Ambiguous Identification Research Articles

Full Transformer Framework for Robust Point Cloud Registration With Deep Information Interaction.

Point cloud registration is an essential technology in computer vision and robotics. Recently, transformer-based methods have achieved advanced performance in point cloud registration by utilizing the advantages of the transformer in order-invariance and modeling dependencies to aggregate information. However, they still suffer from indistinct feature extraction, sensitivity to noise, and outliers, owing to three major limitations: 1) the adoption of CNNs fails to model global relations due to their local receptive fields, resulting in extracted features susceptible to noise; 2) the shallow-wide architecture of transformers and the lack of positional information lead to indistinct feature extraction due to inefficient information interaction; and 3) the insufficient consideration of geometrical compatibility leads to the ambiguous identification of incorrect correspondences. To address the above-mentioned limitations, a novel full transformer network for point cloud registration is proposed, named the deep interaction transformer (DIT), which incorporates: 1) a point cloud structure extractor (PSE) to retrieve structural information and model global relations with the local feature integrator (LFI) and transformer encoders; 2) a deep-narrow point feature transformer (PFT) to facilitate deep information interaction across a pair of point clouds with positional information, such that transformers establish comprehensive associations and directly learn the relative position between points; and 3) a geometric matching-based correspondence confidence evaluation (GMCCE) method to measure spatial consistency and estimate correspondence confidence by the designed triangulated descriptor. Extensive experiments on the ModelNet40, ScanObjectNN, and 3DMatch datasets demonstrate that our method is capable of precisely aligning point clouds, consequently, achieving superior performance compared with state-of-the-art methods. The code is publicly available at https://github.com/CGuangyan-BIT/DIT.

Explainable artificial intelligence-driven prostate cancer screening using exosomal multi-marker based dual-gate FET biosensor

Prostate Imaging Reporting and Data System (PI-RADS) score, a reporting system of prostate MRI cases, has become a standard prostate cancer (PCa) screening method due to exceptional diagnosis performance. However, PI-RADS 3 lesions are an unmet medical need because PI-RADS provides diagnosis accuracy of only 30–40% at most, accompanied by a high false-positive rate. Here, we propose an explainable artificial intelligence (XAI) based PCa screening system integrating a highly sensitive dual-gate field-effect transistor (DGFET) based multi-marker biosensor for ambiguous lesions identification. This system produces interpretable results by analyzing sensing patterns of three urinary exosomal biomarkers, providing a possibility of an evidence-based prediction from clinicians. In our results, XAI-based PCa screening system showed a high accuracy with an AUC of 0.93 using 102 blinded samples with the non-invasive method. Remarkably, the PCa diagnosis accuracy of patients with PI-RADS 3 was more than twice that of conventional PI-RADS scoring. Our system also provided a reasonable explanation of its decision that TMEM256 biomarker is the leading factor for screening those with PI-RADS 3. Our study implies that XAI can facilitate informed decisions, guided by insights into the significance of visualized multi-biomarkers and clinical factors. The XAI-based sensor system can assist healthcare professionals in providing practical and evidence-based PCa diagnoses.

Ambiguous Identification and “Latinx”: Young Mexican Americans on the Border, Social Media, and a Disconnect in Marketing

This study takes up the relationship between young Mexican Americans living on the U.S–Mexico border and the term “Latinx” through their reception of Latinx-related content on Instagram. Using qualitative interviews where four Instagram pages (@wearemitu, @perolike, @mexicansproblemas, & @mijacultura) were used as stimuli for conversation, we found most participants engaged in complementary inclusivity with the label “Latinx.” This means that while they did not use or prefer the term for themselves, they had no issue with others using it. On the other hand, the shift in ethnic labels led to ambiguous identification, where it was unclear which label applied and what it meant. This manifested when interacting with content marketed to a “Latinx” audience that was not specific to their lived experience or contrived. Participants instead appreciated content that included the use of Spanish and quotidian humor, suggesting a disconnect in marketing and an incomplete understanding of what young Mexican Americans see as relatable.

