RESUMEN

Cooking is one of the major sources of indoor particulate matter (PM), which poses significant health risks and is a severe health hazard. Current studies lack an economical and effective analytical framework for quantifying inhalable particles (PM10) and fine particulate matter (PM2.5) from residential cooking activities on a large scale under real-world scenarios. This study bridges this gap by employing computer vision (CV) technology and readily available sensors. We collected data over a month in real-world settings, including cooking videos and air quality data (indoor PM10, PM2.5, CO2, temperature, relative humidity, and outdoor PM10 and PM2.5 concentrations). To classify high-emission (pan-frying, stir-frying, deep-frying) and low-emission (stewing, steaming, boiling, non-cooking) activities, we developed and validated a robust CV model named "Cooking-I3D." This model leverages a pre-trained Two-Stream Inflated 3D ConvNet (I3D) architecture. We then assessed the efficacy of the CV-predicted cooking method in PM characterization using a first-order multivariate autoregressive model, controlling for environmental factors. The Cooking-I3D model achieved exceptional performance, boasting an accuracy of 95 % and an Area Under the Curve (AUC) of 0.98. Our results indicate that a single 6-minute high-emission cooking event triggers a 21-25 % increase in indoor PM concentrations and a 23-24 % increase in the indoor/outdoor ratio, with relative errors in these estimates ranging from 10 to 21 %. This innovative method offers a powerful tool for long-term assessment of cooking-related indoor air pollution and facilitates precision exposure assessment in human health studies.

Asunto(s)

RESUMEN

Background: Alzheimer's disease (AD) is a complex disorder, and its risk is influenced by multiple genetic and environmental factors. In this study, an AD risk gene prediction framework based on spatial and temporal features of gene expression data (STGE) was proposed. Methods: We proposed an AD risk gene prediction framework based on spatial and temporal features of gene expression data. The gene expression data of providers of different tissues and ages were used as model features. Human genes were classified as AD risk or non-risk sets based on information extracted from relevant databases. Support vector machine (SVM) models were constructed to capture the expression patterns of genes believed to contribute to the risk of AD. Results: The recursive feature elimination (RFE) method was utilized for feature selection. Data for 64 tissue-age features were obtained before feature selection, and this number was reduced to 19 after RFE was performed. The SVM models were built and evaluated using 19 selected and full features. The area under curve (AUC) values for the SVM model based on 19 selected features (0.740 [0.690-0.790]) and full feature sets (0.730 [0.678-0.769]) were very similar. Fifteen genes predicted to be risk genes for AD with a probability greater than 90% were obtained. Conclusion: The newly proposed framework performed comparably to previous prediction methods based on protein-protein interaction (PPI) network properties. A list of 15 candidate genes for AD risk was also generated to provide data support for further studies on the genetic etiology of AD.