A Deep Anomaly Detection System for IoT-Based Smart Buildings.

In recent years, technological advancements in sensor, communication, and data storage technologies have led to the increasingly widespread use of smart devices in different types of buildings, such as residential homes, offices, and industrial installations. The main benefit of using these devices is the possibility of enhancing different crucial aspects of life within these buildings, including energy efficiency, safety, health, and occupant comfort. In particular, the fast progress in the field of the Internet of Things has yielded exponential growth in the number of connected smart devices and, consequently, increased the volume of data generated and exchanged. However, traditional Cloud-Computing platforms have exhibited limitations in their capacity to handle and process the continuous data exchange, leading to the rise of new computing paradigms, such as Edge Computing and Fog Computing. In this new complex scenario, advanced Artificial Intelligence and Machine Learning can play a key role in analyzing the generated data and predicting unexpected or anomalous events, allowing for quickly setting up effective responses against these unexpected events. To the best of our knowledge, current literature lacks Deep-Learning-based approaches specifically devised for guaranteeing safety in IoT-Based Smart Buildings. For this reason, we adopt an unsupervised neural architecture for detecting anomalies, such as faults, fires, theft attempts, and more, in such contexts. In more detail, in our proposal, data from a sensor network are processed by a Sparse U-Net neural model. The proposed approach is lightweight, making it suitable for deployment on the edge nodes of the network, and it does not require a pre-labeled training dataset. Experimental results conducted on a real-world case study demonstrate the effectiveness of the developed solution.

A validated LC-MS/MS method for quantitative determination of L-dopa in Fagioli di Sarconi beans (Phaseolus vulgaris L.).

An analytical method based on ultrasound assisted extraction (UAE) and liquid chromatography coupled to electrospray tandem mass spectrometry (LC-ESI/MS/MS) was validated and applied for determining L-dopa in four ecotypes of Fagioli di Sarconi beans (Phaseolus vulgaris L.), marked with the European label PGI (Protected Geographical Indication). The selectivity of the proposed method was ensured by the specific fragmentation of the analyte. Simple isocratic chromatographic conditions and mass spectrometric detection in multiple reaction monitoring (MRM) acquisition mode were used for sensitive quantification. The LC-ESI/MS/MS method was validated within a linear range of 0.001-5.000 µg/mL. Values of 0.4 and 1.1 ng/mL were obtained for the limits of detection and quantification, respectively. The repeatability, inter-day precision, and recovery values ranges were 0.6%-4.5%, 5.4%-9.9%, and 83%-93%, respectively. Fresh and dried beans, as well as pods, cultivated exclusively with organic methods avoiding any synthetic fertilizers and pesticides were analyzed showing an L-dopa content ranging from 0.020 ± 0.005 to 2.34 ± 0.05 µg/g dry weight.

An Amperometric Biosensor Based on a Bilayer of Electrodeposited Graphene Oxide and Co-Crosslinked Tyrosinase for L-Dopa Detection in Untreated Human Plasma.

L-Dopa, a bioactive compound naturally occurring in some Leguminosae plants, is the most effective symptomatic drug treatment for Parkinson's disease. During disease progression, fluctuations in L-DOPA plasma levels occur, causing motor complications. Sensing devices capable of rapidly monitoring drug levels would allow adjusting L-Dopa dosing, improving therapeutic outcomes. A novel amperometric biosensor for L-Dopa detection is described, based on tyrosinase co-crosslinked onto a graphene oxide layer produced through electrodeposition. Careful optimization of the enzyme immobilization procedure permitted to improve the long-term stability while substantially shortening and simplifying the biosensor fabrication. The effectiveness of the immobilization protocol combined with the enhanced performances of electrodeposited graphene oxide allowed to achieve high sensitivity, wide linear range, and a detection limit of 0.84 µM, suitable for L-Dopa detection within its therapeutic window. Interference from endogenous compounds, tested at concentrations levels typically found in drug-treated patients, was not significant. Ascorbic acid exhibited a tyrosinase inhibitory behavior and was therefore rejected from the enzymatic layer by casting an outer Nafion membrane. The proposed device was applied for L-Dopa detection in human plasma, showing good recoveries.

At the Confluence of Artificial Intelligence and Edge Computing in IoT-Based Applications: A Review and New Perspectives.

