RESUMO
Surface-enhanced Raman spectroscopy (SERS) holds exceptional promise as a streamlined chemical detection strategy for biological and environmental contaminants compared with current laboratory methods. Priority pollutants such as polycyclic aromatic hydrocarbons (PAHs), detectable in water and soil worldwide and known to induce multiple adverse health effects upon human exposure, are typically found in multicomponent mixtures. By combining the molecular fingerprinting capabilities of SERS with the signal separation and detection capabilities of machine learning (ML), we examine whether individual PAHs can be identified through an analysis of the SERS spectra of multicomponent PAH mixtures. We have developed an unsupervised ML method we call Characteristic Peak Extraction, a dimensionality reduction algorithm that extracts characteristic SERS peaks based on counts of detected peaks of the mixture. By analyzing the SERS spectra of two-component and four-component PAH mixtures where the concentration ratios of the various components vary, this algorithm is able to extract the spectra of each unknown component in the mixture of unknowns, which is then subsequently identified against a SERS spectral library of PAHs. Combining the molecular fingerprinting capabilities of SERS with the signal separation and detection capabilities of ML, this effort is a step toward the computational demixing of unknown chemical components occurring in complex multicomponent mixtures.
Assuntos
Poluentes Ambientais , Hidrocarbonetos Policíclicos Aromáticos , Humanos , Hidrocarbonetos Policíclicos Aromáticos/análise , Análise Espectral Raman/métodos , Água , Poluentes Ambientais/análise , Misturas Complexas , Aprendizado de MáquinaRESUMO
Adversarial attacks are still a significant challenge for neural networks. Recent efforts have shown that adversarial perturbations typically contain high-frequency features, but the root cause of this phenomenon remains unknown. Inspired by theoretical work on linear convolutional models, we hypothesize that translational symmetry in convolutional operations together with localized kernels implicitly bias the learning of high-frequency features, and that this is one of the main causes of high frequency adversarial examples. To test this hypothesis, we analyzed the impact of different choices of linear and non-linear architectures on the implicit bias of the learned features and adversarial perturbations, in spatial and frequency domains. We find that, independently of the training dataset, convolutional operations have higher frequency adversarial attacks compared to other architectural parameterizations, and that this phenomenon is exacerbated with stronger locality of the kernel (kernel size) end depth of the model. The explanation for the kernel size dependence involves the Fourier Uncertainty Principle: a spatially-limited filter (local kernel in the space domain) cannot also be frequency-limited (local in the frequency domain). Using larger convolution kernel sizes or avoiding convolutions (e.g., by using Vision Transformers or MLP-style architectures) significantly reduces this high-frequency bias. Looking forward, our work strongly suggests that understanding and controlling the implicit bias of architectures will be essential for achieving adversarial robustness.
RESUMO
Background: Junctional ectopic tachycardia (JET) is a prevalent life-threatening arrhythmia in children with congenital heart disease. It has a marked resemblance to normal sinus rhythm, often leading to delay in diagnosis and management. Objective: The study sought to develop a novel multimodal automated arrhythmia detection tool that outperforms existing JET detection tools. Methods: This is a cohort study performed on 40 patients with congenital heart disease at Texas Children's Hospital. Electrocardiogram and central venous pressure waveform data produced by bedside monitors are captured by the Sickbay platform. Convolutional neural networks (CNNs) were trained to classify each heartbeat as either normal sinus rhythm or JET based only on raw electrocardiogram signals. Results: Our best model improved the area under the curve from 0.948 to 0.952 and the true positive rate at 5% false positive rate from 71.8% to 80.6%. Using a 3-model ensemble further improved the area under the curve to 0.953 and the true positive rate at 5% false positive rate to 85.2%. Results on a subset of data show that adding central venous pressure can significantly improve area under the receiver-operating characteristic curve from 0.646 to 0.825. Conclusion: This study validates the efficacy of deep neural networks to notably improve JET detection accuracy. We have built a performant and reliable model that can be used to create a bedside alarm that diagnoses JET, allowing for precise diagnosis of this life-threatening postoperative arrhythmia and prompt intervention. Future validation of the model in a larger cohort is needed.