Imaging of annulus-formation interface using reverse time migration based on ultrasonic pitch-catch measurements: Case studies from an experimental well and a field well

Cased-hole logging using ultrasonic pitch-catch modality is a proven technique for cement bond evaluation in cased holes. However, complex measuring environments such as tool and casing eccentering makes it difficult to precisely separate and pick the third interface echoes from full ultrasonic Lamb waves, leading to ambiguous identification of annulus-formation interface. To overcome this problem, we propose an improved cased-hole reverse time migration approach for ultrasonic pitch-catch measurements to image the annulus-formation interface. The missing ultrasonic Lamb waveforms between far and near receivers due to insufficient spatial sampling are reconstructed based on the optimized theoretical phase velocity of zero-order anti-symmetric Lamb mode waves. Additionally, we apply the hybrid illumination imaging condition to mitigate the imaging noises around the true reflectors and sources. Data examples from a physical experimental well and a field well demonstrate that the proposed approach is an effective method for characterizing the annulus-formation interface without requiring precise knowledge of the velocity distribution in the region behind the casing. Furthermore, test on an experimental well has indicated that the method has the potential to detect the absence of cement in the annulus medium.

Implementation of Intelligent Indoor Service Robot Based on ROS and Deep Learning

When faced with challenges such as adapting to dynamic environments and handling ambiguous identification, indoor service robots encounter manifold difficulties. This paper aims to address this issue by proposing the design of a service robot equipped with precise small-object recognition, autonomous path planning, and obstacle-avoidance capabilities. We conducted in-depth research on the suitability of three SLAM algorithms (GMapping, Hector-SLAM, and Cartographer) in indoor environments and explored their performance disparities. Upon this foundation, we have elected to utilize the STM32F407VET6 and Nvidia Jetson Nano B01 as our processing controllers. For the program design on the STM32 side, we are employing the FreeRTOS operating system, while for the Jetson Nano side, we are employing ROS (Robot Operating System) for program design. The robot employs a differential drive chassis, enabling successful autonomous path planning and obstacle-avoidance maneuvers. Within indoor environments, we utilized the YOLOv3 algorithm for target detection, achieving precise target identification. Through a series of simulations and real-world experiments, we validated the performance and feasibility of the robot, including mapping, navigation, and target detection functionalities. Experimental results demonstrate the robot’s outstanding performance and accuracy in indoor environments, offering users efficient service and presenting new avenues and methodologies for the development of indoor service robots.

DNA Barcoding for Species Identification of Moss-Dwelling Invertebrates: Performance of Nanopore Sequencing and Coverage in Reference Database

In view of the current biodiversity crisis and our need to preserve and improve ecosystem functioning, efficient means for characterizing and monitoring biodiversity are required. DNA barcoding, especially when coupled with new sequencing technologies, is a promising method that can, in principle, also be employed by taxonomic lay people. In this study we compare the performance of DNA barcoding by means of a third-generation sequencing technology, nanopore sequencing with classical Sanger sequencing, based on a sample of invertebrates collected from moss pads in a bog in Austria. We find that our nanopore sequencing pipeline generates DNA barcodes that are at least as good as barcodes generated with Sanger sequencing, with the MinION producing better results than the Flongle flowcell. We further find that while many arthropod taxa are well covered in the international reference DNA barcode database BOLD, this clearly is not the case for important taxa like mites and springtails, which hampers large-scale biodiversity assessments. Based on examples from our study we further highlight which factors might be responsible for ambiguous species identification based on BOLD and how this can, at least partly, be solved.

Optimization Method for Water-Flooded Beach-Bar Sand Bodies: A Case Study of the Fourth Member Red Beds of the Paleogene Shahejie Formation in the Dongying Depression

As water flooding continues to advance in mature oil fields, conventional well logging curve responses exhibit anomalies, particularly in deep-seated beach-bar thin-layer sand bodies. These sand bodies exhibit strong vertical and planar heterogeneity, which hinders a clear understanding of their distribution and connectivity. This paper conducts a sensitivity analysis of the lower part of the 4th member of the Shahejie Formation in fault block C26, based on core data analysis and the integration of drilling, logging, recording, and production dynamic data. Natural potential and resistivity curves are selected as the sensitive curves for identifying sand bodies. Preliminary processing of well logging curves is carried out using the “standard layer constraint and multi-standardization method comparison” principle. Employing sedimentological research methods and the principles of wavelet transformation, the well logging curves within the study area undergo extraction of both high and low frequencies. This procedure accentuates details related to thin sand bodies and responses indicative of sedimentary cyclicity. Through a thoughtful amalgamation of multiple curves, the investigation achieves a systematic fusion of natural potential curves via multi-curve frequency division fusion, employing reconstruction optimization. This method adeptly mitigates interference stemming from water-flooded layers, effectively addressing challenges such as excessive calibration and ambiguous identification of sand bodies. As a result, a comprehensive analytical approach is established for assessing the distribution and connectivity of deep-seated, beach-bar thin sand bodies influenced by water-flooded layers, providing clarity on the connectivity relationships among the sand bodies. Additionally, in combination with the results of mercury injection experiments in the water-flooded layer segments, favorable reservoir criteria for the study area are determined, providing a scientific basis for adjusting future development plans.