Given its advantages in low latency, fast response, context-aware services, mobility, and privacy preservation, edge computing has emerged as the key support for intelligent applications and 5G/6G Internet of things (IoT) networks. This technology extends the cloud by providing intermediate services at the edge of the network and improving the quality of service for latency-sensitive applications. Many AI-based solutions with machine learning, deep learning, and swarm intelligence have exhibited the high potential to perform intelligent cognitive sensing, intelligent network management, big data analytics, and security enhancement for edge-based smart applications. Despite its many benefits, there are still concerns about the required capabilities of intelligent edge computing to deal with the computational complexity of machine learning techniques for big IoT data analytics. Resource constraints of edge computing, distributed computing, efficient orchestration, and synchronization of resources are all factors that require attention for quality of service improvement and cost-effective development of edge-based smart applications. In this context, this paper aims to explore the confluence of AI and edge in many application domains in order to leverage the potential of the existing research around these factors and identify new perspectives. The confluence of edge computing and AI improves the quality of user experience in emergency situations, such as in the Internet of vehicles, where critical inaccuracies or delays can lead to damage and accidents. These are the same factors that most studies have used to evaluate the success of an edge-based application. In this review, we first provide an in-depth analysis of the state of the art of AI in edge-based applications with a focus on eight application areas: smart agriculture, smart environment, smart grid, smart healthcare, smart industry, smart education, smart transportation, and security and privacy. Then, we present a qualitative comparison that emphasizes the main objective of the confluence, the roles and the use of artificial intelligence at the network edge, and the key enabling technologies for edge analytics. Then, open challenges, future research directions, and perspectives are identified and discussed. Finally, some conclusions are drawn.

Development and Validation of a Reversed-Phase HPLC Method with UV Detection for the Determination of L-Dopa in Vicia faba L. Broad Beans.

L-Dopa (LD), a substance used medically in the treatment of Parkinson's disease, is found in several natural products, such as Vicia faba L., also known as broad beans. Due to its low chemical stability, LD analysis in plant matrices requires an appropriate optimization of the chosen analytical method to obtain reliable results. This work proposes an HPLC-UV method, validated according to EURACHEM guidelines as regards linearity, limits of detection and quantification, precision, accuracy, and matrix effect. The LD extraction was studied by evaluating its aqueous stability over 3 months. The best chromatographic conditions were found by systematically testing several C18 stationary phases and acidic mobile phases. In addition, the assessment of the best storage treatment of Vicia faba L. broad beans able to preserve a high LD content was performed. The best LD determination conditions include sun-drying storage, extraction in HCl 0.1 M, chromatographic separation with a Discovery C18 column, 250 × 4.6 mm, 5 µm particle size, and 99% formic acid 0.2% v/v and 1% methanol as the mobile phase. The optimized method proposed here overcomes the problems linked to LD stability and separation, thus contributing to the improvement of its analytical determination.

IoT Platforms and Security: An Analysis of the Leading Industrial/Commercial Solutions.

For simplifying and speeding up the development of the Internet of Things (IoT) ecosystem, there has been a proliferation of IoT platforms, built up according to different design principles, computing paradigms, technologies, and targets. This paper proposes a review of main examples populating the wide landscape of IoT platforms and their comparison based on the IoT-A reference architecture. In such a way, heterogeneous IoT platforms (both current and future) can be analyzed regardless of their low-level specifications but exclusively through the lens of those key functionalities and architectural building blocks that enable the interplay among devices, data flow, software, and stakeholders within the IoT ecosystem. Among these, security by design (i.e., the inclusion of security design principles, technology, and governance at every level) must be integrated into every tier, component, and application to minimize the risk of cyber threats and preserve the integrity of the IoT platforms, not only within individual components but also for all the components working together as a whole.

Gender Identification in a Two-Level Hierarchical Speech Emotion Recognition System for an Italian Social Robot.

The real challenge in Human-Robot Interaction (HRI) is to build machines capable of perceiving human emotions so that robots can interact with humans in a proper manner. Emotion varies accordingly to many factors, and gender represents one of the most influential ones: an appropriate gender-dependent emotion recognition system is recommended indeed. In this article, we propose a Gender Recognition (GR) module for the gender identification of the speaker, as a preliminary step for the final development of a Speech Emotion Recognition (SER) system. The system was designed to be installed on social robots for hospitalized and living at home patients monitoring. Hence, the importance of reducing the software computational effort of the architecture also minimizing the hardware bulkiness, in order for the system to be suitable for social robots. The algorithm was executed on the Raspberry Pi hardware. For the training, the Italian emotional database EMOVO was used. Results show a GR accuracy value of 97.8%, comparable with the ones found in the literature.

