HoloSelecta dataset: 10'035 GTIN-labelled product instances in vending machines for object detection of packaged products in retail environments.

To assess the potential of current neural network architectures to reliably identify packaged products within a retail environment, we created an open-source dataset of 295 shelf images of vending machines with 10'035 labelled instances of 109 products. The dataset contains photos of vending machines by the provider Selecta, the largest European operator of vending machines. The vending machines are a mix of machines in public and private office spaces. The vending machines contain food as well as beverage products. The product instances in the vending machine images are labelled with bounding boxes, where a bounding box encapsulates the entire product with as little overlap as possible. The labels corresponding to the bounding box consist of a structured, human-readable labels including brand, product name and size as well as the GTIN of the product. The GTIN is the global standard to identify products in the retail environment and therefore increases the value as a dataset for the retail industry. Contrary to typical object detection datasets that choose labels at a higher level such as a can or bottle for a much wider variety of objects, this dataset chooses a far more detailed label that depends less on the shape but rather on the exact design of the product. The dataset falls into the category of object detection datasets with a large number of objects, which next to the GTIN label, represents a main differentiator of the dataset to other object detection datasets.

Anthropogenic radionuclides in atmospheric air over Switzerland during the last few decades.

The atmospheric nuclear testing in the 1950s and early 1960s and the burn-up of the SNAP-9A satellite led to large injections of radionuclides into the stratosphere. It is generally accepted that current levels of plutonium and caesium radionuclides in the stratosphere are negligible. Here we show that those radionuclides are present in the stratosphere at higher levels than in the troposphere. The lower content in the troposphere reveals that dry and wet deposition efficiently removes radionuclides within a period of a few weeks to months. Since the stratosphere is thermally stratified and separated from the troposphere by the tropopause, radioactive aerosols remain longer. We estimate a mean residence time for plutonium and caesium radionuclides in the stratosphere of 2.5-5 years. Our results also reveal that strong volcanic eruptions like Eyjafjallajökull in 2010 have an important role in redistributing anthropogenic radionuclides from the stratosphere to the troposphere.

Migration of antimony from PET trays into food simulant and food: determination of Arrhenius parameters and comparison of predicted and measured migration data.

Migration experiments with small sheets cut out from ovenable PET trays were performed in two-sided contact with 3% acetic acid as food simulant at various temperatures. The fraction of diffusible antimony (Sb) was estimated to be 62% in the PET sample under study. Apparent diffusion coefficients of Sb in PET trays were determined experimentally. Measurement of migration between 20 and 150°C yielded a linear Arrhenius plot over a wide temperature range from which the activation energy (E(a)) of 188 ± 36 kJ mol(-1) and the pre-exponential factor (D(0)) of 3.6 × 10(14) cm(2) s(-1) were determined for diffusing Sb species. E (a) was similar to previously reported values for PET bottles obtained with a different experimental approach. E (a) and D (0) were applied as model parameters in migration modelling software for predicting the Sb transfer in real food. Ready meals intended for preparation in a baking oven were heated in the PET trays under study and the actual Sb migration into the food phase was measured by isotope dilution ICP-MS. It was shown that the predictive modelling reproduces correctly experimental data.

Occurrence of uranium in Swiss drinking water.

The results of a nationwide survey of uranium in Swiss drinking water are reported. Elevated concentrations of uranium in groundwater are found mainly in the alpine regions and can be traced back to the geology of the bedrock. Water sources were systematically surveyed and analysed for the presence of Li, B, Si, Sc, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, As, Se, Sr, Cd, Sn, Sb, Ba, Tl, Pb and U and the results were analysed to determine if any correlation with uranium concentration was apparent. No correlation was found. The results are interpreted in relation to the current WHO guideline and those of other countries with a view to determining which areas would be affected if a maximum value were to be adopted and which areas require further investigation. Uranium content varied considerably, from below the limit of detection to almost 100 µg L(-1). Of the 5548 data samples, 98% are below the 2004 WHO provisional guideline value of 15 µg L(-1) and 99.7% below the revised (2011) value of 30 µg L(-1).

Estimation of selenium intake in Switzerland in relation to selected food groups.