Genetic and morphological analysis of complex Giuris group species from Limboto Lake, Gorontalo Province, Indonesia

Abstract. Lamadi A, Paricahya AF, Iranawati F, Widodo MS, Wiadnya DGR. 2023. Genetic and morphological analysis of complex Giuris group species from Limboto Lake, Gorontalo Province, Indonesia. Biodiversitas 24: 5223-5233. Wallacea, especially Sulawesi, has a high endemism level with genuine biodiversity for mammals, reptiles, freshwater, and brackish fishes. Fish biodiversity in Limboto Lake, a great natural lake of Gorontalo Province, Indonesia, still needs to be scientifically explored, in contrast with its utilization for human activities that have anthropogenic effects on their ecosystem. Eleotridae is a taxonomical complex to their identification and distributed well in Sulawesi, including Limboto Lake with genus Giuris, known locally as hulu'u fish. Ambiguous identification in Giuris margaritaceus morphologically causing misidentification in molecular analysis, including using Cytochrome Oxidase subunit I (COI) region's information. Giuris specimens from Limboto Lake were collected from June 2022 until May 2023. From 662 individuals, length and weight relationship analysis results showed the population is mildly positive allometric (W= 0.11TL3.08). Meristic measurement result (D1 VI; D2 I.8-9; A I.9; P 14-16; V I.5; C 13-14) directing to G. margaritaceus with several small differences. COI analysis using six random specimens of the Limboto Lake population, comparing with four population from NCBI and BOLD System identified as G. margaritaceus or its synonym Ophieleotris aporos, indicating all populations is separated species from one another with P-distances more than 2% and made separated groups in phylogenetic tree and haplotype network. Misidentification issues in the Giuris margaritaceus taxonomical name for all five populations indicate they're potentially separated into five different species, with or without including Giuris margaritaceus, and potentially from currently described species or being new species in the future. Limboto Lake population made three haplotypes with high haplotype diversity (0.733) and low nucleotide diversity (0.001), indicating this population is facing high stressors and starting to grow. Understanding Giuris ambiguities could help governments, researchers, conservationists, and other stakeholders to mapping the ecological priorities.

Detailed integrative taxonomic analysis reveals large-scale species misidentification of barnacles based on DNA barcoding data

Abstract Morphological species identification is the traditional way to identify species. More recently, biodiversity studies have depended on DNA barcoding for species identification. Mismatches between morphological and genetic species identification caused by voucher specimen misidentifications lead to ambiguous species identification. We used barnacles of the genera Amphibalanus and Balanus to investigate this issue. The purpose of this study was to assess the accuracy of the Barcoding of Life database and determine the true species affinity of three inconsistent barnacle barcode index numbers using an integrative taxonomic approach. To this end, we re-identified barcoded voucher specimens, or we identified specimens based on a dichotomous key that we barcoded subsequently. We found that 75% of Barcoding of Life clades representing genetic species showed inconsistent species assignments. One clade (BOLD: AAG0069), containing mostly unidentified Balanus sp. individuals, was morphologically and genetically Balanus balanus (Linneaus, 1758); a second clade (BOLD: AAB1410), containing specimens mostly identified as B. balanus, turned out to be morphologically and genetically Balanus crenatus Bruguière, 1789; and a third clade (BOLD: AAE2482), containing different species of Amphibalanus, was Amphibalanus improvisus (Darwin, 1854). This study showed that morphological species identification and taxonomic expertise can resolve the widespread misidentification in barcoding data. To prevent this problem in the future, morphological taxonomy and molecular species identification need to interact more closely.

Characterization and Comparative Analysis of the Complete Plastomes of Five Epidendrum (Epidendreae, Orchidaceae) Species.

