Image learning to accurately identify complex mixture components.

The study of complex mixtures is very important for exploring the evolution of natural phenomena, but the complexity of the mixtures greatly increases the difficulty of material information extraction. Image perception-based machine-learning techniques have the ability to cope with this problem in a data-driven way. Herein, we report a 2D-spectral imaging method to collect matter information from mixture components, and the obtained feature images can be easily provided to deep convolutional neural networks (CNNs) for establishing a spectral network. The results demonstrated that a single CNN trained end-to-end from the proposed images can directly accomplish synchronous measurement of multi-component samples using only raw pixels as inputs. Our strategy has some innate advantages, such as fast data acquisition, low cost, and simple chemical treatment, suggesting that it can be extensively applied in many fields, including environmental science, biology, medicine, and chemistry.

Machine learning for total organic carbon analysis of environmental water samples using high-throughput colorimetric sensors.

Due to the complexity of nonlinear reactions, the analysis of environmental samples often relies on expensive equipment as well as tedious and time-consuming experimental procedures. Currently, the efficient machine learning (ML) strategy based on big data offers some new insights for the analysis of complex components in the environmental field. In this study, ML was applied for the analysis of total organic carbon (TOC). We prepared a special colorimetric sensor (c-sensor) by inkjet printing. The sensor reacted with water samples in a high-throughput process, producing characteristic patterns to map TOC information in water samples. To quickly acquire TOC information on c-sensors, a ML model was proposed to describe the relationship between the c-sensor and TOC value. According to this study, the c-sensor and ML can be effectively applied to TOC information analysis of environmental water samples, which provides convenience for environmental research. It is foreseeable that ML has a broad prospect of application in environmental research.

Machine learning for mixture toxicity analysis based on high-throughput printing technology.

Analysis on mixture toxicity (Mix-tox) of the multi-chemical space is constantly followed with interest for many researchers. Conventional toxicity tests with time-consuming and costly operations make researchers can only establish some toxicity prediction models aiming to a limited sampling dimension. The rapid development of machine learning (ML) algorithm will accelerate the exploration of many fields involving toxicity analysis. Rather than the model calculation capacity, the challenge of this process mainly comes from the lack of toxicology big-data to perform toxicity perception through the ML model. In this paper, a full strategy based a standardized high-throughput experiment was developed for Mix-tox analysis throughout the whole routine, from big-sample dataset design, model building, and training, to the toxicity prediction. Using the concentration variates as input and bio-luminescent inhibition rate as output, it turned out that a well-trained random forest algorithm was successfully applied to assess the mixtures' toxicity effect, suggesting its value in facilitating adoption of Mix-tox analysis.