Search for: All records

Performance improvements obtained by recent principled approaches for pulse rate (PR) estimation from face videos have typically been achieved by adding or modifying certain modules within a reconfigurable system. Yet, evaluations of such remote photoplethysmography (rPPG) are usually performed only at the system level. To better understand each module's contribution and facilitate future research in explainable learning and artificial intelligence for physiological monitoring, this paper conducts a comparative study of video-based, principled PR tracking algorithms, with a focus on challenging fitness scenarios. A review of the progress achieved over the last decade and a half in this field is utilized to construct the major processing modules of a reconfigurable remote pulse rate sensing system. Experiments are conducted on two challenging datasets—an internal collection of 25 videos of two Asian males exercising on stationary-bike, elliptical, and treadmill machines, and 34 videos from a public ECG fitness database of 14 men and 3 women exercising on elliptical and stationary-bike machines. The signal-to-noise ratio (SNR), Pearson's correlation coefficient, error count ratio, error rate, and root mean squared error are used for performance evaluation. The top-performing configuration produces respective values of −0.8 dB, 0.86, 9%, 1.7%, and 3.3 beats per minute (bpm) for the internal dataset and 1.3 dB, 0.77, 28.6%, 6.0%, and 8.1 bpm for the ECG Fitness dataset, achieving significant improvements over alternative configurations. Our results suggest a synergistic effect between pulse color mapping and adaptive motion filtering, as well as the importance of a robust frequency tracking algorithm for PR estimation in low SNR settings. 

Paper surfaces can be used for anticounterfeiting due to their inherent and physically unclonable irregularities. Prior work used mobile cameras to capture paper's microstructure with the help of camera flash. However, prolonged exposure to flash in the workplace may harm the eyes of workers involved in the authentication process. This work proposes an authentication method that exploits indoor lighting without the need for a camera flash. Indoor lighting has a lower strength and leads to interference due to secondary reflections, making it challenging to achieve a good authentication performance. To this end, we create a digital twin (DT) replication of a real world in which paper patches are captured under multiple lights, taking account of key physics and optical laws. From simulations of DT, we identify important factors to the authentication performance and design an authentication method for an office setup. We have experimented with three different types of paper and showed that the DT-guided authentication method can achieve satisfactory authentication performance without using active light sources. 

This article presents a computational solution that enables continuous cardiac monitoring through cross-modality inference of electrocardiogram (ECG). While some smartwatches now allow users to obtain a 30-s ECG test by tapping a built-in bio-sensor, these short-term ECG tests often miss intermittent and asymptomatic abnormalities of cardiac functions. It is also infeasible to expect persistently active user participation for long-term continuous cardiac monitoring in order to capture these and other types of cardiac abnormalities. To alleviate the need for continuous user attention and active participation, we design a lightweight neural network that infers ECG from the photoplethysmogram (PPG) signal sensed at the skin surface by a wearable optical sensor. We also develop a diagnosis-oriented training strategy to enable the neural network to capture the pathological features of ECG, aiming to increase the utility of reconstructed ECG signals for screening cardiovascular diseases (CVDs). We also leverage model interpretation to obtain insights from data-driven models, for example, to reveal some associations between CVDs and ECG/PPG and to demonstrate how the neural network copes with motion artifacts in the ambulatory application. The experimental results on three datasets demonstrate the feasibility of inferring ECG from PPG, achieving a high fidelity of ECG reconstruction with only about 40000 parameters. 

Increases in the deployment of machine learning algorithms for applications that deal with sensitive data have brought attention to the issue of fairness in machine learning. Many works have been devoted to applications that require different demographic groups to be treated fairly. However, algorithms that aim to satisfy inter-group fairness (also called group fairness) may inadvertently treat individuals within the same demographic group unfairly. To address this issue, this article introduces a formal definition of within-group fairness that maintains fairness among individuals from within the same group. A pre-processing framework is proposed to meet both inter- and within-group fairness criteria with little compromise in performance. The framework maps the feature vectors of members from different groups to an inter-group fair canonical domain before feeding them into a scoring function. The mapping is constructed to preserve the relative relationship between the scores obtained from the unprocessed feature vectors of individuals from the same demographic group, guaranteeing within-group fairness. This framework has been applied to the Adult, COMPAS risk assessment, and Law School datasets, and its performance is demonstrated and compared with two regularization-based methods in achieving inter-group and within-group fairness. 

Quantization replaces floating point arithmetic with integer arithmetic in deep neural networks, enabling more efficient on-device inference with less power and memory. However, it also brings in loss of generalization and even potential errors to the models. In this work, we propose a parallelization technique for formally verifying the equivalence between quantized models and their original real-valued counterparts. In order to guarantee both soundness and completeness, mixed integer linear programming (MILP) is deployed as the baseline technique. Nevertheless, the incorporation of two networks as well as the mixture of integer and real number arithmetic make the problem much more challenging than verifying a single network, and thus using MILP alone is inadequate for the non-trivial cases. To tackle this, we design a distributed verification technique that can leverage hundreds of CPUs on high-performance computing clusters. We develop a two-tier parallel framework and propose property- and output-based partition strategies. Evaluated on perception networks quantized with PyTorch, our approach outperforms existing methods in successfully verifying many cases that are otherwise considered infeasible. 

The inverse problem of inferring clinical gold-standard electrocardiogram (ECG) from photoplethysmogram (PPG) that can be measured by affordable wearable Internet of Healthcare Things (IoHT) devices is a research direction receiving growing attention. It combines the easy measurability of PPG and the rich clinical knowledge of ECG for long-term continuous cardiac monitoring. The prior art for reconstruction using a universal basis, such as discrete cosine transform (DCT), has limited fidelity for uncommon ECG shapes due to the lack of representative power. To better utilize the data and improve data representation, we design two dictionary learning frameworks, the cross-domain joint dictionary learning (XDJDL), and the label-consistent XDJDL (LC-XDJDL), to further improve the ECG inference quality and enrich the PPG-based diagnosis knowledge. Building on the K-SVD technique, the proposed joint dictionary learning frameworks extend the expressive power by optimizing simultaneously a pair of signal dictionaries for PPG and ECG with the transforms to relate their sparse codes and disease information. The proposed models are evaluated with a variety of PPG and ECG morphologies from two benchmark datasets that cover various age groups and disease types. The results show the proposed frameworks achieve better inference performance than previous methods with average Pearson coefficients being 0.88 using XDJDL and 0.92 using LC-XDJDL, suggesting an encouraging potential for ECG screening using PPG based on the proactively learned PPG-ECG relationship. By enabling the dynamic monitoring and analysis of the health status of an individual, the proposed frameworks contribute to the emerging digital twins paradigm for personalized healthcare. 

The softmax function is a ubiquitous component at the output of neural networks and increasingly in intermediate layers as well. This paper provides convex lower bounds and concave upper bounds on the softmax function, which are compatible with convex optimization formulations for characterizing neural networks and other ML models. We derive bounds using both a natural exponential-reciprocal decomposition of the softmax as well as an alternative decomposition in terms of the log-sum-exp function. The new bounds are provably and/or numerically tighter than linear bounds obtained in previous work on robustness verification of transformers. As illustrations of the utility of the bounds, we apply them to verification of transformers as well as of the robustness of predictive uncertainty estimates of deep ensembles.