Epidendrum, one of the three largest genera of Orchidaceae, exhibits significant horticultural and ornamental value and serves as an important research model in conservation, ecology, and evolutionary biology. Given the ambiguous identification of germplasm and complex evolutionary relationships within the genus, the complete plastome of this genus (including five species) were firstly sequenced and assembled to explore their characterizations. The plastomes exhibited a typical quadripartite structure. The lengths of the plastomes ranged from 147,902 bp to 150,986 bp, with a GC content of 37.16% to 37.33%. Gene annotation revealed the presence of 78-82 protein-coding genes, 38 tRNAs, and 8 rRNAs. A total of 25-38 long repeats and 130-149 SSRs were detected. Analysis of relative synonymous codon usage (RSCU) indicated that leucine (Leu) was the most and cysteine (Cys) was the least. The consistent and robust phylogenetic relationships of Epidendrum and its closely related taxa were established using a total of 43 plastid genomes from the tribe Epidendreae. The genus Epidendrum was supported as a monophyletic group and as a sister to Cattleya. Meanwhile, four mutational hotspots (trnCGCA-petN, trnDGUC-trnYGUA, trnSGCU-trnGUCC, and rpl32-trnLUAG) were identified for further phylogenetic studies. Our analysis demonstrates the promising utility of plastomes in inferring the phylogenetic relationships of Epidendrum.

From bites to barcodes: uncovering the hidden diversity of black flies (Diptera: Simuliidae) in Vietnam

BackgroundPrompt and precise identification of black flies (Simuliidae) is crucial, given their biting behaviour and significant impact on human and animal health. To address the challenges presented by morphology and chromosomes in black fly taxonomy, along with the limited availability of molecular data pertaining to the black fly fauna in Vietnam, this study employed DNA-based approaches. Specifically, we used mitochondrial and nuclear-encoded genes to distinguish nominal species of black flies in Vietnam.MethodsIn this study, 135 mitochondrial cytochrome c oxidase subunit I (COI) sequences were established for 45 species in the genus Simulium in Vietnam, encompassing three subgenera (Gomphostilbia, Nevermannia, and Simulium), with 64 paratypes of 27 species and 16 topotypes of six species. Of these COI sequences, 71, representing 27 species, are reported for the first time.ResultsCombined with GenBank sequences of specimens from Malaysia, Myanmar, Thailand, and Vietnam, a total of 234 DNA barcodes of 53 nominal species resulted in a 71% success rate for species identification. Species from the non-monophyletic Simulium asakoae, S. feuerborni, S. multistriatum, S. striatum, S. tuberosum, and S. variegatum species groups were associated with ambiguous or incorrect identifications. Pairwise distances, phylogenetics, and species delimitation analyses revealed a high level of cryptic diversity, with discovery of 15 cryptic taxa. The current study also revealed the limited utility of a fast-evolving nuclear gene, big zinc finger (BZF), in discriminating closely related, morphologically similar nominal species of the S. asakoae species group.ConclusionThis study represents the first comprehensive molecular genetic analysis of the black fly fauna in Vietnam to our knowledge, providing a foundation for future research. DNA barcoding exhibits varying levels of differentiating efficiency across species groups but is valuable in the discovery of cryptic diversity.Graphical abstract

On the challenge of unambiguous identification of fentanyl analogs: Exploring measurement diversity using standard reference mass spectral libraries.

Fentanyl analogs are a class of designer drugs that are particularly challenging to unambiguously identify due to the mass spectral and retention time similarities of unique compounds. In this paper, we use agglomerative hierarchical clustering to explore the measurement diversity of fentanyl analogs and better understand the challenge of unambiguous identifications using analytical techniques traditionally available to drug chemists. We consider four measurements in particular: gas chromatography retention indices, electron ionization mass spectra, electrospray ionization tandem mass spectra, and direct analysis in real time mass spectra. Our analysis demonstrates how simultaneously considering data from multiple measurement techniques increases the observable measurement diversity of fentanyl analogs, which can reduce identification ambiguity. This paper further supports the use of multiple analytical techniques to identify fentanyl analogs (among other substances), as is recommended by the Scientific Working Group for the Analysis of Seized Drugs (SWGDRUG).

Characterization and Comparative Analysis of Chloroplast Genomes of Medicinal Herb Scrophularia ningpoensis and Its Common Adulterants (Scrophulariaceae).

