia de COVID-19.Hoy disponemos de herramientas que no solamente pue-den asistir o llevar adelante el proceso de toma de deci-siones basadas en guías o algoritmos, sino que también pueden modificar su desempeño a partir de los procesos previamente realizados. Desde la optimización en la identificación de microorganis-mos resistentes, la selección de candidatos a participar en ensayos clínicos, la búsqueda de nuevos agentes terapéu-ticos antimicrobianos, el desarrollo de nuevas vacunas, la predicción de futuras epidemias y pandemias, y el segui-miento clínico de pacientes con enfermedades infecciosas hasta la asignación de recursos en el curso de manejo de un brote son actividades que hoy ya pueden valerse de la inteligencia artificial para obtener un mejor resultado. El desarrollo de la IA tiene un potencial de aplicación expo-nencial y sin dudas será uno de los determinantes principa-les que moldearán la actividad médica del futuro cercano.Sin embargo, la maduración de esta tecnología, necesaria para su inserción definitiva en las actividades cotidianas del cuidado de la salud, requiere la definición de paráme-tros de referencia, sistemas de validación y lineamientos regulatorios que todavía no existen o son aún solo inci-pientes

We are in the midst of a true technological revolution in healthcare. Processes based upon artificial intelligence and machine learning are progressively touching all disciplinary areas, and its implementation in the field of infectious diseases is astonishing, accelerated by the COVID-19 pandemic. Today we have tools that can not only assist or carry on decision-making processes based upon guidelines or algorithms, but also modify its performance from the previously completed tasks. From optimization of the identification of resistant pathogens, selection of candidates for participating in clinical trials, the search of new antimicrobial therapeutic agents, the development of new vaccines, the prediction of future epidemics and pandemics, the clinical follow up of patients suffering infectious diseases up to the resource allocation in the management of an outbreak, are all current activities that can apply artificial intelligence in order to improve their final outcomes.This development has an exponential possibility of application, and is undoubtedly one of the main determinants that will shape medical activity in the future.Notwithstanding the maturation of this technology that is required for its definitive insertion in day-to-day healthcare activities, should be accompanied by definition of reference parameters, validation systems and regulatory guidelines that do not exist yet or are still in its initial stages

Design and development of an early warning score for covid-19 hospitalized patients.

Pandemics pose a major challenge for public health preparedness, requiring a coordinated international response and the development of solid containment plans. Early and accurate identification of high-risk patients in the course of the current COVID-19 pandemic is vital for planning and making proper use of available resources. The purpose of this study was to identify the key variables that account for worse outcomes to create a predictive model that could be used effectively for triage. Through literature review, 44 variables that could be linked to an unfavorable course of COVID-19 disease were obtained, including clinical, laboratory, and X-ray variables. These were used for a 2-round modified Delphi processing with 14 experts to select a final list of variables with the greatest predictive power for the construction of a scoring system, leading to the creation of a new scoring system: the COVID-19 Severity Index. The analysis of the area under the curve for the COVID-19 Severity Index was 0.94 to predict the need for ICU admission in the following 24 hours against 0.80 for NEWS-2. Additionally, the digital medical record of the Hospital Italiano de Buenos Aires was electronically set for an automatic calculation and constant update of the COVID-19 Severity Index. Specifically designed for the current COVID-19 pandemic, COVID-19 Severity Index could be used as a reliable tool for strategic planning, organization, and administration of resources by easily identifying hospitalized patients with a greater need of intensive care.

