Marco Turco

Comparison of standard reading and computer aided detection (CAD) on a national proficiency test of screening mammography

OBJECTIVE To evaluate the role of computer aided detection (CAD) in improving the interpretation of screening mammograms MATERIAL AND METHODS Ten radiologists underwent a proficiency test of screening mammography first by conventional reading and then with the help of CAD. Radiologists were blinded to test results for the whole study duration. Results of conventional and CAD reading were compared in terms of sensitivity and recall rate. Double reading was simulated combining conventional readings of four expert radiologists and compared with CAD reading. RESULTS Considering all ten readings, cancer was identified in 146 or 153 of 170 cases (85.8 vs. 90.0%; chi(2)=0.99, df=1, P=0.31) and recalls were 106 or 152 of 1330 cases (7.9 vs. 11.4%; chi(2)=8.69, df=1, P=0.003) at conventional or CAD reading, respectively. CAD reading was essentially the same (sensitivity 97.0 vs. 96.0%; chi(2)=7.1, df=1, P=0.93; recall rate 10.7 vs. 10.6%; chi(2)=1.5, df=1, P=0.96) as compared with simulated conventional double reading. CONCLUSION CAD reading seems to improve the sensitivity of conventional reading while reducing specificity, both effects being of limited size. CAD reading had almost the same performance of simulated conventional double reading, suggesting a possible use of CAD which needs to be confirmed by further studies inclusive of cost-effective analysis.

Climate change impacts on wildfires in a Mediterranean environment

We analyse the observed climate-driven changes in summer wildfires and their future evolution in a typical Mediterranean environment (NE Spain). By analysing observed climate and fire data from 1970 to 2007, we estimate the response of fire number (NF) and burned area (BA) to climate trends, disentangling the drivers responsible for long-term and interannual changes by means of a parsimonious Multi Linear Regression model (MLR). In the last forty years, the observed NF trend was negative. Here we show that, if improvements in fire management were not taken into account, the warming climate forcing alone would have led to a positive trend in NF. On the other hand, for BA, higher fuel flammability is counterbalanced by the indirect climate effects on fuel structure (i.e. less favourable conditions for fine-fuel availability and fuel connectivity), leading to a slightly negative trend. Driving the fire model with A1B climate change scenarios based on a set of Regional Climate Models from the ENSEMBLES project indicates that increasing temperatures promote a positive trend in NF if no further improvements in fire management are introduced.

Testing MOS precipitation downscaling for ENSEMBLES regional climate models over Spain

[1] Model Output Statistics (MOS) has been recently proposed as an alternative to the standard perfect prognosis statistical downscaling approach for Regional Climate Model (RCM) outputs. In this case, the model output for the variable of interest (e.g. precipitation) is directly downscaled using observations. In this paper we test the performance of a MOS implementation of the popular analog methodology (referred to as MOS analog) applied to downscale daily precipitation outputs over Spain. To this aim, we consider the state‐of‐the‐art ERA40‐driven RCMs provided by the EU‐funded ENSEMBLES project and the Spain02 gridded observations data set, using the common period 1961–2000. The MOS analog method improves the representation of the mean regimes, the annual cycle, the frequency and the extremes of precipitation for all RCMs, regardless of the region and the model reliability (including relatively low‐performing models), while preserving the daily accuracy. The good performance of the method in this complex climatic region suggests its potential transferability to other regions. Furthermore, in order to test the robustness of the method in changing climate conditions, a cross‐validation in driest or wettest years was performed. The method improves the RCM results in both cases, especially in the former.

