Comparison of random forest and multiple linear regression to model the mass balance of biosolids from a complex biosolids management area.

The use of biosolids as a soil amendment provides an important alternative to disposal and can improve soil health; however, distribution for water resource recovery facilities (WRRFs) in the United States can be challenging due to decreasing cropland, increased precipitation, variable plant operations, and financial constraints. Although statistical modeling is commonly used in the water sector, machine learning is still an emerging tool and can provide insights to optimize operations. Random forest (RF), a machine learning model, and multiple linear regression (MLR) were used in this study to model the mass balance of biosolids from a complex biosolids management area. The RF model outperformed (R2 = 0.89) the MLR model (R2 = 0.49) and showed that rainfall was a major factor impacting distribution. Storage for dried biosolids would help decouple drying operations from wet weather and increase distribution. This study demonstrated how machine learning can assist in decision-making processes for long-term planning at WRRFs. PRACTITIONER POINTS: Random forest predicted the 7-day average mass balance of biosolids from a complex biosolids management area. Decoupling biosolids drying operations from wet weather was identified as the highest operational priority. Machine learning outperformed multiple linear regression and can be an important tool for the water sector.

Advanced machine learning application for odor and corrosion control at a water resource recovery facility.

The objective of this study was to develop a machine learning (ML) application to determine the optimal dosage of sodium hypochlorite (NaOCl) to curtail corrosion and odor by H2 S in the headworks of a water resource recovery facility (WRRF) without overly consuming volatile fatty acids (VFAs) that are essential for the enhanced biological phosphorus removal. Given the highly diverse datasets available, three subproblems were formulated, and three cascaded ML modules were developed accordingly. The final ML models, chosen based on performance, were able to predict various targeted variables. More specifically, in Module 1, a recurrent neural network (RNN) was designed to predict wastewater characteristics. In Module 2, a random forest (RF) classifier and a support vector machine (SVM) classifier were built with the information from Module 1 along with other datasets to predict the concentrations of VFAs and H2 S, respectively. Finally, in Module 3, with the information obtained from Module 2, another RF classifier was developed to predict NaOCl dosage to reduce H2 S but keeping VFAs within the target range. These efforts are relevant and informative for WRRFs that are considering developing Intelligent Water Systems to predict the wastewater characteristics to make operational improvements. PRACTITIONER POINTS: A recurrent neural network (RNN) using long short-term memory (LSTM) successfully predicted influent wastewater parameters. A support vector machine classifier predicted hydrogen sulfide (H2 S) with 97.6% accuracy. The concentration of VFAs, an important parameter in EBPR, was predicted using a random forest classifier with 93.4% accuracy. The optimal NaOCl dosage for H2 S control can be predicted with a random forest classifier using H2 S, VFAs, and flow.

Epidemiological profile of cancer patients from an area with high pesticide use / Perfil epidemiológico de pacientes com câncer de uma área com alto uso de agrotóxico

RESUMO O consumo de agrotóxicos aumentou consideravelmente nos últimos anos. Embora os agrotóxicos tenham ajudado a aumentar a produtividade das culturas, também têm sido associados ao câncer. O objetivo deste estudo foi descrever o perfil epidemiológico de pacientes oncológicos localizados em uma área com alto uso de agrotóxicos. Empregou-se coorte retrospectiva para descrever os casos de câncer. Incluíram-se pacientes diagnosticados com algum tipo de câncer entre 2005 e 2016, residindo, no momento do diagnóstico, em uma das 69 cidades brasileiras estudadas e com idade igual ou superior a 12 anos. Utilizou-se regressão multinível para modelar o coeficiente de morbidade por câncer. Estudaram-se 10.640 pacientes com câncer. Os coeficientes de morbidade por câncer aumentaram com a idade e foram significativamente maiores entre as pessoas que residiam em áreas rurais, quando comparadas com as residentes das áreas urbanas (p<0,0001). Em ambas as áreas, os homens apresentaram coeficientes de morbidade por câncer significativamente maiores do que as mulheres. Este estudo sugere que a maior incidência de câncer está relacionada a fatores da vida rural, como a exposição à agrotóxicos, visto que a área estudada é conhecida por sua economia baseada na agricultura e no elevado uso de agrotóxicos.

