Multichannel Hierarchical Analysis of Time-Resolved Hyperspectral Data for Advanced Colorimetric E-Nose.

The colorimetric sensor-based electronic nose has been demonstrated to discriminate specific gaseous molecules for various applications, including health or environmental monitoring. However, conventional colorimetric sensor systems rely on RGB sensors, which cannot capture the complete spectral response of the system. This limitation can degrade the performance of machine learning analysis, leading to inaccurate identification of chemicals with similar functional groups. Here, we propose a novel time-resolved hyperspectral (TRH) data set from colorimetric array sensors consisting of 1D spatial, 1D spectral, and 1D temporal axes, which enables hierarchical analysis of multichannel 2D spectrograms via a convolution neural network (CNN). We assessed the outstanding classification performance of the TRH data set compared to an RGB data set by conducting a relative humidity (RH) concentration classification. The time-dependent spectral response of the colorimetric sensor was measured and trained as a CNN model using TRH and RGB sensor systems at different RH levels. While the TRH model shows a high classification accuracy of 97.5% for the RH concentration, the RGB model yields 72.5% under identical conditions. Furthermore, we demonstrated the detection of various functional volatile gases with the TRH system by using experimental and simulation approaches. The results reveal distinct spectral features from the TRH system, corresponding to changes in the concentration of each substance.

3D Superclusters with Hybrid Bioinks for Early Detection in Breast Cancer.

The surface-enhanced Raman scattering (SERS) technique has garnered significant interest due to its ultrahigh sensitivity, making it suitable for addressing the growing demand for disease diagnosis. In addition to its sensitivity and uniformity, an ideal SERS platform should possess characteristics such as simplicity in manufacturing and low analyte consumption, enabling practical applications in complex diagnoses including cancer. Furthermore, the integration of machine learning algorithms with SERS can enhance the practical usability of sensing devices by effectively classifying the subtle vibrational fingerprints produced by molecules such as those found in human blood. In this study, we demonstrate an approach for early detection of breast cancer using a bottom-up strategy to construct a flexible and simple three-dimensional (3D) plasmonic cluster SERS platform integrated with a deep learning algorithm. With these advantages of the 3D plasmonic cluster, we demonstrate that the 3D plasmonic cluster (3D-PC) exhibits a significantly enhanced Raman intensity through detection limit down to 10-6 M (femtomole-(10-17 mol)) for p-nitrophenol (PNP) molecules. Afterward, the plasma of cancer subjects and healthy subjects was used to fabricate the bioink to build 3D-PC structures. The collected SERS successfully classified into two clusters of cancer subjects and healthy subjects with high accuracy of up to 93%. These results highlight the potential of the 3D plasmonic cluster SERS platform for early breast cancer detection and open promising avenues for future research in this field.

Non-intrusive quality appraisal of differentiation-induced cardiovascular stem cells using E-Nose sensor technology.

Stem cell technology holds immense potential for revolutionizing medicine, particularly in regenerative treatment for heart disease. The unique capacity of stem cells to differentiate into diverse cell types offers promise in repairing damaged tissues and implanting organs. Ensuring the quality of differentiated cells, essential for specific functions, demands in-depth analysis. However, this process consumes time and incurs substantial costs while invasive methods may alter stem cell features during differentiation and deplete cell numbers. To address these challenges, we propose a non-invasive strategy, using cellular respiration, to assess the quality of differentiation-induced stem cells, notably cardiovascular stem cells. This evaluation employs an electronic nose (E-Nose) and neural pattern separation (NPS). Our goal is to assess differentiation-induced cardiac stem cells (DICs) quality through E-Nose data analysis and compare it with standard commercial human cells (SCHCs). Sensitivity and specificity were evaluated by interacting SCHCs and DICs with the E-Nose, achieving over 90% classification accuracy. Employing selective combinations optimized by NPS, E-Nose successfully classified all six cell types. Consequently, the relative similarity among DICs like cardiomyocytes, endothelial cells with SCHCs was established relied on comparing response data from the E-Nose sensor without resorting to complex evaluations.