A Crosstalk- and Interferent-Free Dual Electrode Amperometric Biosensor for the Simultaneous Determination of Choline and Phosphocholine.

Choline (Ch) and phosphocholine (PCh) levels in tissues are associated to tissue growth and so to carcinogenesis. Till now, only highly sophisticated and expensive techniques like those based on NMR spectroscopy or GC/LC- high resolution mass spectrometry permitted Ch and PCh analysis but very few of them were capable of a simultaneous determination of these analytes. Thus, a never reported before amperometric biosensor for PCh analysis based on choline oxidase and alkaline phosphatase co-immobilized onto a Pt electrode by co-crosslinking has been developed. Coupling the developed biosensor with a parallel sensor but specific to Ch, a crosstalk-free dual electrode biosensor was also developed, permitting the simultaneous determination of Ch and PCh in flow injection analysis. This novel sensing device performed remarkably in terms of sensitivity, linear range, and limit of detection so to exceed in most cases the more complex analytical instrumentations. Further, electrode modification by overoxidized polypyrrole permitted the development of a fouling- and interferent-free dual electrode biosensor which appeared promising for the simultaneous determination of Ch and PCh in a real sample.

Untargeted analysis of pure snail slime and snail slime-induced Au nanoparticles metabolome with MALDI FT-ICR MS.

Chronic wounds result from the failure of the normal wound healing process. Any delay during the tissue repair process could be defined as chronic wound healing, potentially having a highly detrimental impact on human health. To face this problem, in the last years, the use of different technologies alternative to therapeutic agents is gaining more attention. The Helix aspersa snail slime-based products are increasingly being used for skin injury, thanks to their ability to make tissue repair processes faster. To date, a comprehensive overview of pure snail slime metabolome is not available. Besides, Au nanoparticles (AuNPs) technology is spreading rapidly in the medical environment, and the search for AuNPs "green" synthetic routes that involve natural products as precursor agents is demanded, alongside with a deep comprehension of the kind of species that actively take part in synthesis and product stabilization. The aim of this work is to characterize the metabolic profile of a pure snail slime sample, by an untargeted high-resolution mass spectrometry-based analysis. In addition, insights on AuNPs synthesis and stabilization by the main components of pure snail slime used to induce the synthesis were obtained. The untargeted analysis provided a large list of important classes of metabolites, that is, fatty acid derivatives, amino acids and peptides, carbohydrates and polyphenolic compounds that could be appreciated in both samples of slime, with and without AuNPs. Moreover, a direct comparison of the obtained results suggests that mostly nitrogen and sulfur-bearing metabolites take part in the synthesis and stabilization of AuNPs.

Smart Monitoring and Control in the Future Internet of Things.

The Internet of Things (IoT) and related technologies are promising in terms of realizing pervasive and smart applications, which, in turn, have the potential to improve the quality of life of people living in a connected world [...].

Biomolecules from snail mucus (Helix aspersa) conjugated gold nanoparticles, exhibiting potential wound healing and anti-inflammatory activity.

In this work, for the first time, snail slime from garden snails "Helix Aspersa Müller", has been used to induce the formation of eco-friendly gold nanoparticles (AuNPs-SS) suitable for biomedical applications. An AuNPs-SS comprehensive investigation was performed and AuNPs with an average particle size of 14 ± 6 nm were observed, stabilized by a slime snail-based organic layer. Indeed, as recognized in high-resolution MALDI-MS analyses, and corroborated by FESEM, UV-Vis, ATR-FTIR, and XPS results, it was possible to assess the main presence of peptides and amino acids as the main components of the slime, that, combined with the AuNPs confers on them interesting properties. More specifically, we tested, in vitro, the AuNPs-SS safety in human keratinocytes and their potential effect on wound healing as well as their anti-inflammatory properties in murine macrophages. Moreover, the AuNPs-SS treatment resulted in a significant increase of the urokinase-type plasminogen activator receptor (uPAR), essential for keratinocyte adhesion, spreading, and migration, together with the reduction of LPS-induced IL1-ß and IL-6 cytokine levels, and completely abrogated the synthesis of inducible nitric oxide synthase (iNOS).

