We can't find the internet
Attempting to reconnect
Something went wrong!
Hang in there while we get back on track
Papers
20 resultsShowing papers similar to Deep learning-enabled Inference of 3D molecular absorption distribution of biological cells from IR spectra
ClearA3M-CIR: An Attention-Aware Adversarial Masking Based Self-Supervised Chemometric Framework for Robust Infrared Spectral Analysis
Scientists developed a new computer system called A3M-CIR that can better analyze infrared light readings to identify chemicals and materials. The system learns to recognize patterns without needing lots of labeled examples, making it more reliable when analyzing samples under different conditions or with background noise. This could improve how we detect harmful substances like microplastics in food, water, and medical samples, leading to better health monitoring and safety testing.
Detection of Unlabeled Polystyrene Micro- and Nanoplastics in Mammalian Tissue by Optical Photothermal Infrared Spectroscopy
Researchers demonstrated that a new imaging technique called O-PTIR spectroscopy can detect unlabeled plastic particles as small as 200 nanometers inside mammalian tissues without damaging the samples. Combined with machine learning for faster analysis, this method significantly outperforms traditional infrared spectroscopy for finding nanoplastics in biological tissue. Better detection tools like this are essential for understanding how much plastic actually accumulates in human organs.
PlasticNet: Deep Learning for Automatic Microplastic Recognition via FT-IR Spectroscopy
Researchers developed PlasticNet, a deep learning algorithm that automatically identifies microplastic types from infrared spectral data, outperforming conventional library matching approaches. Automating microplastic identification could dramatically speed up the analysis of environmental samples and reduce human error.
Detection of Unlabeled Micro- and Nanoplastics in Unstained Tissue with Optical Photothermal Infrared Spectroscopy
Researchers demonstrated that optical photothermal infrared spectroscopy can detect unlabeled micro- and nanoplastics as small as 250 nanometers in mammalian tissue samples without staining or labeling. The technique significantly outperformed traditional FTIR spectroscopy in spatial resolution and signal quality when imaging particles in complex biological matrices. The study also introduced a semi-automated machine learning analysis to speed up detection, offering a powerful new tool for studying nanoplastic accumulation in tissues.
A Universal Approach to Mie Scatter Correction in FTIR Analysis of Microsized Samples
Researchers developed a deep-learning-based method to correct Mie scattering distortions in infrared microspectroscopy, enabling accurate chemical identification of microscopic samples including microplastic beads. The universal approach works across different sample types and spectroscopic setups without requiring prior knowledge of sample absorption properties, offering a significant improvement for microplastic analysis and other applications.
Deep Learning for Reconstructing Low-Quality FTIR and Raman Spectra─A Case Study in Microplastic Analyses
Researchers developed a deep learning method to reconstruct low-quality FTIR and Raman spectra, demonstrating its effectiveness for automated microplastic analysis where rapid measurement workflows produce noisy, challenging spectral datasets.
Infrared spectroscopic laser scanning confocal microscopy for whole-slide chemical imaging
Scientists developed a new infrared microscope that can create chemical images of entire tissue slides in about 3 minutes, far faster than existing methods. While not directly about microplastics, this type of imaging technology could significantly speed up the detection and identification of microplastic particles in human tissue samples. Faster, more detailed chemical imaging tools are needed to better understand where microplastics accumulate in the body and what damage they cause.
RepDwNet: Lightweight Deep Learning Model for Special Biological Blood Raman Spectra Analysis
Researchers developed a lightweight deep learning model called RepDwNet for analyzing Raman spectroscopy data from biological blood samples. The model achieved high accuracy while being small enough to run on portable spectrometer devices used in the field. The study demonstrates that advanced AI analysis of Raman spectra can be made practical for point-of-care and on-site testing applications without sacrificing analytical performance.
Computer-Assisted Analysis of Microplastics in Environmental Samples Based on μFTIR Imaging in Combination with Machine Learning
Researchers developed machine learning approaches for automated microplastic identification in environmental samples from micro-FTIR imaging data, demonstrating improved accuracy and speed compared to traditional spectral library search methods for scalable analysis.