La pandemia por COVID-19 planteó un desafío para el sistema salud, debido a la gran demanda de pacientes hospitalizados. La identificación temprana de pacientes hospitalizados con riesgo de evolución desfavorable es vital para asistir en forma oportuna y planificar la demanda de recursos. El propósito de este estudio fue identificar las variables predictivas de mala evolución en pacientes hospitalizados por COVID-19 y crear un modelo predictivo que pueda usarse como herramienta de triage. A través de una revisión narrativa, se obtuvieron 44 variables vinculadas a una evolución desfavorable de la enfermedad COVID-19, incluyendo variables clínicas, de laboratorio y radiográficas. Luego se utilizó un procesamiento por método Delphi modificado de 2 rondas para seleccionar una lista final de variables incluidas en el score llamado COVID-19 Severity Index. Luego se calculó el Área Bajo la Curva (AUC) del score para predecir el pase a terapia intensiva en las próximas 24 horas. El score presentó un AUC de 0,94 frente a 0,80 para NEWS-2. Finalmente se agregó el COVID-19 Severity Index a la historia clínica electrónica de un hospital universitario de alta complejidad. Se programó para que el mismo se actualice de manera automática, facilitando la planificación estratégica, organización y administración de recursos a través de la identificación temprana de pacientes hospitalizados con mayor riesgo de transferencia a la Unidad de Cuidados Intensivos.

Design and development of an early warning score for COVID-19 hospitalized patients / Diseño y desarrollo de un sistema de alerta temprana para pacientes hospitalizados por COVID-19

Abstract Pandemics pose a major challenge for public health preparedness, requiring a coordinated international response and the development of solid containment plans. Early and accurate identifica tion of high-risk patients in the course of the current COVID-19 pandemic is vital for planning and making proper use of available resources. The purpose of this study was to identify the key variables that account for worse outcomes to create a predictive model that could be used effectively for triage. Through literature review, 44 variables that could be linked to an unfavorable course of COVID-19 disease were obtained, including clinical, laboratory, and X-ray variables. These were used for a 2-round modified Delphi processing with 14 experts to select a final list of variables with the greatest predictive power for the construction of a scoring system, leading to the creation of a new scoring system: the COVID-19 Severity Index. The analysis of the area under the curve for the COVID-19 Severity Index was 0.94 to predict the need for ICU admission in the following 24 hours against 0.80 for NEWS-2. Additionally, the digital medical record of the Hospital Italiano de Buenos Aires was electronically set for an automatic calculation and constant update of the COVID-19 Severity Index. Specifically designed for the current COVID-19 pandemic, COVID-19 Severity Index could be used as a reliable tool for strategic planning, organization, and administration of resources by easily identifying hospitalized patients with a greater need of intensive care.

Resumen La pandemia por COVID-19 planteó un desafío para el sistema salud, debido a la gran demanda de pacientes hospitalizados. La identificación temprana de pacientes hospitalizados con riesgo de evo lución desfavorable es vital para asistir en forma oportuna y planificar la demanda de recursos. El propósito de este estudio fue identificar las variables predictivas de mala evolución en pacientes hospitalizados por COVID-19 y crear un modelo predictivo que pueda usarse como herramienta de triage. A través de una revisión narrativa, se obtuvieron 44 variables vinculadas a una evolución desfavorable de la enfermedad COVID-19, incluyendo variables clínicas, de laboratorio y radiográficas. Luego se utilizó un procesamiento por método Delphi modificado de 2 rondas para seleccionar una lista final de variables incluidas en el score llamado COVID-19 Severity Index. Luego se calculó el Área Bajo la Curva (AUC) del score para predecir el pase a terapia intensiva en las próximas 24 horas. El score presentó un AUC de 0,94 frente a 0,80 para NEWS-2. Finalmente se agregó el COVID-19 Severity Index a la historia clínica electrónica de un hospital universitario de alta complejidad. Se programó para que el mismo se actualice de manera automática, facilitando la planificación estratégica, organización y administración de recursos a través de la identificación temprana de pacientes hospitalizados con mayor riesgo de transferencia a la Unidad de Cuidados Intensivos.