Allosteric Enzyme-Based Biosensors-Kinetic Behaviours of Immobilised L-Lysine-α-Oxidase from Trichoderma viride: pH Influence and Allosteric Properties.

The present study describes the kinetics of L-lysine-α-oxidase (LO) from Trichoderma viride immobilised by co-crosslinking onto the surface of a Pt electrode. The resulting amperometric biosensor was able to analyse L-lysine, thus permitting a simple but thorough study of the kinetics of the immobilised enzyme. The kinetic study evidenced that LO behaves in an allosteric fashion and that cooperativity is strongly pH-dependent. Not less important, experimental evidence shows that cooperativity is also dependent on substrate concentration at high pH and behaves as predicted by the Monod-Wyman-Changeux model for allosteric enzymes. According to this model, the existence of two different conformational states of the enzyme was postulated, which differ in Lys species landing on LO to form the enzyme-substrate complex. Considerations about the influence of the peculiar LO kinetics on biosensor operations and extracorporeal reactor devices will be discussed as well. Not less important, the present study also shows the effectiveness of using immobilised enzymes and amperometric biosensors not only for substrate analysis, but also as a convenient tool for enzyme kinetic studies.

Validation of an Analytical Method for Nitrite and Nitrate Determination in Meat Foods for Infants by Ion Chromatography with Conductivity Detection.

Nitrate and nitrite as sodium or potassium salts are usually added to meat products to develop the characteristic flavor, to inhibit the growth of microorganisms (particularly Clostridium botulinum), and effectively control rancidity by inhibiting lipid oxidation. However, both nitrate and nitrite ions need to be monitored for ensuring the quality and safety of cured meats. In this work, for the first time the content of nitrite and nitrate ions in homogenized meat samples of baby foods was determined by a validated method based on ion chromatography (IC) coupled with conductivity detection. Recoveries of nitrate and nitrite ions in meat samples were not lower than 84 ± 6%. The detection limits of nitrate and nitrite were 0.08 mg L-1 and 0.13 mg L-1, respectively. Five commercial samples of homogenized meat, namely lamb, rabbit, chicken, veal, and beef, for infant feeding were investigated; while nitrite content was below the detection limit, nitrate ranged from 10.7 to 21.0 mg kg-1. The results indicated that nitrate contents were below the European (EU) fixed value of 200 mg kg-1, and an acceptable daily intake of 3.7 mg kg-1 was estimated.

Profiling of quercetin glycosides and acyl glycosides in sun-dried peperoni di Senise peppers (Capsicum annuum L.) by a combination of LC-ESI(-)-MS/MS and polarity prediction in reversed-phase separations.

Interest in targeted profiling of quercetin glycoconjugates occurring in edible foodstuffs continues to expand because of their recognized beneficial health effects. Quercetin derivatives encompass several thousands of chemically distinguishable compounds, among which there are several compounds with different glycosylations and acylations. Since reference standards and dedicated databases are not available, the mass spectrometric identification of quercetin glycoconjugates is challenging. A targeted liquid chromatography (LC) coupled with tandem mass spectrometry (MS/MS) was applied for screening quercetin glycoconjugates in edible peperoni di Senise peppers (Capsicum annuum L.), protected by the European Union with the mark PGI (i.e., Protected Geographical Indication), and cultivated in Basilicata (Southern Italy). Chromatographic separation was accomplished by reversed-phase liquid chromatography (RPLC) using water/acetonitrile as the mobile phase and detection was performed on a linear ion trap mass spectrometer fitted with an electrospray ionization (ESI) source operating in negative ion mode. A correlation between experimental RP chromatographic retention time and those predicted by partition coefficients (log P) along with MS/MS data and an in-house developed database (named QUEdb) provided deep coverage for sixteen quercetin glycoconjugates. Among them, eleven quercetin glycoconjugates were already described in the literature and five were reported for the first time. These last acyl glycosidic quercetin derivatives were tentatively identified as quercetin-(galloyl-rhamnoside)-hexoside, [C34H33O20]- at m/z 761.1; quercetin-(sinapoyl-hexoside)-rhamnoside, [C38H39O20]- at m/z 815.4; quercetin-(galloyl-caffeoyl-hexoside)-rhamnoside, [C43H39O23]- at m/z 923.0; quercetin-(feruloyl-hexoside)-rhamnoside, [C37H37O19]- at m/z 785.1; and quercetin-(succinyl-rhamnoside)-rhamnoside, [C31H33O18]- at m/z 693.1. Graphical abstract.