RESUMO
Since its discovery, surface-enhanced Raman spectroscopy (SERS) has shown outstanding promise of identifying trace amounts of unknown molecules in rapid, portable formats. However, the many different types of nanoparticles or nanostructured metallic SERS substrates created over the past few decades show substantial variability in the SERS spectra they provide. These inconsistencies have even raised speculation that substrate-specific SERS spectral libraries must be compiled for practical use of this type of spectroscopy. Here, we report a machine learning (ML) algorithm that can identify chemicals by matching their SERS spectra to those of a standard Raman spectral library. We use an approach analogous to facial recognition that utilizes feature extraction in the presence of multiple nuisance variables for spectral recognition. The key element is a metric we call "Characteristic Peak Similarity" (CaPSim) that focuses on the characteristic peaks in the SERS spectra. It has the flexibility to accommodate substrate-specific variability when quantifying the degree of similarity to a Raman spectrum. Analysis shows that CaPSim substantially outperforms existing spectral matching algorithms in terms of accuracy. This ML-based approach could greatly facilitate the spectroscopic identification of molecules in fieldable SERS applications.
RESUMO
Background: Junctional ectopic tachycardia (JET) is a prevalent life-threatening arrhythmia in children with congenital heart disease (CHD), with marked resemblance to normal sinus rhythm (NSR) often leading to delay in diagnosis. Objective: To develop a novel automated arrhythmia detection tool to identify JET. Methods: A single-center retrospective cohort study of children with CHD was performed. Electrocardiographic (ECG) data produced by bedside monitors is captured automatically by the Sickbay platform. Based on the detection of R and P wave peaks, 2 interpretable ECG features are calculated: P prominence median and PR interval interquartile range (IQR). These features are used as input to a simple logistic regression classification model built to distinguish JET from NSR. Results: This study analyzed a total of 64.5 physician-labeled hours consisting of 509,833 cardiac cycles (R-R intervals), from 40 patients with CHD. The extracted P prominence median feature is much smaller in JET compared to NSR, whereas the PR interval IQR feature is larger in JET compared to NSR. The area under the receiver operating characteristic curve for the unseen patient test cohort was 93%. Selecting a threshold of 0.73 results in a true-positive rate of 90% and a false-positive rate of 17%. Conclusion: This novel arrhythmia detection tool identifies JET, using 2 distinctive features of JET in ECG-the loss of a normal P wave and PR relationship-allowing for early detection and timely intervention.
RESUMO
Stereocomplex (SC) crystallites, formed between poly(l-lactide) (PLLA) and poly(d-lactide), exhibit great potential to substantially enhance crystallization rate of PLLA-based materials as an eco-friendly nucleating agent. However, the nucleation efficiency of the SC crystallites is still far below an expected level, mostly on account of their strong aggregation tendency in PLLA/PDLA melts. Herein, taking PLLA/poly(ethylene-methyl acrylate-glycidyl methacrylate) (E-MA-GMA) blends as an example, we report a unique and facile strategy to control the dispersion and distribution of SC crystallites within the PLLA matrix by using elastomeric E-MA-GMA as carrier for the incorporation of PDLA. To do this, PDLA was first blended with E-MA-GMA or chemically grafted onto the E-MA-GMA. During subsequent melt-blending of PLLA and the E-MA-GMA/PDLA master batch, the PDLA chain clusters predispersed in the E-MA-GMA phase can gradually migrate into PLLA matrix and then collaborate with the matrix chains to form large amounts of tiny and well-dispersed SC crystallites. Compared with the SC-crystallite agglomerates formed by the direct melt-blending of PLLA and PDLA components, such tiny SC crystallites are much more effective in accelerating PLLA matrix crystallization. More interestingly, when PDLA chains are grafted onto the EMA-GMA, the formed SC crystallites tend to preferentially distribute at the blend interface and thus induce not only optimal nucleation efficiency but also superior impact toughness because these interface-localized SC crystallites can also serve as bridges to enhance interface adhesion. This work could open a new avenue in designing heat-resistant and supertough PLLA blends via controllable construction of SC crystallites.