Investigación traslacional en microbioma (InTraMic): un área de interés común y creciente en el IMTIB, Hospital Italiano e Instituto Universitario / Translational research in microbiome (InTraMic): an area of common and growing interest at IMTIB, Hospital Italiano and Instituto Universitario

Se estima que aproximadamente 100 trillones de microorganismos (incluidos bacterias, virus y hongos) residen en el intestino humano adulto y que el total del material genético del microbioma es 100 veces superior al del genoma humano. Esta comunidad, conocida como microbioma se adquiere al momento del nacimiento a través de la flora comensal de la piel, vagina y heces de la madre y se mantiene relativamente estable a partir de los dos años desempeñando un papel crítico tanto en el estado de salud como en la enfermedad. El desarrollo de nuevas tecnologías, como los secuenciadores de próxima generación (NGS), permiten actualmente realizar un estudio mucho más preciso de ella que en décadas pasadas cuando se limitaba a su cultivo. Si bien esto ha llevado a un crecimiento exponencial en las publicaciones, los datos sobre las poblaciones Latinoamérica son casi inexistentes. La investigación traslacional en microbioma (InTraMic) es una de las líneas que se desarrollan en el Instituto de Medicina Traslacional e Ingeniería Biomédica (IMTIB). Esta se inició en 2018 con la línea de cáncer colorrectal (CCR) en una colaboración con el Colorectal Cancer Research Group del Leeds Institute of Medical Research en el proyecto Large bowel microbiome disease network: Creation of a proof of principle exemplar in colorectal cancer across three continents. A fines de 2019 se cumplió el objetivo de comprobar la factibilidad de la recolección, envío y análisis de muestras de MBF en 5 continentes, incluyendo muestras provenientes de la Argentina, Chile, India y Vietnam. Luego de haber participado de capacitaciones en Inglaterra, se ha cumplido con el objetivo de la etapa piloto, logrando efectivizar la recolección, envío y análisis metagenómico a partir de la secuenciación de la región V4 del ARNr 16S. En 2019, la línea de enfermedad de hígado graso no alcohólico se sumó a la InTraMic iniciando una caracterización piloto en el marco de una colaboración con el laboratorio Novartis. Los resultados de ese estudio, así como el de cáncer colorrectal, están siendo enviados a publicación. En 2020, con la incorporación de la línea de trasplante alogénico de células progenitoras hematopoyéticas, fue presentado un proyecto para un subsidio del CONICET que ha superado la primera etapa de evaluación. En el presente artículo se brinda una actualización sobre la caracterización taxonómica de microbioma y se describen las líneas de investigación en curso. (AU)

It is estimated that approximately 100 trillion microorganisms (including bacteria, viruses, and fungi) reside in the adult human intestine, and that the total genetic material of the microbiome is 100 times greater than that of the human genome. This community, known as the microbiome, is acquired at birth through the commensal flora of the mother's skin, vagina, and feces and remains relatively stable after two years, playing a critical role in both the state of health and in disease. The development of new technologies, such as next-generation sequencers (NGS), currently allow for a much more precise study of it than in past decades when it was limited to cultivation. Although this has led to exponential growth in publications, data on Latin American populations is almost non-existent. Translational research in microbiome (InTraMic) is one of the lines developed at the Instituto de Medicina Traslacional e Ingeniería Biomédica (IMTIB). This started in 2018 with the Colorectal Cancer Line (CRC) in a collaboration with the Colorectal Cancer Research Group of the Leeds Institute of Medical Research in the project "Large bowel microbiome disease network: Creation of a proof of principle exemplar in colorectal cancer across three continents". At the end of 2019, the objective of verifying the feasibility of collecting, sending and analyzing MBF samples on 5 continents, including samples from Argentina, Chile, India and Vietnam, was met. After having participated in training in England, the objective of the pilot stage has been met, achieving the collection, delivery and metagenomic analysis from the sequencing of the V4 region of the 16S rRNA. In 2019, the non-alcoholic fatty liver disease line joined InTraMic, initiating a pilot characterization in the framework of a collaboration with the Novartis laboratory. The results of that study, as well as that of colorectal cancer, are being published. In 2020, with the incorporation of the allogeneic hematopoietic stem cell transplantation line, a project was presented for a grant from the CONICET that has passed the first stage of evaluation. This article provides an update on the taxonomic characterization of the microbiome and describes the lines of ongoing research. (AU)