Assay of Phospholipase D Activity by an Amperometric Choline Oxidase Biosensor.

A novel electrochemical method to assay phospholipase D (PLD) activity is proposed based on the employment of a choline biosensor realized by immobilizing choline oxidase through co-crosslinking on an overoxidized polypyrrole film previously deposited on a platinum electrode. To perform the assay, an aliquot of a PLD standard solution is typically added to borate buffer containing phosphatidylcholine at a certain concentration and the oxidation current of hydrogen peroxide is then measured at the rotating modified electrode by applying a detection potential of + 0.7 V vs. SCE. Various experimental parameters influencing the assay were studied and optimized. The employment of 0.75% (v/v) Triton X-100, 0.2 mM calcium chloride, 5 mM phosphatidylcholine, and borate buffer at pH 8.0, ionic strength (I) 0.05 M allowed to achieve considerable current responses. In order to assure a controlled mass transport and, at the same time, high sensitivity, an electrode rotation rate of 200 rpm was selected. The proposed method showed a sensitivity of 24 (nA/s)(IU/mL)-1, a wide linear range up to 0.33 IU/mL, fast response time and appreciable long-term stability. The limit of detection, evaluated from the linear calibration curve, was 0.005 IU/mL (S/N = 3). Finally, due to the presence of overoxidized polypyrrole film characterized by notable rejection properties towards electroactive compounds, a practical application to real sample analysis can be envisaged.

Detection of choline in biological fluids from patients on haemodialysis by an amperometric biosensor based on a novel anti-interference bilayer.

A new and highly selective amperometric biosensor able to analyse choline in clinical samples from patients suffering from renal diseases and receiving repetitive haemodialysis treatment is described. The proposed biosensor is based on choline oxidase immobilized by co-crosslinking onto a novel anti-fouling and anti-interferent membrane. Between the several polymeric films electrosynthesized on a Pt electrode whose permselective behaviours were here investigated, those based on overoxidized polypyrrole/poly(o-aminophenol) bilayer revealed the most effective in rejecting common interferents usually present in biological fluids. The so realized biosensor showed notably analytical performances, displaying linear choline responses up to 100 µM, a sensitivity of 156 nA mM-1 mm-2 and a limit of detection, calculated at a signal-to-noise ratio equal to 3, of 1 µM; further, the within-a-day coefficients of variation for replicate (n = 3) were 2.7% and 1.2% at 100 µM and 10 µM choline levels, respectively. The remarkable performances and anti-interference behaviour allowed us the use of the proposed biosensor for the selective and fouling-free detection of choline in dialysate coming from patients on haemodialysis and even in their unpretreated human sera. Preliminary results gave choline levels in good agreement with the expected values.

Interactions between elastin-like peptides and an insulating poly(ortho-aminophenol) membrane investigated by AFM and XPS.

This investigation was undertaken to explore the mutual recognition of the pentapeptide (ValGlyGlyValGly)n, a hydrophobic elastin-like peptide (ELP), suspended in deionized water in monomer (n = 1) and trimer (n = 3) forms and the outer surface of a very thin, insulating polymer, poly(ortho-aminophenol) (PoAP), electrochemically grown on a platinum foil by cyclic voltammetry in a neutral medium (phosphate-buffered saline, I = 0.1M) immersed in the suspension. As a prior task, the proved propensity of the ValGlyGlyValGly sequence, at the given minimal length (three or more repeats), to self-assemble into amyloid-like fibrils when solubilized in an aqueous environment was considered within the framework of testing PoAP surfaces for the specific detection of amyloid precursors. From our knowledge of the chemical structure and physical properties of both biomacromolecule families obtained in previous studies, we focused on the efficacy of the binding sites offered to ELP fibrils by PoAP in its as-prepared form or properly modified either by postsynthesis oxidation or by adsorption/entrapping of ELP monomer(s) with or without protecting terminal groups. Consistent with all methods of preparation, the best surfaces, recognizable by the trimer fibrils, are those modified to carry a larger number of carbonyls, particularly by entrapment of ELP monomer(s) during PoAP electrosynthesis using an imprinting-inspired method. The degree of attachment of fibrillar aggregates, detected by atomic force microscopy and X-ray photoelectron spectroscopy, provides unequivocal evidence of the cooperative forces involving PoAP-ELP interactions. The results obtained suggest the prospect of using the proposed Pt/PoAP/ELP systems as biodetectors in Alzheimer disease. Graphical abstract Synthesis steps of Pt/PoAP/ELP electrodes for amyloid detection. AFM = Atomic Force Microscopy, CV = Cyclic Voltammetry, ELPs = Elastin like Peptides, PoAP = Poly ortho-Aminophenol, Pt = Platinum, XPS = X-ray Photoelectron Spectroscopy.