Does family history influence survival in breast cancer cases

A few studies have suggested a relatively better prognosis for breast cancer (BC) cases reporting a positive family history (FH). We aimed at comparing the survival of patients according to FH in a large hospital‐based series of 1,278 BC cases. Information on FH for BC was obtained at diagnosis by interview. All cases reporting a first‐ or second‐degree FH for breast carcinoma were compared with cases without FH. Overall survival was estimated using a product‐limit method. Hazard ratios (HRs) and the corresponding 95% confidence intervals (95% CIs), adjusted for confounding factors, were computed using proportional hazard models. Overall, 240 (18.8%) cases reporting, at diagnosis, a positive FH (156 with at least 1 first‐degree relative and 84 with at least 1 second‐degree relative) were compared with 1,038 patients without FH for BC. No significant differences were found in terms of distribution of age at diagnosis, tumor stage, nodal involvement, receptor status and histology. Cumulative survival rates at 5 years for cases without FH and with first‐degree and second‐degree FH for BC were 79.8 (95% CI 77.0–83.0), 78.6 (95% CI 70.0–88.0) and 80.2 (95% CI 68.0–92.0), respectively (log‐rank test, χ22 = 0.02, p = 1.0). After adjustment for age, pathologic size and nodal involvement, the HR among cases of invasive cancer with a first‐degree FH of BC was 0.91 (95% CI 0.55–1.48); however, the HR for cases with second‐degree FH was 1.18 (95% CI 0.62–2.25) compared to cases without FH. Our study, based on a large series of consecutive invasive BC cases, did not find any significant survival differences associated with a positive FH for breast carcinoma, suggesting the existence of a large heterogeneity among BC cases with FH.

Decreasing fires in Mediterranean Europe

Forest fires are a serious environmental hazard in southern Europe. Quantitative assessment of recent trends in fire statistics is important for assessing the possible shifts induced by climate and other environmental/socioeconomic changes in this area. Here we analyse recent fire trends in Portugal, Spain, southern France, Italy and Greece, building on a homogenized fire database integrating official fire statistics provided by several national/EU agencies. During the period 1985-2011, the total annual burned area (BA) displayed a general decreasing trend, with the exception of Portugal, where a heterogeneous signal was found. Considering all countries globally, we found that BA decreased by about 3020 km2 over the 27-year-long study period (i.e. about -66% of the mean historical value). These results are consistent with those obtained on longer time scales when data were available, also yielding predominantly negative trends in Spain and France (1974-2011) and a mixed trend in Portugal (1980-2011). Similar overall results were found for the annual number of fires (NF), which globally decreased by about 12600 in the study period (about -59%), except for Spain where, excluding the provinces along the Mediterranean coast, an upward trend was found for the longer period. We argue that the negative trends can be explained, at least in part, by an increased effort in fire management and prevention after the big fires of the 1980’s, while positive trends may be related to recent socioeconomic transformations leading to more hazardous landscape configurations, as well as to the observed warming of recent decades. We stress the importance of fire data homogenization prior to analysis, in order to alleviate spurious effects associated with non-stationarities in the data due to temporal variations in fire detection efforts.

Large biases and inconsistent climate change signals in ENSEMBLES regional projections

In this paper we analyze some caveats found in the state-of-the-art ENSEMBLES regional projections dataset focusing on precipitation over Spain, and highlight the need of a task-oriented validation of the GCM-driven control runs. In particular, we compare the performance of the GCM-driven control runs (20C3M scenario) with the ERA40-driven ones (“perfect” boundary conditions) in a common period (1961–2000). Large deviations between the results indicate a large uncertainty/bias for the particular RCM-GCM combinations and, hence, a small confidence for the corresponding transient simulations due to the potential nonlinear amplification of biases. Specifically, we found large biases for some RCM-GCM combinations attributable to RCM in-house problems with the particular GCM coupling. These biases are shown to distort the corresponding climate change signal, or “delta”, in the last decades of the 21st century, considering the A1B scenario. Moreover, we analyze how to best combine the available RCMs to obtain more reliable projections.

