Using Computational Learning for Non-Melanoma Skin Cancer and Actinic Keratosis Near-Infrared Hyperspectral Signature Classification.

BACKGROUND: The early detection of Non-Melanoma Skin Cancer (NMSC) is essential to ensure patients receive the most effective treatment. Diagnostic screening tools for NMSC are crucial due to high confusion rates with other types of skin lesions, such as Actinic Keratosis. Nevertheless, current means of diagnosing and screening patients rely on either visual criteria, that are often conditioned by subjectivity and experience, or highly invasive, slow, and costly methods, such as histological diagnoses. From this, the objectives of the present study are to test if classification accuracies improve in the Near-Infrared region of the electromagnetic spectrum, as opposed to previous research in shorter wavelengths. METHODS: This study utilizes near-infrared hyperspectral imaging, within the range of 900.6 and 1454.8 nm. Images were captured for a total of 125 patients, including 66 patients with Basal Cell Carcinoma, 42 with cutaneous Squamous Cell Carcinoma, and 17 with Actinic Keratosis, to differentiate between healthy and unhealthy skin lesions. A combination of hybrid convolutional neural networks (for feature extraction) and support vector machine algorithms (as a final activation layer) was employed for analysis. In addition, we test whether transfer learning is feasible from networks trained on shorter wavelengths of the electromagnetic spectrum. RESULTS: The implemented method achieved a general accuracy of over 80%, with some tasks reaching over 90%. F1 scores were also found to generally be over the optimal threshold of 0.8. The best results were obtained when detecting Actinic Keratosis, however differentiation between the two types of malignant lesions was often noted to be more difficult. These results demonstrate the potential of near-infrared hyperspectral imaging combined with advanced machine learning techniques in distinguishing NMSC from other skin lesions. Transfer learning was unsuccessful in improving the training of these algorithms. CONCLUSIONS: We have shown that the Near-Infrared region of the electromagnetic spectrum is highly useful for the identification and study of non-melanoma type skin lesions. While the results are promising, further research is required to develop more robust algorithms that can minimize the impact of noise in these datasets before clinical application is feasible.

Deep Convolutional Neural Support Vector Machines for the Classification of Basal Cell Carcinoma Hyperspectral Signatures.

Non-melanoma skin cancer, and basal cell carcinoma in particular, is one of the most common types of cancer. Although this type of malignancy has lower metastatic rates than other types of skin cancer, its locally destructive nature and the advantages of its timely treatment make early detection vital. The combination of multispectral imaging and artificial intelligence has arisen as a powerful tool for the detection and classification of skin cancer in a non-invasive manner. The present study uses hyperspectral images to discern between healthy and basal cell carcinoma hyperspectral signatures. Upon the combined use of convolutional neural networks, with a final support vector machine activation layer, the present study reaches up to 90% accuracy, with an area under the receiver operating characteristic curve being calculated at 0.9 as well. While the results are promising, future research should build upon a dataset with a larger number of patients.