Investigación y terapéutica en el trastorno del espectro autista: una experiencia de desarrollo interdisciplinario / Research and therapeutics in the disorder of autistic spectrum: a development experience interdisciplinary

El trastorno del espectro autista (TEA) es un trastorno del neurodesarrollo con amplia problemática psicosocial y neurocognitiva. El uso de estrategias estructuradas en el abordaje de las personas con TEA es una metodología con reconocimiento internacional de utilidad. En la práctica clínica del equipo de diagnóstico y tratamiento del Servicio de Salud Mental Pediátrica del Hospital Italiano de Buenos aires (HIBA) se utilizan las mismas y también otras herramientas como la adquisición del lenguaje a través de la escritura en computadora. Así, es importante compartir la experiencia clínica en el empleo de diferentes estrategias de abordaje en la población con TEA. También, debatir los alcances de la utilización de herramientas tecnológicas en este abordaje y entretanto pensar diferentes formas de validar la utilización de dichas aplicaciones e intervenciones tecnológicas en el TEA. Evaluando sus ventajas y desventajas, buscamos generar un modelo educativo y terapéutico más integrador, centrado en la comunicación y la facilitación de la motivación y la expresividad, para homogeneizar a profesionales y padres con el mundo de los niños con TEA. Esperamos que el uso de dispositivos tecnológicos permita recolectar datos e información clave que luego se podrán utilizar para realizar distintos estudios que arrojen luz sobre aspectos vinculados al desarrollo psicológico-cognitivo de personas con TEA. Es una valiosa opción para considerar por los profesionales entre las alternativas de intervención terapéutica, y que a su vez ayude tanto a plantear nuevas hipótesis en este campo como a ofrecer herramientas ccesibles, innovadoras y eficaces que entrenen y faciliten la tarea de los profesionales de la salud mental. (AU)

The Autism Spectrum Disorder is a neurodevlopmental disorder with a wider psicosocial and neurocognitive problematic. The use of structured strategies in dealing with people with autism spectrum disorder (ASD) is an useful and internationally recognized methodology. In the clinical practice of diagnostic and treatment team of the service of Pediatric Mental Health of the HIBA, these and also other tools such as language acquisition through writing computer are used. So it is important to share clinical experience using different strategies in the population with ASD. Also, discuss the scope of the use of technological tools in this approach and while thinking about different ways to validate the use of such applications and technological interventions in ASD. Assessing their advantages and disadvantages, we seek to create a more inclusive educational and therapeutic model, focused on communication and facilitation of motivation and expression, to homogenize professionals and arents with the ASD children world. We hope that the use of technological devices allow the collection of data and valuable key information which can then be used to conduct studies that shed light on aspects related to psychological and cognitive development in people with ASD. It is a valuable option to consider for professionals in alternative therapeutic intervention, which in turn will help both raise new hypotheses in this field as well as providing accessible, innovative and efficient tools to train and facilitate the work of mental health professionals. (AU)

Minería de Datos y Genómica / Data Mining and Genomics

El amplio acceso a computadoras de alto desempeño y dispositivos electrónicos de gran almacenamiento, entre otros, ha permitido en los últimos años la generación de cantidades masivas de datos, concepto que puede ser representado por Velocidad, Volumen y Variabilidad. La Minería de Datos es un proceso que permite descubrir patrones o asociaciones relevantes, no plenamente descubiertas en principio con los métodos tradicionales de análisis, en grandes bases de datos y generar modelos. Para ello, usa herramientas de áreas tales como Sistemas de Bases de Datos, Almacenamiento, Aprendizaje Automático, Estadística, Visualización de la Información y Computación de Alto Desempeño. En las últimas décadas, la biología molecular ha pasado del análisis de genes individuales a estudios más complejos que abarcan el genoma completo de un individuo. El desarrollo de tecnologías genómicas de alto desempeño, como los microarrays y la secuenciación de próxima generación (NGS), ha hecho posible producir de manera exponencial información, con la expansión de nuestro conocimiento de las bases genéticas de varias enfermedades. En la Medicina Genómica, el uso de la Minería de Datos para el análisis de la información genómica se está convirtiendo en una necesidad cada vez más buscada, contribuyendo así hacia una medicina personalizada tal que permite inferir modelos clínicamente relevantes y definir estrategias terapéuticas individualizadas a partir de datos moleculares de pacientes. (AU)

