Evaluation of an autoencoder as a feature extraction tool for near-infrared spectroscopic discriminant analysis

Safe-by-Design Strategies for Intranasal Drug Delivery Systems: Machine and Deep Learning Solutions to Differentiate Epithelial Tissues via Attenuated Total Reflection Fourier Transform Infrared Spectroscopy

The development of nasal drug delivery systems requires advanced analytical techniques and tools that allow for distinguishing between the nose-to-brain epithelial tissues with better precision, where traditional bioanalytical methods frequently fail. In this study, attenuated total reflection Fourier transform infrared (ATR-FTIR) spectroscopy is coupled to machine learning (ML) and deep learning (DL) techniques to discriminate effectively between epithelial tissues. The primary goal of this work was to develop Safe-by-Design models for intranasal drug delivery using ex vivo pig tissues experiment, which were analyzed by way of ML modeling. We compiled an ATR-FTIR spectral data set from olfactory epithelium (OE), respiratory epithelium (RE), and tracheal tissues. The data set was used to train and test different ML algorithms. Accuracy, sensitivity, specificity, and F1 score metrics were used to evaluate optimized model performance and their abilities to identify specific spectral signatures relevant to each tissue type. The used feedforward neural network (FNN) has shown 0.99 accuracy, indicating that it had performed a discrimination with a high level of trueness estimates, without overfitting, unlike the built support vector machine (SVM) model. Important spectral features detailing the assignment and site of two-dimensional (2D) protein structures per tissue type were determined by the SHapley Additive exPlanations (SHAP) value analysis of the FNN model. Furthermore, a denoising autoencoder was built to improve spectral quality by reducing noise, as confirmed by higher Pearson correlation coefficients for denoised spectra. The combination of spectroscopic analysis with ML modeling offers a promising strategy called, Safe-by-Design, as a monitoring strategy for intranasal drug delivery systems, also for designing the analysis of tissue for diagnosis purposes.

CVAE-DF: A hybrid deep learning framework for fertilization status detection of pre-incubation duck eggs based on VIS/NIR spectroscopy

Unfertilized duck eggs not removed prior to incubation will deteriorate quickly, posing a risk of contaminating the normally fertilized duck eggs. Thus, detecting the fertilization status of breeding duck eggs as early as possible is a meaningful and challenging task. Most existing work usually focus on the characteristics of chicken eggs during mid-term hatching. However, little attention has been paid to the detection for duck eggs prior to incubation. In this paper, we present a novel hybrid deep learning detection framework for the fertilization status of pre-incubation duck eggs, termed CVAE-DF, based on visible/near-infrared (VIS/NIR) transmittance spectroscopy. The framework comprises the encoder of a convolutional variational autoencoder (CVAE) and an improved deep forest (DF) model. More specifically, we first collected transmittance spectral data (400-1000 nm) of 255 duck eggs before hatching. The multiplicative scatter correction (MSC) method was then used to eliminate noise and extraneous information of the raw spectral data. Two efficient data augmentation methods were adopted to provide sufficient data. After that, CVAE was applied to extract representative features and reduce the feature dimension for the detection task. Finally, an improved DF model was employed to build the classification model on the enhanced feature set. The CVAE-DF model achieved an overall accuracy of 95.94 % on the test dataset. These experimental results in terms of four metrics demonstrate that our CVAE-DF method outperforms the traditional methods by a significant margin. Furthermore, the results also indicate that CVAE holds great promise as a novel feature extraction method for the VIS/NIR spectral analysis of other agricultural products. It is extremely beneficial to practical engineering.

Semi-Supervised Autoencoder for Chemical Gas Classification with FTIR Spectrum

Chemical warfare agents pose a serious threat due to their extreme toxicity, necessitating swift the identification of chemical gases and individual responses to the identified threats. Fourier transform infrared (FTIR) spectroscopy offers a method for remote material analysis, particularly in detecting colorless and odorless chemical agents. In this paper, we propose a deep neural network utilizing a semi-supervised autoencoder (SSAE) for the classification of chemical gases based on FTIR spectra. In contrast to traditional methods, the SSAE concurrently trains an autoencoder and a classifier attached to a latent vector of the autoencoder, enhancing feature extraction for classification. The SSAE was evaluated on laboratory-collected FTIR spectra, demonstrating a superior classification performance compared to existing methods. The efficacy of the SSAE lies in its ability to generate denser cluster distributions in latent vectors, thereby enhancing gas classification. This study established a consistent experimental environment for hyperparameter optimization, offering valuable insights into the influence of latent vectors on classification performance.