Impact of climate variability on summer fires in a Mediterranean environment (northeastern Iberian Peninsula)

We analyse the impact of climate interannual variability on summer forest fires in Catalonia (northeastern Iberian Peninsula). The study period covers 25 years, from 1983 to 2007. During this period more than 16000 fire events were recorded and the total burned area was more than 240 kha, i.e. around 7.5% of whole Catalonia. We show that the interannual variability of summer fires is significantly correlated with summer precipitation and summer maximum temperature. In addition, fires are significantly related to antecedent climate conditions, showing positive correlation with lagged precipitation and negative correlation with lagged temperatures, both with a time lag of two years, and negative correlation with the minimum temperature in the spring of the same year. The interaction between antecedent climate conditions and fire variability highlights the importance of climate not only in regulating fuel flammability, but also fuel structure. On the basis of these results, we discuss a simple regression model that explains up to 76% of the variance of the Burned Area and up to 91% of the variance of the number of fires. This simple regression model produces reliable out-of-sample predictions of the impact of climate variability on summer forest fires and it could be used to estimate fire response to different climate change scenarios, assuming that climate-vegetation-humans-fire interactions will not change significantly.

Observed climate change hotspots

We quantify climate change hotspots from observations, taking into account the differences in precipitation and temperature statistics (mean, variability, and extremes) between 1981–2010 and 1951–1980. Areas in the Amazon, the Sahel, tropical West Africa, Indonesia, and central eastern Asia emerge as primary observed hotspots. The main contributing factors are the global increase in mean temperatures, the intensification of extreme hot-season occurrence in low-latitude regions and the decrease of precipitation over central Africa. Temperature and precipitation variability have been substantially stable over the past decades, with only a few areas showing significant changes against the background climate variability. The regions identified from the observations are remarkably similar to those defined from projections of global climate models under a “business-as-usual” scenario, indicating that climate change hotspots are robust and persistent over time. These results provide a useful background to develop global policy decisions on adaptation and mitigation priorities over near-time horizons.

Inter-annual and decadal changes in teleconnections drive continental-scale synchronization of tree reproduction

Climate teleconnections drive highly variable and synchronous seed production (masting) over large scales. Disentangling the effect of high-frequency (inter-annual variation) from low-frequency (decadal trends) components of climate oscillations will improve our understanding of masting as an ecosystem process. Using century-long observations on masting (the MASTREE database) and data on the Northern Atlantic Oscillation (NAO), we show that in the last 60 years both high-frequency summer and spring NAO, and low-frequency winter NAO components are highly correlated to continent-wide masting in European beech and Norway spruce. Relationships are weaker (non-stationary) in the early twentieth century. This finding improves our understanding on how climate variation affects large-scale synchronization of tree masting. Moreover, it supports the connection between proximate and ultimate causes of masting: indeed, large-scale features of atmospheric circulation coherently drive cues and resources for masting, as well as its evolutionary drivers, such as pollination efficiency, abundance of seed dispersers, and natural disturbance regimes.Climate oscillations affect weather on different temporal-spatial scales, which poses difficulty in understanding how they influence tree reproduction. Here Ascoli et al. show relationships between low- and high-frequency components of the NAO and masting in two European tree species across multiple decades.

On the key role of droughts in the dynamics of summer fires in Mediterranean Europe

Summer fires frequently rage across Mediterranean Europe, often intensified by high temperatures and droughts. According to the state-of-the-art regional fire risk projections, in forthcoming decades climate effects are expected to become stronger and possibly overcome fire prevention efforts. However, significant uncertainties exist and the direct effect of climate change in regulating fuel moisture (e.g. warmer conditions increasing fuel dryness) could be counterbalanced by the indirect effects on fuel structure (e.g. warmer conditions limiting fuel amount), affecting the transition between climate-driven and fuel-limited fire regimes as temperatures increase. Here we analyse and model the impact of coincident drought and antecedent wet conditions (proxy for the climatic factor influencing total fuel and fine fuel structure) on the summer Burned Area (BA) across all eco-regions in Mediterranean Europe. This approach allows BA to be linked to the key drivers of fire in the region. We show a statistically significant relationship between fire and same-summer droughts in most regions, while antecedent climate conditions play a relatively minor role, except in few specific eco-regions. The presented models for individual eco-regions provide insights on the impacts of climate variability on BA, and appear to be promising for developing a seasonal forecast system supporting fire management strategies.