The selenium concentration in foods was analysed in order to identify principal sources of this trace element in Switzerland. Selenium intake estimations based on three different approaches were carried out. From the relationship between intake and serum/plasma concentration, the selenium intake was estimated to 66 µg day(-1). The second approach based on measured food groups combined with consumption statistics; and the third approach consisted of duplicate meal samples. With the last two methods, over 75% of the serum/plasma based intake was confirmed. Swiss pasta made of North American durum wheat was the food with the highest contribution to the dietary intake, followed by meat. The strong decrease in imports of selenium-rich North American wheat of the last years was not reflected in the present intake estimations. It appears that this intake loss was compensated by a consumption increase of other foods. Compared with former intake estimations, selenium intake seems to be in Switzerland nearly constant for the last 25 years.

Exposure to antimony from polyethylene terephthalate (PET) trays used in ready-to-eat meals.

Antimony residues, a result of the use of a polycondensation catalyst in the production of polyethylene terephthalate (PET) oven-proof trays, were analysed in ready-to-eat meals. The toxicity of antimony has raised concerns about consumer safety; therefore, the migration of small fractions of these residues into ready meals and foods as a result of cooking directly in the PET trays was studied. A straightforward approach of measuring real samples was selected to obtain accurate exposure data. Background antimony concentration was determined separately from a series of lunch meals, which ranged from not detectable to 3.4 microg kg(-1). Microwave and conventional oven-cooking caused a distinct increase in the concentration of antimony in food and ready meals of 0-17 and 8-38 microg kg(-1), respectively, depending, to a certain extent, on the industrial preparations. The migrated quantities of antimony corresponded to 3-13 microg. For comparison, PET roasting bags and ready-made dough products in PET baking dishes were also evaluated. About half of the products prepared at a temperature of 180 degrees C exceeded the specific migration limit set for food contact material by the European Commission. However, the migrated amounts of antimony relative to the accepted tolerable daily intake (TDI) show that exposure from this type of food is currently not of toxicological concern.

Pilot study of urinary iodine concentration and of biochemical thyroid parameters before and after cautious public health intervention on salt iodide content: the Swiss longitudinal 1996-2000 iodine study.

OBJECTIVE AND DESIGN: Iodide concentration (IC) in salt was cautiously increased in Switzerland (15 --> 20 ppm iodide). We evaluated the dynamics of the effect of this intervention on urinary iodine concentration (UIC, microg/l) and on thyroid parameters. SETTING: University Hospital in Bern, Switzerland. SUBJECTS: A cohort of 36 subjects (12 children, 11 women, 13 men) out of 44 were recruited. INTERVENTIONS: During the study periods PRE (May 1996-May 1998) and POST (October 1998-December 2000, subdivided into equal subperiods POST1 and POST2), that is, before and after the increase of IC in salt, subjects collected 6248 urine spots for analysis of UIC. Thyroid volumes (n=2/subject) and serum thyroid parameters (n=8/subject) were sequentially evaluated. METHODS: Average PRE-POST data were compared (multiple regression analysis). RESULTS: UIC increased overall by 5.1% (P=0.0003). Increase of UIC was highest in children (11.3%, P<0.0001), significant in women (8%, P=0.0016), but not significant in men (P=0.143). Comparison between periods POST1 and POST2 showed that UIC changed more gradually in women than in children. Thyroid volumes were normal, no nonphysiological change occurred. TSH indicated euthyroidism; it decreased in children (1.98 ==> 1.74 mU/l, P=0.04) and increased in men (1.65 ==> 1.91mU/l, P=0.025). FT3 decreased in children (P<0.004) and FT4 decreased in men (P=0.017), both within normal ranges. TSH, FT3 and FT4 were unchanged in women. FT3/FT4 ratios were stable. Anti-TPO-Ab titers were stable (P=0.9). Anti-Tg-Ab titers decreased (P=0.009). CONCLUSION: The significant UIC effects were of uncertain metabolic relevance. No pathological side effects occurred. Differential delays and penetrances of UIC increase in children and adults were hitherto unknown. The unspectacular stepwise policy seems to be safe. Our pilot results in a population with moderate iodine deficiency in women should be confirmed in population-based cluster studies. SPONSORSHIP: This work was supported by grants from the University Hospital in Bern, the Swiss Federal Office of Public Health (SFOPH), the 'Swiss National Foundation for Scientific Research' (32-49424.96), the 'Fondation Genevoise de Bienfaisance Valerie Rossi di Montelera', the 'Schweizerische Lebensversicherungs- und Rentenanstalt' and the 'Schüpbach Foundation of the University of Bern'.