Scrophularia ningpoensis, a perennial medicinal plant from the Scrophulariaceae family, is the original species of Scrophulariae Radix (SR) in the Chinese Pharmacopoeia. This medicine is usually deliberately substituted or accidentally contaminated with other closely related species including S. kakudensis, S. buergeriana, and S. yoshimurae. Given the ambiguous identification of germplasm and complex evolutionary relationships within the genus, the complete chloroplast genomes of the four mentioned Scrophularia species were sequenced and characterized. Comparative genomic studies revealed a high degree of conservation in genomic structure, gene arrangement, and content within the species, with the entire chloroplast genome spanning 153,016-153,631 bp in full length, encoding 132 genes, including 80 protein-coding genes, 4 rRNA genes, 30 tRNA genes, and 18 duplicated genes. We identified 8 highly variable plastid regions and 39-44 SSRs as potential molecular markers for further species identification in the genus. The consistent and robust phylogenetic relationships of S. ningpoensis and its common adulterants were firstly established using a total of 28 plastid genomes from the Scrophulariaceae family. In the monophyletic group, S. kakudensis was determined to be the earliest diverging species, succeeded by S. ningpoensis. Meanwhile, S. yoshimurae and S. buergeriana were clustered together as sister clades. Our research manifestly illustrates the efficacy of plastid genomes in identifying S. ningpoensis and its counterfeits and will also contribute to a deeper understanding of the evolutionary processes within Scrophularia.

Bayesian-Optimized Hybrid Kernel SVM for Rolling Bearing Fault Diagnosis.

We propose a new fault diagnosis model for rolling bearings based on a hybrid kernel support vector machine (SVM) and Bayesian optimization (BO). The model uses discrete Fourier transform (DFT) to extract fifteen features from vibration signals in the time and frequency domains of four bearing failure forms, which addresses the issue of ambiguous fault identification caused by their nonlinearity and nonstationarity. The extracted feature vectors are then divided into training and test sets as SVM inputs for fault diagnosis. To optimize the SVM, we construct a hybrid kernel SVM using a polynomial kernel function and radial basis kernel function. BO is used to optimize the extreme values of the objective function and determine their weight coefficients. We create an objective function for the Gaussian regression process of BO using training and test data as inputs, respectively. The optimized parameters are used to rebuild the SVM, which is then trained for network classification prediction. We tested the proposed diagnostic model using the bearing dataset of the Case Western Reserve University. The verification results show that the fault diagnosis accuracy is improved from 85% to 100% compared with the direct input of vibration signal into the SVM, and the effect is significant. Compared with other diagnostic models, our Bayesian-optimized hybrid kernel SVM model has the highest accuracy. In laboratory verification, we took sixty sets of sample values for each of the four failure forms measured in the experiment, and the verification process was repeated. The experimental results showed that the accuracy of the Bayesian-optimized hybrid kernel SVM reached 100%, and the accuracy of five replicates reached 96.7%. These results demonstrate the feasibility and superiority of our proposed method for fault diagnosis in rolling bearings.

DNA Barcoding Culicoides Biting Midges (Diptera: Ceratopogonidae) in Northeast Brazil.

Biting midges of the genus Culicoides are small insects associated with the transmission of several pathogens, which requires the correct identification of the species, for implementation of effective strategies against these insects. However, many species are difficult to identify only by morphological characters. Therefore, the use of molecular methods can help in the taxonomy and systematics of this group. Here, the DNA barcode approach was evaluated for nine species of Culicoides from the State of Maranhão, Brazil. We generated 39 sequences from a 476 bp (base pairs) fragment of the cytochrome c oxidase subunit I (COI) mitochondrial gene. To assess the usefulness of COI barcodes for the identification of these species, paired genetic distances from intra and interspecific comparisons and phylogenetic trees were generated in MEGA and RAxML/BEAST softwares, respectively. In addition, species delimitation was performed using the PTP, GMYC, and ABGD algorithms. The intra and interspecific genetic distances showed a clear distinction between them, demonstrating that, for the taxa studied, there can hardly be ambiguous identifications with barcodes. In the same sense, the phylogenetic reconstruction resulted in well-supported clades for all morphospecies analyzed.

Structural insights into thrombolytic activity of destabilase from medicinal leech

Destabilase from the medical leech Hirudo medicinalis belongs to the family of i-type lysozymes. It has two different enzymatic activities: microbial cell walls destruction (muramidase activity), and dissolution of the stabilized fibrin (isopeptidase activity). Both activities are known to be inhibited by sodium chloride at near physiological concentrations, but the structural basis remains unknown. Here we present two crystal structures of destabilase, including a 1.1 Å-resolution structure in complex with sodium ion. Our structures reveal the location of sodium ion between Glu34/Asp46 residues, which were previously recognized as a glycosidase active site. While sodium coordination with these amino acids may explain inhibition of the muramidase activity, its influence on previously suggested Ser49/Lys58 isopeptidase activity dyad is unclear. We revise the Ser49/Lys58 hypothesis and compare sequences of i-type lysozymes with confirmed destabilase activity. We suggest that the general base for the isopeptidase activity is His112 rather than Lys58. pKa calculations of these amino acids, assessed through the 1 μs molecular dynamics simulation, confirm the hypothesis. Our findings highlight the ambiguity of destabilase catalytic residues identification and build foundations for further research of structure–activity relationship of isopeptidase activity as well as structure-based protein design for potential anticoagulant drug development.