The availability of use of high-performance computers and large-storage electronic devices, among others, has allowed the generation of a huge masses of digital data, an idea that can be represented by velocity, volume and variety. Data mining is a process that permits to discover relevant patterns or relations, not previously seen with traditional methods of analysis, in large databases and generate models. It uses tools from Database Systems, Data Warehouse, Machine Learning, Statistics, Information Visualization and High-Performance Computing. In the last decades, molecular biology has moved from individual gene analysis to more complex studies that involve the complete genome. The development of high-throughput genomic technologies, such as microarrays and next-generation sequencing, has promoted the exponential growth of a huge amount of information, expanding our knowledge on the genetic basis of various diseases. In genomics medicine, the application of data mining techniques has become an increasingly important process that contributes towards a personalized medicine, that involves the inference of clinically relevant models and defines individualized therapeutic strategies based on the molecular data of patients. (AU)

Inhibition of brain ST8SiaIII sialyltransferase leads to impairment of procedural memory in mice.

Several glycoproteins in mammalian brains contain α2,8-linked disialic acid residues. We previously showed a constant expression of disialic acid (DiSia) in the hippocampus, olfactory bulb and cortex, and a gradual decrease of expression in the cerebellum from neonatal to senile mice. Previous publications indicate that neurite extension of neuroblastoma-derived Neuro2A cells is inhibited in the presence of DiSia antibody. Based on this, we treated Neuro2A cell cultures with RNA interference for ST8SiaIII mRNA, the enzyme responsible for DiSia formation. We observed that neurite extension was inhibited by this treatment. Taking this evidence into consideration and the relationship of the cerebellum with learning and memory, we studied the role of DiSia expression in a learning task. Through delivery of pST8SiaIII into the brains of C57BL/6 neonatal mice, we inhibited the expression of ST8SiaIII. ST8SiaIII mRNA and protein expressions were analyzed by real-time PCR and western blot, respectively. In this work, we showed that pST8SiaIII-treated mice presented a significantly reduced level of ST8SiaIII mRNA in the cerebellum (p<0.01) in comparison to control mice at 8 days after treatment. It is also noted that these levels returned to baseline values in the adulthood. Then, we evaluated behavioural performance in the T-Maze, a learning task that estimates procedural memory. At all ages, pST8SiaIII-treated mice showed a lower performance in the test session, being most evident at older ages (p<0.001). Taken all together, we conclude that gene expression of ST8SiaIII is necessary for some cognitive tasks at early postnatal ages, since reduced levels impaired procedural memory in adult mice.

Differential expression of disialic acids in the cerebellum of senile mice.

It is known that disialic acids (diSia) are present in the mammalian brain. However, the precise anatomical distribution and the chronology of its expression along life are not well studied yet. It is accepted that the transfer of diSia in the brain is mediated mainly by the enzyme ST8Sia III (α2,8-sialyltransferase III). We studied the expression of diSia glycoepitopes and of the ST8Sia III gene in different structures of the mouse brain at different postnatal stages by immunohistochemistry and real-time polymerase chain reaction, respectively. C57BL/6 mice of different stages were used. Samples of hippocampus, olfactory bulb, cortex and cerebellum were processed for studies of molecular biology and immunohistochemistry. Histological analysis revealed an important decrease in diSia labeling in the senile cerebellum compared with other structures and stages (P â‰ª 0.001). In concordance with these results, a significant decrease in ST8Sia III gene expression was found in the cerebellum of senile animals (P < 0.001). These results suggest that diSia are constantly expressed but with differential expression in various areas of the mouse central nervous system. On the other hand, the concordance in the decreased expression of ST8Sia III and the diSia epitope in the cerebellum of senile animals suggest a role of diSia in this structure or, inversely, an influence of aging on the expression of diSia in the cerebellum. Further research in that direction could elucidate the roles of diSia in brain function in health and disease.