Identification of extracellular vesicles from their Raman spectra via self-supervised learning
Researchers developed a deep learning method to identify and classify tiny biological particles called extracellular vesicles — which cells release and which may signal disease — using Raman spectroscopy without any chemical labels. The model achieved over 92% accuracy in distinguishing vesicles from different biological sources, including cancer patients versus healthy controls.
Functional Group Identification for FTIR Spectra Using Image-Based Machine Learning Models
Researchers developed a machine learning model that uses images of FTIR spectra to automatically identify chemical functional groups in unknown substances. This approach could speed up the identification of microplastic polymer types in environmental samples, making large-scale monitoring more efficient.
Construction of an Intelligent Identification Model for Drugs in Near Infrared Spectroscopy and Research on Drog Classification based on Improved Deep Algorithm
Researchers built an intelligent near-infrared spectroscopy model to identify pharmaceutical compounds, training and validating a machine learning classifier on spectral data from multiple drug types. The model achieved high classification accuracy and demonstrated the potential of NIR spectroscopy combined with AI for rapid, non-destructive drug identification.
Deep convolutional neural networks for aged microplastics identification by Fourier transform infrared spectra classification
This study developed a deep learning model using convolutional neural networks to automatically identify aged microplastics from their infrared spectra. Aging changes the chemical signature of plastics, making them harder to identify with conventional spectral databases. The AI approach achieved high accuracy and could significantly speed up the analysis of environmental samples where weathered microplastics are the norm.
Photothermal Infrared Imaging of Nanoplastics in Human Cells with Nanoscale Resolution
Researchers demonstrated a new photothermal infrared imaging technique capable of detecting and localizing nanoplastics inside individual human cells at nanoscale resolution. The study found that polystyrene nanoparticles accumulated around cell nuclei, and that this advanced imaging approach overcomes the spatial resolution limitations of conventional infrared spectroscopy for tracking nanoplastics in biological tissues.
Optical photothermal infrared spectroscopic assessment of microplastics in tissue models and non-digested human tissue sections
Researchers developed a method using optical photothermal infrared spectroscopy to detect and map microplastics directly within tissue sections without requiring chemical or enzymatic digestion. The study suggests this approach preserves spatial information about where microplastics are located within tissue architecture, overcoming a key limitation of conventional digestion-based methods that can lose some particles.
Infrared chemical imaging through non-degenerate two-photon absorption in silicon-based cameras
Researchers demonstrated a way to take mid-infrared chemical images — which reveal what materials are made of — using ordinary silicon-based camera chips instead of expensive specialized infrared detectors, by exploiting a quantum light absorption effect. This affordable approach successfully identified different types of polymers and biological samples, and could make chemical imaging tools widely accessible for detecting microplastics and other materials.
Label-free nano- and microplastics detection in mammalian tissue by photothermal infrared spectroscopy
Researchers applied optical photothermal infrared spectroscopy to detect and identify nano- and microplastics smaller than 1 µm in mammalian tissue sections without requiring labels or lengthy digestion protocols. The method successfully localized polystyrene particles in tissue samples with chemical specificity, offering a faster workflow for nanoplastic detection in biological matrices.
Inverse Infrared Spectral Deconvolution for Quantitative Analysis of Polymer Mixtures in Scattering Media
Researchers developed an inverse infrared spectral deconvolution algorithm capable of nondestructively identifying polymer components in strongly scattering mixed-composition microspheres without chromatographic separation or calibration. The method reconstructs pure absorption spectra for each component, determines volume fractions, and can be fully automated, addressing key challenges in analytical and forensic polymer characterization.
Saccharide concentration prediction from proxy sea surface microlayer samples analyzed via infrared spectroscopy and quantitative machine learning
Researchers developed infrared spectroscopy combined with machine learning to analyze dissolved organic carbon in sea surface microlayer samples more efficiently. The sea surface microlayer is a critical zone where microplastics concentrate and interact with marine chemistry.
Machine learning outperforms humans in microplastic characterization and reveals human labelling errors in FTIR data
Researchers developed a small but powerful neural network that can identify microplastic types from infrared spectroscopy data more accurately than human experts. The AI model classified 16 different categories of microplastics and even revealed errors in human-labeled data. This technology could dramatically speed up microplastic analysis in environmental and health studies, making it easier to understand the scale and types of microplastic contamination people are exposed to.