Swiss pilot study of individual seasonal fluctuations of urinary iodine concentration over two years: is age-dependency linked to the major source of dietary iodine?

OBJECTIVE: Because of known significant seasonal fluctuations of iodine concentration in cow's milk (ICM) in Switzerland (winter/summer (w/s) ICM ratio averages 5.6), we looked for seasonal variations and familial aspects in urinary iodine and sodium concentrations (UIC, UNaC). DESIGN: : Prospective sequential cohort investigation. SETTING: University hospital. SUBJECTS: Thirteen children (six girls, seven boys) aged 3-10 y and their parents (n=9) aged 30-47 y. INTERVENTIONS: The volunteers collected 1729 spot urine samples (5.1996-5.1998). UIC data from winter (UIC(w): December-February) and summer seasons (UIC(s): July-September) were compared with UNaC and average milk consumption. Iodine intakes from milk and salt were modelized. RESULTS: Highest seasonal fluctuations of UIC were found in six pre-school children (w/s UIC group average=1.56, significant), followed by seven school children (w/s UIC group average=1.24, N.S.); none existed in adults. UIC/UNaC showed corresponding seasonal fluctuations in children, but not in adults. Winter milk was an important iodine source for children, as proven indirectly by similar seasonal fluctuations of ICM, UIC, UIC/UNaC and an important part within UIC due to milk. CONCLUSIONS: Contribution to UIC from milk intake during winter was high in children (40-50%) and lower in adults (about 20%). Compared with children, dietary habits of adults are more complex and their iodine supply depends mainly on iodized salt and not on milk, so the effect of seasonal ICM variations on UIC is less marked. Because of significant seasonal UIC fluctuations in consumers of fresh milk products in Switzerland, results of future UIC studies conducted during summer and winter seasons should be compared cautiously, especially in young children. Furthermore, consumption of milk is to be promoted, since, besides calcium and vitamins, it is an essential source of iodine. UNaC determination should be included in epidemiologic studies if the dietary source of iodine is questioned.

Age- and gender-dependent urinary iodine concentrations in an area-covering population sample from the Bernese region in Switzerland.

OBJECTIVE: Before a scheduled increase of salt iodide content up to 20 p.p.m. in Switzerland, urinary iodine concentration (UIC, microg/l) of a randomly selected, area-covering, age-stratified population subgroup was sampled in spring 1997. In contrast to former published studies on UIC, we minimized biases by age, gender, rural/urban, socio-economic and cultural influences. METHODS: We contacted 750 households in the Bernese region using random telephone numbers. Per household, one proband selected by play dice collected a spot urine sample. We defined seven age categories: infants, children, adolescents, adults I-III and seniors, aged 0-5, 6-12, 13-20, 21-35, 36-50, 51-65 and >65 years respectively. Randomized UIC data were compared with WHO criteria of iodine deficiency (ID) and previous non-randomized, age-dependent Swiss UIC data. To recruit a perfect random group of volunteers proved difficult, as study participation of adolescent and male populations was somewhat lower than that of elder and female (P<0.005) populations respectively. Study participation of inhabitants of rural and suburban regions was comparable (P=0.139). Out of 413 obtained urine spots, 412 (55%) were analyzed statistically (58% women, 42% men, P<0.005). RESULTS: Ninety per cent of all UIC values ranged between 25 and 200 microg/l. Median UIC showed mild ID (94 microg/l). UIC varied significantly between age categories (P=0. 0006). Women had lower UIC than men (P=0.014). Infants had no ID. Women in child-bearing age (13-35 years), adults II, adults III and seniors had mild ID. Compared with recommended supplies of 150 and 200 microg I/day in children and male adolescents respectively, we consider their UIC (110 and 144 microg/l) as at least borderline deficient. Absolute UIC was lowest in seniors. Probands from rural/suburban households had similar UICs. Our UIC data were comparable to published ones. CONCLUSIONS: Despite long-term national efforts with iodized salt (15 p.p.m. in 1997), mild ID still prevailed in Bern in spring 1997, with rare extreme UIC values. Mild ID with women in child-bearing age (and probably also with children and male adolescents) was a serious concern, as goiter growth with puberty or pregnancy might be the consequence. Mild ID in adults and seniors might be due to restricted NaCl intake. Results of our randomly selected subgroup from a 650 000 population show that age- and gender-dependency are important when comparing our collected data with published UIC data. This fact must be considered with non-randomly selected population subgroups.

