Machine learning toward high-performance electrochemical sensors.

The so-coined fourth paradigm in science has reached the sensing area, with the use of machine learning (ML) toward data-driven improvements in sensitivity, reproducibility, and accuracy, along with the determination of multiple targets from a single measurement using multi-output regression models. Particularly, the use of supervised ML models trained on large data sets produced by electrical and electrochemical bio/sensors has emerged as an impacting trend in the literature by allowing accurate analyses even in the presence of usual issues such as electrode fouling, poor signal-to-noise ratio, chemical interferences, and matrix effects. In this trend article, apart from an outlook for the coming years, we present examples from the literature that demonstrate how helpful ML algorithms can be for dispensing the adoption of experimental methods to address the aforesaid interfering issues, ultimately contributing to translate testing technologies into on-site, practical, and daily applications.

Potential displacement of the native Tenuisvalvae notata by the invasive Cryptolaemus montrouzieri in South America suggested by differences in climate suitability.

Tenuisvalvae notata (Mulsant) (Coccinellidae) is a predatory ladybird beetle native to South America. It specializes in mealybugs prey (Pseudococcidae), but relatively little is known about its ecology. In contrast, the ladybird beetle Cryptolaemus montrouzieri Mulsant (Coccinellidae) is indigenous to Australia and has been introduced to many countries worldwide including Brazil for biological control of mealybugs. The potential impacts of these introductions to native coccinellids have rarely been considered. The software CLIMEX estimated the climate suitability for both species as reflected in the Ecoclimatic Index (EI). Much of South America, Africa, and Australia can be considered climatically suitable for both species, but in most cases, the climate is considerably more favorable for C. montrouzieri than T. notata, especially in South America. The CLIMEX model also suggests seasonal differences in growth conditions (e.g. rainfall and temperature) that could affect the phenology of both species. These models suggest that few locations in South America would be expected to provide T. notata climatic refugia from C. montrouzieri. Although other ecological factors will also be important, such as prey availability, this analysis suggests a strong potential for displacement of a native coccinellid throughout most of its range as a consequence of the invasion by an alien competitor.

Real-Time and In Situ Monitoring of the Synthesis of Silica Nanoparticles.

The real-time and in situ monitoring of the synthesis of nanomaterials (NMs) remains a challenging task, which is of pivotal importance by assisting fundamental studies (e.g., synthesis kinetics and colloidal phenomena) and providing optimized quality control. In fact, the lack of reproducibility in the synthesis of NMs is a bottleneck against the translation of nanotechnologies into the market toward daily practice. Here, we address an impedimetric millifluidic sensor with data processing by machine learning (ML) as a sensing platform to monitor silica nanoparticles (SiO2NPs) over a 24 h synthesis from a single measurement. The SiO2NPs were selected as a model NM because of their extensive applications. Impressively, simple ML-fitted descriptors were capable of overcoming interferences derived from SiO2NP adsorption over the signals of polarizable Au interdigitate electrodes to assure the determination of the size and concentration of nanoparticles over synthesis while meeting the trade-off between accuracy and speed/simplicity of computation. The root-mean-square errors were calculated as ∼2.0 nm (size) and 2.6 × 1010 nanoparticles mL-1 (concentration). Further, the robustness of the ML size descriptor was successfully challenged in data obtained along independent syntheses using different devices, with the global average accuracy being 103.7 ± 1.9%. Our work advances the developments required to transform a closed flow system basically encompassing the reactional flask and an impedimetric sensor into a scalable and user-friendly platform to assess the in situ synthesis of SiO2NPs. Since the sensor presents a universal response principle, the method is expected to enable the monitoring of other NMs. Such a platform may help to pave the way for translating "sense-act" systems into practice use in nanotechnology.

Converging Multidimensional Sensor and Machine Learning Toward High-Throughput and Biorecognition Element-Free Multidetermination of Extracellular Vesicle Biomarkers.

Extracellular vesicles (EVs) are a frontier class of circulating biomarkers for the diagnosis and prognosis of different diseases. These lipid structures afford various biomarkers such as the concentrations of the EVs (CV) themselves and carried proteins (CP). However, simple, high-throughput, and accurate determination of these targets remains a key challenge. Herein, we address the simultaneous monitoring of CV and CP from a single impedance spectrum without using recognizing elements by combining a multidimensional sensor and machine learning models. This multidetermination is essential for diagnostic accuracy because of the heterogeneous composition of EVs and their molecular cargoes both within the tumor itself and among patients. Pencil HB cores acting as electric double-layer capacitors were integrated into a scalable microfluidic device, whereas supervised models provided accurate predictions, even from a small number of training samples. User-friendly measurements were performed with sample-to-answer data processing on a smartphone. This new platform further showed the highest throughput when compared with the techniques described in the literature to quantify EVs biomarkers. Our results shed light on a method with the ability to determine multiple EVs biomarkers in a simple and fast way, providing a promising platform to translate biofluid-based diagnostics into clinical workflows.

Freeze-drying of silica nanoparticles: redispersibility toward nanomedicine applications.

AIM: To study freeze-drying of silica nanoparticles (SiO2NPs) in order to find suitable conditions to produce lyophilized powders with no aggregation after resuspension and storage. METHODS: SiO2NPs were synthesized using a Stöber-based procedure, and characterized by scanning electron microscopy, dynamic light scattering and nitrogen adsorption/desorption isotherms. SiO2NPs hydrodynamic diameters were compared prior and after freeze-drying in the presence/absence of carbohydrate protectants. RESULTS: Glucose was found to be the most suitable protectant against the detrimental effects of lyophilization. The minimum concentration of carbohydrate required to effectively protect SiO2NPs from aggregation during freeze-drying is influenced by the nanoparticle's size and texture. Negligible aggregation was observed during storage. CONCLUSION: Carbohydrates can be used during SiO2NPs freeze-drying process to obtain redispersable solids that maintain original sizes without residual aggregation.