Interpretable machine learning with an eye for the physics: Hyperspectral Vis/NIR “video” of drying wood analyzed by hybrid subspace modeling

Chemometric multivariate analysis based on low-dimensional linear and bilinear data modelling is presented as a fast and interpretable alternative to more fancy “AI” for practical use of Big Data streams from hyperspectral “video” cameras. The purpose of the present illustration is to find, quantify and understand the various known and unknown factors affecting the process of drying moist wood. It involves an “interpretable machine learning” that analyses more than 350 million absorbance spectra, requiring 418 GB of data storage, without the use of black box operations. The 159-channel high-resolution hyperspectral wood “video” in the 500–1005 nm range was reduced to five known and four unknown variation components of physical and chemical nature, each with its spectral, spatial and temporal parameters quantified. Together, this 9-dimensional linear model explained more than 99.98% of the total input variance.

Enhancing the Throughput of FT Mass Spectrometry Imaging Using Compressed Sensing and Subspace Modeling

Mass spectrometry imaging (MSI) allows for untargeted mapping of the chemical compositions of tissues with attomole detection limits. MSI using Fourier transform-based mass spectrometers, such as FT-ion cyclotron resonance (FT-ICR), grants the ability to examine the chemical space with unmatched mass resolution and mass accuracy. However, direct imaging of large tissue samples on FT-ICR is restrictively slow. In this work, we present an approach that combines the subspace modeling of ICR temporal signals with compressed sensing to accelerate high-resolution FT-ICR MSI. A joint subspace and sparsity constrained reconstruction enables the creation of high-resolution imaging data from the sparsely sampled and short-time acquired transients. Simulation studies and experimental implementation of the proposed acquisition in investigation of brain tissues demonstrate a factor of 10 enhancement in throughput of FT-ICR MSI, without the need for instrumental or hardware modifications.

Visual Semantic-Based Representation Learning Using Deep CNNs for Scene Recognition

In this work, we address the task of scene recognition from image data. A scene is a spatially correlated arrangement of various visual semantic contents also known as concepts, e.g., “chair,” “car,” “sky,” etc. Representation learning using visual semantic content can be regarded as one of the most trivial ideas as it mimics the human behavior of perceiving visual information. Semantic multinomial (SMN) representation is one such representation that captures semantic information using posterior probabilities of concepts. The core part of obtaining SMN representation is the building of concept models. Therefore, it is necessary to have ground-truth (true) concept labels for every concept present in an image. Moreover, manual labeling of concepts is practically not feasible due to the large number of images in the dataset. To address this issue, we propose an approach for generating pseudo-concepts in the absence of true concept labels. We utilize the pre-trained deep CNN-based architectures where activation maps (filter responses) from convolutional layers are considered as initial cues to the pseudo-concepts. The non-significant activation maps are removed using the proposed filter-specific threshold-based approach that leads to the removal of non-prominent concepts from data. Further, we propose a grouping mechanism to group the same pseudo-concepts using subspace modeling of filter responses to achieve a non-redundant representation. Experimental studies show that generated SMN representation using pseudo-concepts achieves comparable results for scene recognition tasks on standard datasets like MIT-67 and SUN-397 even in the absence of true concept labels.