Urinary iodine concentration follows a circadian rhythm: a study with 3023 spot urine samples in adults and children.

Our overall aim is to monitor iodine supply in a prospective study before and after the September 1998 increase of salt iodide content in Switzerland. Because iodide is supplied by alimentation, we moreover wondered whether urinary iodine concentration (UI) is governed by circadian rhythmicity. Forty-two subjects (18 males and 24 females, including 13 children) collected 3023 urine spots between May 1996 and May 1998, at a rate of three to five samples per month, at any time of the day. The results show that circadian rhythmicity of UI in adults and children was found independent of the individual subject, age, gender, and season. Lowest UI levels were found between 8-11 h. A curve increasing progressively between 12 and 24 h was obtained. UI returned to base-line levels between 21 and 22 h in children only. UI peaks occurred 4-5 h after main meals; children's peaks occurred later than that of adults. Although the existence of a circadian rhythm of UI is probably universal, its profile, however, depends on alimentation. Because nadir of UI is represented by morning spots, this might seem an appropriate collecting period. In view of the significant circadian rhythmicity of UI, studies with restriction of sampling time to morning hours, for example, cannot be directly compared with studies in which urine is sampled all over the day.

Improved measurement of iodine in food samples using inductively coupled plasma isotope dilution mass spectrometry.

A new ICP-MS method for the determination of iodine in food samples is presented. The method makes use of a new miniature cyclonic spray chamber and a concentric glass nebulizer that is designed for low sample uptakes and is operated in a self-aspirating mode. As a consequence the wash-out was accelerated over conventional systems. This configuration allows the direct determination of iodine in mineralized solutions following digestion with nitric acid only. No strong oxidizing reagents such as perchloric acid or lengthy sample preparation were necessary to alter the chemical form of potentially volatile species. The isotope dilution technique using the long-lived isotope 129I was applied to obtain freedom from matrix effects. The present study reports on results for total iodine in selected nutritional and biological reference materials and makes a comparison with instrumental neutron activation analysis.

Direct determination of urinary iodine by inductively coupled plasma mass spectrometry using isotope dilution with iodine-129.

An inductively coupled mass spectrometric method was developed for the direct determination of iodine in urine. The application of isotope dilution analysis with added 129I offers new possibilities for automatic and accurate determinations. The sample preparation consists of dilution with an ammonia solution containing 129I. The validation was made by comparison with the results obtained in another laboratory by a spectrophotometric method based on the Sandell-Kolthoff reaction. Different regression models, including maximum likelihood estimation, were used to compare the methods. None of the models revealed analytical bias between the two methods. The urine samples analyzed for validation were from three persons previously exposed to an iodine bath and covered a concentration range of 0.2 to 2.8 micromol/L. A detection limit of 0.02 micromol/L, a within-run CV of 2.5%, and a between-run CV of 11.9% were estimated for the proposed method.

Determination of selenium in the serum of healthy Swiss adults and correlation to dietary intake.

The serum selenium (Se) concentrations of apparently healthy 20-40 year old blood donors from different parts of Switzerland were determined by electrothermal atomic absorption spectrometry (GFAAS). Application of a rhodium/magnesium matrix-modifier resulted in improved performance in comparison with a palladium modifier. The method was validated by hydride ICP-MS and quality-controlled by independent analysis using GFAAS with palladium as matrix-modifier; no bias was detected. The serum Se concentrations for male (n = 387) and female (n = 243) subjects fell into a normal distribution with mean values and standard deviations of 96.0 +/- 13.3 micrograms/L (1.22 +/- 0.17 mumol/L) and 87.9 +/- 14.4 micrograms/L (1.11 +/- 0.18 mumol/L), respectively. These values corresponded well to the formerly estimated mean daily intakes. Small but significant differences in mean serum Se concentrations were found between genders, ethnic groups as well as geographic regions, whereas age had no influence. The overall Se status of the Swiss population is assessed as adequate, somewhat higher than in the countries adjacent to Switzerland, but lower than in the U.S.A. or Finland. There is no evidence that the Se status of the population has changed over the past 10 years. However, it appears that some Swiss population groups may have a borderline Se status.