Reconsolidation of declarative memory in humans.

The reconsolidation hypothesis states that a consolidated memory could again become unstable and susceptible to facilitation or impairment for a discrete period of time after a reminder presentation. The phenomenon has been demonstrated in very diverse species and types of memory, including the human procedural memory of a motor skill task but not the human declarative one. Here we provide evidence for both consolidation and reconsolidation in a paired-associate learning (i.e., learning an association between a cue syllable and the respective response syllable). Subjects were given two training sessions with a 24-h interval on distinct verbal material, and afterward, they received at testing two successive retrievals corresponding to the first and second learning, respectively. Two main results are noted. First, the first acquired memory was impaired when a reminder was presented 5 min before the second training (reconsolidation), and also when the second training was given 5 min instead of 24 h after the first one (consolidation). Second, the first retrieval proved to influence negatively on the later one (the retrieval-induced forgetting [RIF] effect), and we used the absence of this RIF effect as a very indicator of the target memory impairment. We consider the demonstration of reconsolidation in human declarative memory as backing the universality of this phenomenon and having potential clinical relevance. On the other hand, we discuss the possibility of using the human declarative memory as a model to address several key topics of the reconsolidation hypothesis.

Differential expression of glycans in the hippocampus of rats trained on an inhibitory learning paradigm.

The glycan chains of glycoconjugates play important roles in cell-cell and cell-matrix interactions. In the CNS, previous studies on learning and memory suggest the importance of oligosaccharides attached to glycoconjugates in the modulation of synaptic connections. We studied the hippocampal glycan distribution of rats subject to an inhibitory avoidance task. The expression of glycans was examined by lectin-histochemistry using Vicia villosa lectin (VVL) for terminal alpha/beta N-acetylgalactosamine (alpha/beta GalNAc); Galanthus nivalus lectin (GNL) for terminal mannose alpha-1,3 (Man alpha-1,3); Peanut agglutinin (PNA) for galactose beta-1,3N-acetylgalactosamine (Gal beta-1,3 GalNAc); Erythrina cristagalli lectin (ECL) for galactose beta-1,4 N-acetylglucosamine (Gal beta-1,4 GlcNAc); Sambucus nigra lectin (SNA) for sialic acid alpha-2.6 galactose (SA alpha-2,6 Gal); Maackia amurensis lectin II (MAL II) for sialic acid alpha-2,3 (SA alpha-2,3); Wheat germ agglutinin (WGA) for terminal N-acetylglucosamine with/ without sialic acid (GlcNAc wo SA); succynilated WGA (sWGA) for terminal N-acetylglucosamine without sialic acid (terminal GlcNAc without SA); Griffonia simplicifolia lectin II (GSL II) for terminal alpha/beta N-acetylglucosamine (alpha/beta GlcNAc terminal); and Lotus tetragonolobus lectin (LTL) alpha-fucose. Two groups of 10 animals were examined: non-trained (Control) and Trained rats. ECL, sWGA and GSL II were negative for both groups in all the hippocampal subfields studied. For both groups, VVL was negative in CA4 and granular cells of the Dentate Gyrus (DG) and LTL was negative in the CA4 subfield. Expression of alpha/beta GalNAc, alpha-fucose and GlcNAc in other hippocampal subflields was positive, with no differences between groups. However, expression of Man alpha-1,3 was significantly higher in the CA1, CA2, CA3, and CA4 subfields in the Trained group. On the other hand, expression of Gal beta-1,3 GalNAc was significantly low in CA4 and DG in the Trained group. In conclusion, the results here presented indicate that the exposure of rats to an associative behavioral paradigm related to declarative memory, involves some regulatory mechanism/s for the differential patterns of glycan expression.