High Interobserver Variability Among Pathologists Using Combined Positive Score to Evaluate PD-L1 Expression in Gastric, Gastroesophageal Junction, and Esophageal Adenocarcinoma

Reliable, reproducible methods to interpret programmed death ligand-1 (PD-L1) expression on tumor cells (TC) and immune cells (IC) are needed for pathologists to inform decisions associated with checkpoint inhibitor therapies. Our international study compared interpathologist agreement of PD-L1 expression using the combined positive score (CPS) under standardized conditions on samples from patients with gastric/gastroesophageal junction/esophageal adenocarcinoma. Tissue sections from 100 adenocarcinoma pretreatment biopsies were stained in a single laboratory using the PD-L1 immunohistochemistry 28-8 and 22C3 (Agilent) pharmDx immunohistochemical assays. PD-L1 CPS was evaluated by 12 pathologists on scanned whole slide images of these biopsies before and after a 2-hour CPS training session by Agilent. Additionally, pathologists determined PD-L1–positive TC, IC, and total viable TC on a single tissue fragment from 35 of 100 biopsy samples. Scoring agreement among pathologists was assessed using the intraclass correlation coefficient (ICC). Interobserver variability for CPS for 100 biopsies was high, with only fair agreement among pathologists both pre- (range, 0.45-0.55) and posttraining (range, 0.56-0.57) for both assays. For the 35 single biopsy samples, poor/fair agreement was also observed for the total number of viable TC (ICC, 0.09), number of PD-L1–positive IC (ICC, 0.19), number of PD-L1–positive TC (ICC, 0.54), and calculated CPS (ICC, 0.14), whereas calculated TC score (positive TC/total TC) showed excellent agreement (ICC, 0.82). Retrospective histologic review of samples with the poorest interpathologist agreement revealed the following as possible confounding factors: (1) ambiguous identification of positively staining stromal cells, (2) faint or variable intensity of staining, (3) difficulty in distinguishing membranous from cytoplasmic tumor staining, and (4) cautery and crush artifacts. These results emphasize the need for objective techniques to standardize the interpretation of PD-L1 expression when using the CPS methodology on gastric/gastroesophageal junction cancer biopsies to accurately identify patients most likely to benefit from immune checkpoint inhibitor therapy.

Remembering the Herero-Nama Genocide in Namibia

Remembering the Herero-Nama Genocide in Namibia

Novel strategy for space group determination in real space

Traditional space group determination methods are all in reciprocal space, which involves ambiguous identification on some space groups which have glide plane and screw axes. The novel strategy herein for space group determination in real space is based on the atom resolution high angle annular dark field (HAADF) technology. Three HAADF images in three specific crystal zone axes are needed at most. The proposed strategy for space group determination is easy and effective.

A machine learning-based approach for mapping leachate contamination using geoelectrical methods

Leachate is the main source of pollution in landfills and its negative impacts continue for several years even after landfill closure. In recent years, geophysical methods are recognized as effective tools for providing an imaging of the leachate plume. However, they produce subsurface cross-sections in terms of individual physical quantities, leaving room for ambiguities on interpretation of geophysical models and uncertainties in the definition of contaminated zones. In this work, we propose a machine learning-based approach for mapping leachate contamination through an effective integration of geoelectrical tomographic data. We apply the proposed approach for the characterization of two urban landfills. For both cases, we perform a multivariate analysis on datasets consisting of electrical resistivity, chargeability and normalized chargeability (chargeability-to-resistivity ratio) data extracted from previously inverted model sections. By executing a K-Means cluster analysis, we find that the best partition of the two datasets contains ten and eleven classes, respectively. From such classes and also introducing a distance-based colour code, we get updated cross-sections and provide an easy and less ambiguous identification of the leachate accumulation zones. The latter turn out to be characterized by coordinate values of cluster centroids<3 Ωm and >27 mV/V and 11 mS/m. Our findings, also supported by borehole data for one of the investigation sites, show that the combined use of geophysical imaging and unsupervised machine learning is promising and can yield new perspectives for the characterization of leachate distribution and pollution assessment in landfills.