ABSTRACT Pesticide consumption has increased considerably in the last years. Although pesticides have helped to increase the crops' productivity, they have been associated with the incidence of cancers. The objective of this study was to describe the epidemiological profile of cancer patients living in an area with high use of pesticide. A retrospective cohort design was used to describe cancer cases. Patients were included whenever (i) diagnosed with any type of cancer between 2005 and 2016; (ii) living, at the time of diagnosis, in one of the 69 Brazilian cities studied; and (iii) aged 12 years old or over. A multilevel regression model was used to model the cancer morbidity coefficient. A total of 10,640 cancer patients were studied. Cancer morbidity coefficients increased with age and were significantly higher among people residing in rural areas than among those living in urban areas (p-value<0.0001). In both urban and rural areas, males showed significantly higher cancer morbidity coefficients than females. This study suggests that higher cancer incidence is related to factors of rural life, which can include pesticide exposure since the studied area is known for its agriculture based economy and high pesticide use.

Pesticide exposure and cancer: an integrative literature review / Exposição a agrotóxicos e câncer: uma revisão integrativa da literatura

ABSTRACT We conducted an integrative literature review of published studies on pesticide and cancer exposure, focusing on farmers, rural population, pesticide applicators, and rural workers. The Medline/PubMed was used as searching database. After the retrieval, 74 articles were selected according to pre-established criteria, which design involved 39 case-controls, 32 cohorts, 2 ecological ones, and 1 cross-sectional. Among them, 64 studies showed associations between pesticides and cancer while 10 did not find any significant association. The studies found 53 different types of pesticides significantly associated with at least one type of cancer and 19 different types of cancers linked to at least one type of pesticide. Although few studies presented contradictory results, the sole fact of being a farmer or living near crops or high agricultural areas have also been used as a proxy for pesticide exposure and significantly associated with higher cancer risk. The literature well illustrates the case of prostate cancer, Non-Hodgkin lymphoma, leukemia, multiple myeloma, bladder and colon cancers. Studies are recommended to further investigate the relationship between pesticide and neoplasm of testis, breast, esophagus, kidney, thyroid, lip, head and neck, and bone.

RESUMO Trata-se de revisão integrativa da literatura sobre estudos publicados em relação à exposição a agrotóxicos e câncer, com foco em agricultores, população rural, aplicadores de agrotóxicos e trabalhadores rurais. A busca dos artigos foi realizada por meio do banco de dados Medline/PubMed. Após a triagem, 74 artigos foram selecionados de acordo com critérios pré-estabelecidos, sendo 39 caso-controle, 32 coortes, dois ecológicos e um transversal. Desses, 64 estudos mostraram associação entre agrotóxicos e câncer, enquanto dez não encontraram associação significativa. Nesses 64, 53 diferentes tipos de agrotóxicos foram significativamente associados com pelo menos um tipo de câncer e, inversamente, 19 diferentes tipos de câncer foram associados a pelo menos um tipo de agrotóxico. Embora alguns estudos tenham apresentado resultados contraditórios, ser um agricultor ou morar perto de plantações ou de áreas densamente agrícolas também tem sido motivo para representar a exposição a agrotóxicos e considerado significativamente associado a um maior risco de câncer. A literatura ilustra bem o câncer de próstata, linfoma não-Hodgkin, leucemia, mieloma múltiplo, bexiga e câncer de cólon. Recomendam-se estudos que investiguem mais a relação entre agrotóxicos e neoplasmas de testículos, mama, esôfago, rim, tireoide, lábio, cabeça e pescoço e osso.