A novel approach for the selective analysis of l-lysine in untreated human serum by a co-crosslinked l-lysine-α-oxidase/overoxidized polypyrrole bilayer based amperometric biosensor.

An amperometric biosensor based on an l-lysine-α-oxidase (LO) layer immobilized by co-crosslinking onto the surface of an overoxidized polypyrrole modified Pt electrode (Pt/oPPy) and able to analyse l-lysine (Lys) in untreated human serum is described. The sensing electrode has been characterised and a proper enzyme kinetics optimisation permits to use a low specific enzyme as LO from Trichoderma viride for the selective biorecognition of Lys in the presence of other interferent amino acids; a kinetics study of LO evidenced also the allosteric behaviour of this enzyme, a kinetic feature which was never reported before for this enzyme. The biosensor showed a sensitivity of 0.11 µA/mM mm2, linear responses up to 4 mM and a limit of detection of 2 µM; the within-a-day coefficients of variation for replicate (n = 5) were 0.92% and 1.35% at 4 mM and 0.2 mM Lys levels, respectively. The permselective behaviour of Pt/oPPy modified electrode assured an interference- and fouling-free determination of Lys even in untreated serum samples. The determination of Lys in human serum from healthy donors gave Lys levels in good agreement with the expected values so that the use of the proposed biosensor appears promising in the relevant clinical fields.

Assay of serum cholinesterase activity by an amperometric biosensor based on a co-crosslinked choline oxidase/overoxidized polypyrrole bilayer.

Based on choline oxidase immobilized by co-crosslinking on an overoxidised polypyrrole modified platinum electrode, a novel electrochemical assay for cholinesterase activity in human serum was developed. The assay was performed by adding an aliquot of cholinesterase standard solution or serum sample to phosphate buffer containing choline or thiocholine ester and measuring the oxidation current of hydrogen peroxide at the rotating modified electrode polarized at +0.7 V vs. SCE. The influence of some experimental parameters such as pH of the assay, mass transport at the electrode, type and concentration of the cholinesterase substrate was studied and optimised. Reversible inhibition of choline oxidase from cholinesterase substrates was evidenced for the first time, which increases in the order of acetylcholine, butyrylcholine and s-butyrylthiocholine. Wide linear range, fast response time and appreciable long-term stability were assured for both acethyl- and butyrylcholinesterase assays. On allowing the polypyrrole layer to efficiently remove interferences from the electroactive compounds in the sample, the present method revealed to be suitable for the detection of butyrylcholinesterase in human serum at activities as low as 0.5 U L-1. The validation with a reference spectrophotometric method showed no significant differences when human serum samples were analysed.

Improved separation and size characterization of gold nanoparticles through a novel capillary zone electrophoresis method using poly(sodium4-styrenesulfonate) as stabiliser and a stepwise field strength gradient.

A novel capillary zone electrophoresis (CZE) method was developed for an improved separation and size characterization of pristine gold nanoparticles (AuNP) using uncoated fused-silica capillaries with UV-Vis detection at 520 nm. To avoid colloid aggregation and/or adsorption during runs, poly(sodium 4-styrenesulfonate) (PSS) was added (1%, w/v) in the running buffer (CAPS 10 mM, pH 11). This polyelectrolyte conferred an enhanced stabilization to AuNP, both steric and electrostatic, exalting at the same time their differences in electrophoretic mobility. Resolution was further and successfully improved through a stepwise field strength gradient by the application of 25 kV for the first 5 min and then 10 kV. Migration times varied linearly with particles diameters showing relative standard deviations better than 1% for daily experiments and 3% for interday experiments. A comparison with the size distribution obtained by transmission electron microscopy (TEM) allowed assessing that the electrophoretic profile can reasonably be considered as representative of the effective size heterogeneity of each colloid. Finally, the practical utility of the proposed method was demonstrated by measuring the core diameter of a gold colloid sample produced by chemical synthesis which was in good agreement with the value obtained by TEM measurements.