Evaluation of the Use of the 12 Bands vs. NDVI from Sentinel-2 Images for Crop Identification

Today, machine learning applied to remote sensing data is used for crop detection. This makes it possible to not only monitor crops but also to detect pests, a lack of irrigation, or other problems. For systems that require high accuracy in crop identification, a large amount of data is required to generate reliable models. The more plots of and data on crop evolution used over time, the more reliable the models. Here, a study has been carried out to analyse neural network models trained with the Sentinel satellite's 12 bands, compared to models that only use the NDVI, in order to choose the most suitable model in terms of the amount of storage, calculation time, accuracy, and precision. This study achieved a training time gain of 59.35% for NDVI models compared with 12-band models; however, models based on 12-band values are 1.96% more accurate than those trained with the NDVI alone when it comes to making predictions. The findings of this study could be of great interest to administrations, businesses, land managers, and researchers who use satellite image data mining techniques and wish to design an efficient system, particularly one with limited storage capacity and response times.

Exploitation of technological variability among wild non-Saccharomyces yeasts to select mixed starters for the production of low alcohol wines

Increase of the sugar content in grape must, and consequently, alcohol levels in wine are some of the principal problems affecting the winemaking industry. High alcohol content can compromise wine quality, creating sensory imbalances, as well as decreasing the perception of some flavors. The technological approaches proposed at this aim, although allowing achievement of the purpose, can determine negative influence on quality of wine. A promising strategy is based on the use of specific microorganisms, such as selected yeast strains, mainly non-Saccharomyces, able to convert grape must sugars towards secondary metabolites rather than ethanol. This study aims at screening of wild non-Saccharomyces strains in order to identify those suitable for the use in mixed starter for the production of wine with reduced alcohol content and, at the same time, with improved aromatic characteristics.

Ultrasonographic measurement of thymus size in IUGR fetuses: a marker of the fetal immunoendocrine response to malnutrition

OBJECTIVE

To test the hypothesis that intrauterine growth restriction (IUGR) is associated with decreased thymus size in the human fetus.

METHODS

The thymus perimeter was measured in 60 consecutive IUGR fetuses at prenatal ultrasound examination. IUGR was defined as an abdominal circumference (AC) <5(th) centile. Sixty controls were identified by selection of the next consecutive appropriately grown fetus of similar gestational age (+/-1 week). To exclude fetal size effects, ratios between thymus perimeter and fetal biometry measurements including biparietal diameter (BPD), AC and femur length (FL), as well as estimated fetal weight (EFW) were compared between IUGR fetuses and controls.

RESULTS

The proportion of fetuses with thymus perimeter <5(th) centile for gestation was significantly higher in IUGR fetuses than in controls (58/60 vs. 7/60, P < 0.0001). The mean thymus perimeter/BPD ratio (0.87 +/- 0.20 vs. 1.13 +/- 0.13, P < 0.0001), thymus perimeter/AC ratio (0.28 +/- 0.06 vs. 0.35 +/- 0.03, P < 0.0001), thymus perimeter/FL ratio (1.18 +/- 0.26 vs. 1.51 +/- 0.19, P < 0.001) and thymus perimeter/EFW ratio (0.05 +/- 0.01 vs. 0.06 +/- 0.01, P = 0.02) were significantly lower in IUGR fetuses than in controls. There was a significant positive correlation between the observed-to-expected mean for gestation thymus perimeter ratio and the enrollment-to-delivery interval (r = 0.44, P < 0.001).

CONCLUSION

IUGR is associated with a disproportionately small thymus. This supports the hypothesis that thymic involution may be part of the fetal neuroendocrine response to intrauterine starvation.

Should adnexal mass size influence surgical approach? A series of 186 laparoscopically managed large adnexal masses

OBJECTIVE

To evaluate the feasibility and safety of laparoscopic management of adnexal masses > or = 10 cm in size.

DESIGN

Prospective cohort study.

SETTING

Two Gynecology Departments of University Hospitals.

POPULATION

All women presenting with an adnexal mass > or = 10 cm in diameter were candidates for laparoscopic management. Women were excluded from laparoscopic approach if there was evidence of ascites or gross metastatic disease. Neither the sonographic features of the cyst nor elevated serum CA125 level was used to exclude women from having a laparoscopic approach.

METHODS

A single operative protocol was followed for all women. All removed specimens were sent for immediate pathological evaluation.

MAIN OUTCOME MEASURES

Rate of conversion to laparotomy, incidence of cancer encountered, and operative complications.

RESULTS

One hundred and eighty-six women underwent laparoscopic evaluation for an adnexal mass of 10 cm or larger in size. The average preoperative mass size was 12.1 +/- 4.9 cm. A benign pathological condition was found in 86.6% (161/186) of the women, primary ovarian cancer in 16 (8.6%) women, a metastatic tumour of gastrointestinal origin in 1 (0.5%) woman, and a low malignant potential ovarian tumour in 8 (4.3%) women. Laparoscopic management was successful for 174 (93.5%) women. Reasons for conversion to laparotomy included anticipated technical difficulty (n = 7) and malignancy (n = 5). No intraoperative complications occurred in the entire study group.

CONCLUSIONS

The vast majority of large adnexal masses can be safely resected laparoscopically, provided that there is expertise in laparoscopic surgery, immediate access to frozen section diagnosis, and preparation of patient to receive an adequate cancer surgery where indicated.

Large cross-sectional area of the umbilical cord as a predictor of fetal macrosomia

OBJECTIVE

To determine whether a large cross-sectional area of the umbilical cord is a predictor of fetal macrosomia.

METHODS

Consecutive patients of > 34 weeks' gestation, who presented for sonographic examination and who delivered within 4 weeks of the examination, were included in the study. The sonographic cross-sectional areas of the umbilical cord, the umbilical vessels and the Wharton's jelly were measured in a free loop of the umbilical cord. Logistic regression analysis was used to determine significant predictors of macrosomia (actual birth weight > 4000 g and > 4500 g). Fetal biometric parameters (biparietal diameter, abdominal circumference and femur length), sonographic estimated fetal weight and umbilical cord area > 95(th) centile for gestational age were used as covariates.

RESULTS

During the study period, 1026 patients were enrolled. Fifty-three (5.2%) newborns had a birth weight > 4000 g, and 22 (2.1%) weighed > 4500 g. The proportion of cases with a large umbilical cord was significantly higher in the group of macrosomic compared with non-macrosomic infants (54.7% vs. 8.7%, P < 0.0001). Multiple regression models demonstrated an independent contribution of the large cord in the prediction of birth weight > 4000 g and > 4500 g (odds ratio (95% CI), 20.6 (9.2-45.9) and 4.2 (1.2-17.7), respectively). The sensitivity, specificity and positive and negative predictive values of a sonographic large umbilical cord were 54.7%, 91.3%, 25.4%, and 97.4%, respectively. The combination of abdominal circumference > 95(th) centile and large cord predicted 100% of macrosomic infants. The proportion of umbilical cords with a Wharton's jelly area > 95(th) centile for gestation was significantly higher in macrosomic fetuses of diabetic compared with non-diabetic mothers.

CONCLUSIONS

Sonographic assessment of umbilical cord area may improve the prediction of fetal macrosomia.