A multi-step machine learning approach to assess the impact of COVID-19 lockdown on NO2 attributable deaths in Milan and Rome, Italy

Background Air pollution is one of the main concerns for the health of European citizens, and cities are currently striving to accomplish EU air pollution regulation. The 2020 COVID-19 lockdown measures can be seen as an unintended but effective experiment to assess the impact of traffic restriction policies on air pollution. Our objective was to estimate the impact of the lockdown measures on NO2 concentrations and health in the two largest Italian cities. Methods NO2 concentration datasets were built using data deriving from a 1-month citizen science monitoring campaign that took place in Milan and Rome just before the Italian lockdown period. Annual mean NO2 concentrations were estimated for a lockdown scenario (Scenario 1) and a scenario without lockdown (Scenario 2), by applying city-specific annual adjustment factors to the 1-month data. The latter were estimated deriving data from Air Quality Network stations and by applying a machine learning approach. NO2 spatial distribution was estimated at a neighbourhood scale by applying Land Use Random Forest models for the two scenarios. Finally, the impact of lockdown on health was estimated by subtracting attributable deaths for Scenario 1 and those for Scenario 2, both estimated by applying literature-based dose–response function on the counterfactual concentrations of 10 μg/m3. Results The Land Use Random Forest models were able to capture 41–42% of the total NO2 variability. Passing from Scenario 2 (annual NO2 without lockdown) to Scenario 1 (annual NO2 with lockdown), the population-weighted exposure to NO2 for Milan and Rome decreased by 15.1% and 15.3% on an annual basis. Considering the 10 μg/m3 counterfactual, prevented deaths were respectively 213 and 604. Conclusions Our results show that the lockdown had a beneficial impact on air quality and human health. However, compliance with the current EU legal limit is not enough to avoid a high number of NO2 attributable deaths. This contribution reaffirms the potentiality of the citizen science approach and calls for more ambitious traffic calming policies and a re-evaluation of the legal annual limit value for NO2 for the protection of human health.

Diagnostic and predictive accuracy of anti-mullerian hormone for ovarian function after chemotherapy in premenopausal women with early breast cancer

Purpose Accurate diagnosis and prediction of loss of ovarian function after chemotherapy for premenopausal women with early breast cancer (eBC) is important for future fertility and clinical decisions regarding the need for subsequent adjuvant ovarian suppression. We have investigated the value of anti-mullerian hormone (AMH) as serum biomarker for this. Methods AMH was measured in serial blood samples from 206 premenopausal women aged 40–45 years with eBC, before and at intervals after chemotherapy. The diagnostic accuracy of AMH for loss of ovarian function at 30 months after chemotherapy and the predictive value for that of AMH measurement at 6 months were analysed. Results Undetectable AMH showed a high diagnostic accuracy for absent ovarian function at 30 months with AUROC 0.89 (96% CI 0.84–0.94, P < 0.0001). PPV of undetectable AMH at 6 months for a menopausal estradiol level at 30 months was 0.77. In multivariate analysis age, pre-treatment AMH and FSH, and taxane treatment were significant predictors, and combined with AMH at 6 months, gave AUROC of 0.90 (95% CI 0.86–0.94), with PPV 0.79 for loss of ovarian function at 30 months. Validation by random forest models with 30% data retained gave similar results. Conclusions AMH is a reliable diagnostic test for lack of ovarian function after chemotherapy in women aged 40–45 with eBC. Early analysis of AMH after chemotherapy allows identification of women who will not recover ovarian function with good accuracy. These analyses will help inform treatment decisions regarding adjuvant endocrine therapy in women who were premenopausal before starting chemotherapy.

