Information fusion is an essential part of numerous engineering systems and biological functions, e.g., human cognition. Fusion occurs at many levels, ranging from the low-level combination of signals to the high-level aggregation of heterogeneous decision-making processes. While the last decade has witnessed an explosion of research in deep learning, fusion in neural networks has not observed the same revolution. Specifically, most neural fusion approaches are ad hoc, are not understood, are distributed versus localized, and/or explainability is low (if present at all). Herein, we prove that the fuzzy Choquet integral (ChI), a powerful nonlinear aggregation function, can be represented as a multi-layer network, referred to hereafter as ChIMP. We also put forth an improved ChIMP (iChIMP) that leads to a stochastic gradient descent-based optimization in light of the exponential number of ChI inequality constraints. An additional benefit of ChIMP/iChIMP is that it enables eXplainable AI (XAI). Synthetic validation experiments are provided and iChIMP is applied to the fusion of a set of heterogeneous architecture deep models in remote sensing. We show an improvement in model accuracy and our previously established XAI indices shed light on the quality of our data, model, and its decisions.

To develop quantitative methods for the clinical interpretation of the ballistocardiogram (BCG). A closed-loop mathematical model of the cardiovascular system is proposed to theoretically simulate the mechanisms generating the BCG signal, which is then compared with the signal acquired via accelerometry on a suspended bed. Simulated arterial pressure waveforms and ventricular functions are in good qualitative and quantitative agreement with those reported in the clinical literature. Simulated BCG signals exhibit the typical I, J, K, L, M, and N peaks and show good qualitative and quantitative agreement with experimental measurements. Simulated BCG signals associated with reduced contractility and increased stiffness of the left ventricle exhibit different changes that are characteristic of the specific pathological condition. The proposed closed-loop model captures the predominant features of BCG signals and can predict pathological changes on the basis of fundamental mechanisms in cardiovascular physiology. This paper provides a quantitative framework for the clinical interpretation of BCG signals and the optimization of BCG sensing devices. The present paper considers an average human body and can potentially be extended to include variability among individuals.

We propose a non-wearable hydraulic bed sensor system that is placed underneath the mattress to estimate the relative systolic blood pressure of a subject, which only differs from the actual blood pressure by a scaling and an offset factor. Two types of features are proposed to obtain the relative blood pressure, one based on the strength and the other on the morphology of the bed sensor BCG pulses. The relative blood pressure is related to the actual by a scale and an offset factor that can be obtained through calibration. The proposed system is able to extract the relative blood pressure more accurately with a less sophisticated sensor system compared to those from the literature. We tested the system using a dataset collected from 48 subjects right after active exercises. Comparison with the ground truth obtained from the blood pressure cuff validates the promising performance of the proposed system, where the mean correlation between the estimate and the ground truth is near 90% for the strength feature and 83% for the morphology feature.

A novel approach is proposed for combining multiple anomaly algorithm decisions with image space cell-structured features in a long wave infrared (LWIR) system in the context of forward looking (FL) buried explosive hazard detection along a road. A pre-screener is applied first, which is an ensemble of trainable size-contrast filters with mean shift clustering in Universal Transverse Mercator (UTM) space. Next, features from image chips representing anomaly decisions from different algorithms are extracted from UTM confidence maps based on maximally stable extremal regions (MSERs) and Gaussian mixture models (GMMs). Pre-screener hits in UTM space are back-projected into the video at multiple standoff distances and cell-structured local binary patterns (LBPs), histogram of gradients (HOGs) and mean-variance descriptors are extracted. Experiments are conducted using buried materials with varying metal contents and depths at a U.S. Army test site. Results are extremely encouraging for FL imaging and show a significant decrease in the number of false alarms (FAs). Targets not currently detected by our system are also not detected by a human under manual visual inspection.

This paper presents a new technique for clustering either object or relational data. First, the data are represented as a matrix D of dissimilarity values. D is reordered to D* using a visual assessment of cluster tendency algorithm. If the data contain clusters, they are suggested by visually apparent dark squares arrayed along the main diagonal of an image I(D*) of D*. The suggested clusters in the object set underlying the reordered relational data are found by defining an objective function that recognizes this blocky structure in the reordered data. The objective function is optimized when the boundaries in I(D*) are matched by those in an aligned partition of the objects. The objective function combines measures of contrast and edginess and is optimized by particle swarm optimization. We prove that the set of aligned partitions is exponentially smaller than the set of partitions that needs to be searched if clusters are sought in D. Six numerical examples are given to illustrate various facets of the algorithm.