The Spanish diet: an update / La dieta española: una actualización

Background/Aims: The Food Consumption Survey, conducted for over 20 years by the Spanish Ministry of Agriculture, Food and Environment (MAGRAMA), is the most reliable source of data to evaluate the food consumption and dietary patterns of Spain. The aim of the present article was to review the diet trends in Spain and its evolution. Food availability assessment per capita per day, which allows the calculation of energy and nutrient intake and comparison with the Recommended Nutrient Intakes for the Spanish population is described. In addition, different markers of the quality of the diet have been also evaluated. Methods: The sample consisted of consumption and distribution data, obtained from the nationwide representative Food Consumption Survey for the period 2000-2012. A two-stage sampling method was applied, where in the first stage the units to be sampled were towns or local entities, and in the second stage households which were going to be part of the final sample from those entities were selected. Units consisted of towns or local entities in the national territory. The data allowed the calculation of energy and nutrient intakes, using the Food Composition Tables (Moreiras et al, 2013). The quality of the diet was also evaluated: the adequacy of the diet in meeting the recommended intakes for energy and nutrients; energy profile; dietary fat quality; dietary protein quality; nutrient density; Mediterranean diet adequacy indices. The present data were compared with previous data obtained by our research group in 1964, 1981 and 1991. Results: Using the most recent data, average intake comprised: milk and derivatives (356 g/person/day), fruits (323 g/person/day), vegetables and greens (339 g/ person/day), cereals and derivatives (197 g/person/day), meat and meat products (181 g/day), fish (88,6 g/person/ day), oils and fats (41,6 g/person/day), sugar and derivatives (25,6 g/person/day), eggs (27,1 g/person/day), legumes (13,9 g/person/day). There was also a high consumption of non-alcoholic beverages (437 g/person/day) and decreasing for alcoholic beverages (192 g/person/day) compared to previous surveys. In consequence, meat and meat product consumption was higher than the recommendations, whereas for cereals and their derivatives, vegetables and greens, fruit, and legumes and pulses, consumption was below recommendations for the Spanish population (GRUNUMUR, 2004; SENC, 2007). Some staple and traditional Mediterranean foods (bread, potatoes and olive oil) showed a dramatic decline when compared to data from Household Budget Surveys in 1964 data. Energy intake showed a marked decline when compared to the 1960's mean consumption, and show marked differences for food groups contributors. Energy profile shows too much coming from lipids vs carbohydrates and slightly higher from proteins. Conclusion: Food consumption patterns in Spain and energy and nutrient intakes have changed markedly in the last forty years, differing somewhat at present from the traditional and healthy Mediterranean Diet (AU)

Antecedentes/objetivos: La Encuesta de Consumo de Alimentos, realizada durante 20 años por el Ministerio de Agricultura, Alimentación y Medio Ambiente (MAGRAMA), es la fuente más fiable de datos para evaluar el consumo de alimentos y patrones dietéticos en España. El objetivo de este artículo fue revisar las tendencias dietéticas en España y su evolución. Se describe la evaluación de la disponibilidad de alimentos per cápita y día, que permite el cálculo de consumo de energía y nutrientes y su comparación con el Consumo Recomendado de Nutrientes para la población española. Además, se han evaluado diferentes marcadores de la calidad de la dieta. Métodos: La muestra consistió en los datos de consumo y distribución, obtenidos de la Encuesta Nacional de Consumo de Alimentos para el período 2000-2012. Se aplicó un método de muestreo en dos etapas en el que, en la primera etapa, las unidades que se muestreaban fueron ciudades y entidades locales y, en la segunda, se seleccionaron los hogares que conformaron la muestra final a partir de las entidades locales. Las unidades consistieron en ciudades o entidades locales del territorio nacional. Los datos permitieron el cálculo de consumo de energía y nutrientes utilizando las tablas de Consumo de Alimentos (Moreiras et al., 2013). También se evaluó la calidad de la dieta: la adecuación de la dieta para alcanzar los consumos de energía y nutrientes recomendados; perfil de energía; calidad de la grasa de la dieta; calidad de la proteína de la dieta; densidad de nutrientes; índices de adecuación de la dieta mediterránea. Los datos actuales se compararon con los datos previos obtenidos por nuestro grupo de investigación en 1964, 1981 y 1991 Resultados: Utilizando los datos más recientes, el consumo promedio comprendía: leche y derivados (356 g/persona/día), frutas (323 g/persona/día), verduras y hortalizas (339 g/persona/día), cereales y derivados (197 g/persona/día), carne y productos cárnicos (181 g/día), pescado (88,6 g/persona/día), aceites y grasas (41,6 g/persona/día), azúcar y derivados (25,6 g/persona/día), huevos (27,1 g/persona/día), legumbres (13,9 g/persona/día) . También un consumo elevado de bebidas no alcohólicas (437 g/persona/día) y un descenso del consumo de bebidas alcohólicas (192 g/persona/día) en comparación con las encuestas previas. En consecuencia, el consumo de carne y productos cárnicos fue superior al recomendado mientras que el consumo de cereales y sus derivados, verduras y hortalizas, fruta y legumbres estaba por debajo de las recomendaciones para la población española (GRUNU-MUR, 2004; SENC, 2007). Algunos alimentos mediterráneos de consumo habitual y tradicionales (pan, patatas y aceite de oliva) mostraron un declive notable en comparación con los datos de las Encuestas de Economía Doméstica de 1964. El consumo de energía mostró un marcado declive en comparación con el consumo medio de los años sesenta y mostró marcadas diferencias para los distintos grupos de alimentos contribuyentes. El perfil de energía mostró que una gran parte procedía de los lípidos, frente a los hidratos de carbono y algo superior con respecto a las proteínas. Conclusión: Los patrones de consumo de alimentos en España y los consumos de energía y nutrientes han cambiado notablemente en los últimos 40 años, difiriendo algo en la actualidad de la dieta mediterránea tradicional y saludable (AU)