Multimodal User Feedback During Adaptive Robot-Human Presentations

Feedback is an essential part of all communication, and agents communicating with humans must be able to both give and receive feedback in order to ensure mutual understanding. In this paper, we analyse multimodal feedback given by humans towards a robot that is presenting a piece of art in a shared environment, similar to a museum setting. The data analysed contains both video and audio recordings of 28 participants, and the data has been richly annotated both in terms of multimodal cues (speech, gaze, head gestures, facial expressions, and body pose), as well as the polarity of any feedback (negative, positive, or neutral). We train statistical and machine learning models on the dataset, and find that random forest models and multinomial regression models perform well on predicting the polarity of the participants' reactions. An analysis of the different modalities shows that most information is found in the participants' speech and head gestures, while much less information is found in their facial expressions, body pose and gaze. An analysis of the timing of the feedback shows that most feedback is given when the robot makes pauses (and thereby invites feedback), but that the more exact timing of the feedback does not affect its meaning.

Random Forest for Genomic Prediction

We give a detailed description of random forest and exemplify its use with data from plant breeding and genomic selection. The motivations for using random forest in genomic-enabled prediction are explained. Then we describe the process of building decision trees, which are a key component for building random forest models. We give (1) the random forest algorithm, (2) the main hyperparameters that need to be tuned, and (3) different splitting rules that are key for implementing random forest models for continuous, binary, categorical, and count response variables. In addition, many examples are provided for training random forest models with different types of response variables with plant breeding data. The random forest algorithm for multivariate outcomes is provided and its most popular splitting rules are also explained. In this case, some examples are provided for illustrating its implementation even with mixed outcomes (continuous, binary, and categorical). Final comments about the pros and cons of random forest are provided.

Prediction of Gas Turbine Trip: A Novel Methodology Based On Random Forest Models

A gas turbine trip is an unplanned shutdown, of which the most relevant consequences are business interruption and a reduction of equipment remaining useful life. Thus, understanding the underlying causes of gas turbine trip would allow predicting its occurrence in order to maximize gas turbine profitability and improve its availability. In the ever competitive Oil & Gas sector, data mining and machine learning are increasingly being employed to support a deeper insight and improved operation of gas turbines. Among the various machine learning tools, Random Forests are an ensemble learning method consisting of an aggregation of decision tree classifiers. This paper presents a novel methodology aimed at exploiting information embedded in the data and develops Random Forest models, aimed at predicting gas turbine trip based on information gathered during a timeframe of historical data acquired from multiple sensors. The novel approach exploits time series segmentation to increase the amount of training data, thus reducing overfitting. First, data are transformed according to a feature engineering methodology developed in a separate work by the same authors. Then, Random Forest models are trained and tested on unseen observations to demonstrate the benefits of the novel approach. The superiority of the novel approach is proved by considering two real-word case-studies, involving field data taken during three years of operation of two fleets of gas turbines located in different regions. The novel methodology allows values of Precision, Recall and Accuracy in the range 75 - 85 %, thus demonstrating the industrial feasibility of the predictive methodology.

Evaluation of the Performance of Time-Series Sentinel-1 Data for Discriminating Rock Units

The potential use of time-series Sentinel-1 synthetic aperture radar (SAR) data for rock unit discrimination has never been explored in previous studies. Here, we employed time-series Sentinel-1 data to discriminate Dananhu formation, Xinjiang group, Granite, Wusu group, Xishanyao formation, and Diorite in Xinjiang, China. Firstly, the temporal variation of the backscatter metrics (backscatter coefficient and coherence) from April to October derived from Sentinel-1, was analyzed. Then, the significant differences of the time-series SAR metrics among different rock units were checked using the Kruskal–Wallis rank sum test and Tukey’s honest significant difference test. Finally, random forest models were used to discriminate rock units. As for the input features, there were four groups: (1) time-series backscatter metrics, (2) single-date backscatter metrics, (3) time-series backscatter metrics at VV, and (4) VH channel. In each feature group, there were three sub-groups: backscatter coefficient, coherence, and combined use of backscatter coefficient and coherence. Our results showed that time-series Sentinel-1 data could improve the discrimination accuracy by roughly 9% (from 55.4% to 64.4%), compared to single-date Sentinel-1 data. Both VV and VH polarization provided comparable results. Coherence complements the backscatter coefficient when discriminating rock units. Among the six rock units, the Granite and Xinjiang group can be better differentiated than the other four rock units. Though the result still leaves space for improvement, this study further demonstrates the great potential of time-series Sentinel-1 data for rock unit discrimination.