Spectral data fusion in nondestructive detection of food products: Strategies, recent applications, and future perspectives

In recent years, the food industry has shown a growing interest in the development of rapid and nondestructive analytical methods. However, the utilization of a solitary nondestructive detection technique offers only a constrained extent of physical or chemical insights regarding the sample under examination. To overcome this limitation, the amalgamation of spectroscopy with data fusion strategies has emerged as a promising approach. This comprehensive review delves into the fundamental principles and merits of low-level, mid-level, and high-level data fusion strategies within the domain of food analysis. Various data fusion techniques encompassing spectra-to-spectra, spectra-to-machine vision, spectra-to-electronic nose, and spectra-to-nuclear magnetic resonance are summarized. Moreover, this review also provides an overview of the latest applications of spectral data fusion techniques (SDFTs) for classification, adulteration, quality evaluation, and contaminant detection within the purview of food safety analysis. It also addresses current challenges and future prospects associated with SDFTs in real-world applications. Despite the extant technical intricacy, the ongoing evolution of online data fusion platforms and the emergence of smartphone-based multi-sensor fusion detection technology augur well for the pragmatic realization of SDFTs, endowing them with formidable capabilities for both qualitative and quantitative analysis in the realm of food analysis.

Recent trends of machine learning applied to multi-source data of medicinal plants

In traditional medicine and ethnomedicine, medicinal plants have long been recognized as the basis for materials in therapeutic applications worldwide. In particular, the remarkable curative effect of traditional Chinese medicine during corona virus disease 2019 (COVID-19) pandemic has attracted extensive attention globally. Medicinal plants have, therefore, become increasingly popular among the public. However, with increasing demand for and profit with medicinal plants, commercial fraudulent events such as adulteration or counterfeits sometimes occur, which poses a serious threat to the clinical outcomes and interests of consumers. With rapid advances in artificial intelligence, machine learning can be used to mine information on various medicinal plants to establish an ideal resource database. We herein present a review that mainly introduces common machine learning algorithms and discusses their application in multi-source data analysis of medicinal plants. The combination of machine learning algorithms and multi-source data analysis facilitates a comprehensive analysis and aids in the effective evaluation of the quality of medicinal plants. The findings of this review provide new possibilities for promoting the development and utilization of medicinal plants.

Exploring the Steps of Infrared (IR) Spectral Analysis: Pre-Processing, (Classical) Data Modelling, and Deep Learning

Infrared (IR) spectroscopy has greatly improved the ability to study biomedical samples because IR spectroscopy measures how molecules interact with infrared light, providing a measurement of the vibrational states of the molecules. Therefore, the resulting IR spectrum provides a unique vibrational fingerprint of the sample. This characteristic makes IR spectroscopy an invaluable and versatile technology for detecting a wide variety of chemicals and is widely used in biological, chemical, and medical scenarios. These include, but are not limited to, micro-organism identification, clinical diagnosis, and explosive detection. However, IR spectroscopy is susceptible to various interfering factors such as scattering, reflection, and interference, which manifest themselves as baseline, band distortion, and intensity changes in the measured IR spectra. Combined with the absorption information of the molecules of interest, these interferences prevent direct data interpretation based on the Beer-Lambert law. Instead, more advanced data analysis approaches, particularly artificial intelligence (AI)-based algorithms, are required to remove the interfering contributions and, more importantly, to translate the spectral signals into high-level biological/chemical information. This leads to the tasks of spectral pre-processing and data modeling, the main topics of this review. In particular, we will discuss recent developments in both tasks from the perspectives of classical machine learning and deep learning.