Programa Thao-Salud Infantil en Villanueva de la Cañada (Madrid). Prevalencia de sobrepeso y obesidad en los escolares de 3 a 12 años / No disponible

Fundamentos: Enmarcado en los planteamientos de laEstrategia NAOS, en 2007 se puso en marcha en cincociudades piloto españolas el Programa THAO-Salud Infantil.Su objetivo es frenar la progresión de la obesidad infantila través de la promoción de una alimentación saludable ypráctica de actividad física regular, mediante un plan de accionesque implican transversalmente a toda la comunidady con gran refuerzo mediático.Objetivo: Establecer datos de prevalencia de sobrepesoy obesidad, de los escolares en Villanueva de la Cañada(Madrid), obtenidos durante el curso académico 2007-2008.Métodos: La evaluación de la efectividad del programa serealiza sobre la población escolar de 3-12 años del municipio.Fueron medidos y pesados 2347 escolares, se calcularon susIMC y se clasificaron según su estado ponderal (Orbegozo).Resultados: El 15.1% tiene sobrepeso u obesidad, siendomayor en niñas. Según edad, el 17.4% de los escolares entre3-5 años, el 14.9% de 6-9 y el 12.8% de 10-12 superansus recomendaciones. Existe una dependencia altamentesignificativa entre la prevalencia de sobrepeso y obesidad y eltipo de centro al que pertenecen los escolares, creciendo ensentido: privados, concertados y públicos, mayor en niñas.Conclusiones: Las prevalencias de sobrepeso, obesidad yconjunta son menores en los escolares de 3-12 años delmunicipio que a nivel nacional, y superior en niñas. Laprevalencia de sobrepeso disminuye conforme aumenta laedad. Existe una dependencia estadísticamente significativade las prevalencias de sobrepeso y obesidad y el tipo decolegio al que asisten los escolares(AU)

Background: In the context of the NAOS Strategy, in 2007the program THAO-Childhood health was pilot-tested in5 Spanish towns. The aim of the program is to stop theprogressive trend of childhood obesity through the promotionof healthier eating and regular physical activity bymeans of action plans which involve the whole communityand reinforced by mass media. The objective of this paperis to describe the prevalence of obesity and overweight inschoolchildren of Villanueva de la Cañada (Madrid), basedon data collected during the school ear 2007-2008.Methods: The effectiveness of the programs is evaluatedin school children aged 3-12 yr in the municipality. Bodyweight and height was individually measured in 2347children, BMI computed and classified according by weightdistribution (Orbegozo).Results: 15.1% was classified as being overweight or obese,a higher rate among girls. The prevalence by age groupwasas follows: 17.4% in the 3-5 yr olds; 14.9% in the 6-9 yrold group and 12.8% in the 10-12 yr old group. There isan association between overweight prevalence rates andthe type of school, increasing from private schools towardspublic schools, which show the highest rate, particularlyamong girls.Conclusions: The prevalence of overweight and obesity islower in school boys in the municipality under study comparedto the average national rate aged, but higher amonggirls. Prevalence of overweight decreases in older children.There is an association between the type of school andprevalence of obesity, with higher rates in public schools(AU)