Encoder-decoder neural networks for predicting future FTIR spectra - application to enzymatic protein hydrolysis

In the process of converting food-processing by-products to value-added ingredients, fine grained control of the raw materials, enzymes and process conditions ensures the best possible yield and economic return. However, when raw material batches lack good characterization and contain high batch variation, online or at-line monitoring of the enzymatic reactions would be beneficial. We investigate the potential of deep neural networks in predicting the future state of enzymatic hydrolysis as described by Fourier-transform infrared spectra of the hydrolysates. Combined with predictions of average molecular weight, this provides a flexible and transparent tool for process monitoring and control, enabling proactive adaption of process parameters.

Classification of Textile Samples Using Data Fusion Combining Near- and Mid-Infrared Spectral Information

There is an urgent need to reuse and recycle textile fibers, since today, low recycling rates are achieved. Accurate classification methods for post-consumer textile waste are needed in the short term for a higher circularity in the textile and fashion industries. This paper compares different spectroscopic data from textile samples in order to correctly classify the textile samples. The accurate classification of textile waste results in higher recycling rates and a better quality of the recycled materials. The data fusion of near- and mid-infrared spectra is compared with single-spectrum information. The classification results show that data fusion is a better option, providing more accurate classification results, especially for difficult classification problems where the classes are wide and close to one another. The experimental results presented in this paper prove that the data fusion of near- and mid-infrared spectra is a good option for accurate textile-waste classification, since this approach allows the classification results to be significantly improved.

Image classification of forage grasses on Etuoke Banner using edge autoencoder network

Automatically identifying the forage is the basis of intelligent fine breeding of cattle and sheep. In specific, it is a key step to study the relationship between the type and quantity of forage collected by cattle and sheep and their own growth, cashmere fineness, milk quality, meat quality and flavor, and so on. However, traditional method mainly rely on manual observation, which is time-consuming, laborious and inaccurate, and affects the normal grazing behavior of livestock. In this paper, the optimized Convolution Neural Network(CNN): edge autoencoder network(E-A-Net) algorithm is proposed to accurately identify the forage species, which provides the basis for ecological workers to carry out grassland evaluation, grassland management and precision feeding. We constructed the first forage grass dataset about Etuoke Banner. This dataset contains 3889 images in 22 categories. In the data preprocessing stage, the random cutout data enhancement is adopted to balance the original data, and the background is removed by employing threshold value-based image segmentation operation, in which the accuracy of herbage recognition in complex background is significantly improved. Moreover, in order to avoid the phenomenon of richer edge information disappearing in the process of multiple convolutions, a Sobel operator is utilized in this E-A-Net to extract the edge information of forage grasses. Information is integrated with the features extracted from the backbone network in multi-scale. Additionally, to avoid the localization of the whole information during the convolution process or alleviate the problem of the whole information disappearance, the pre-training autoencoder network is added to form a hard attention mechanism, which fuses the abstracted overall features of forage grasses with the features extracted from the backbone CNN. Compared with the basic CNN, E-A-Net alleviates the problem of edge information disappearing and overall feature disappearing with the deepening of network depth. Numerical simulations show that, compared with the benchmark VGG16, ResNet50 and EfficientNetB0, the f1 − score of the proposed method is improved by 1.6%, 2.8% and 3.7% respectively.

Terahertz spectroscopy combined with data dimensionality reduction algorithms for quantitative analysis of protein content in soybeans

Protein content in soybean is a key determinant of its nutritional and economic value. The paper investigated the feasibility of terahertz (THz) spectroscopy and dimensionality reduction algorithms for the determination of protein content in soybean. First of all, the THz sample spectrum was data processed by pre-processing or dimensionality reduction algorithms. Secondly, by calibration set, using partial least squares regression (PLSR), genetic algorithms-support vector regression (GA-SVR), grey wolf optimizer-support vector regression (GWO-SVR) and back propagation neural network (BPNN) were respectively used to model protein content determination. Afterwards, the model was validated by the prediction set. Ultimately, the BPNN model combined with linear discriminant analysis (LDA) for related coefficient of prediction set (Rp), root mean square error of prediction set (RMSEP), relative standard deviation (RSD), the time required for the operation was respectively 0.9677, 1.2467%, 3.3664%, and 53.51 s. The experimental results showed that the rapid and accurate quantitative determination of protein in soybean using THz spectroscopy is feasible after a suitable dimensionality reduction algorithm.