An Evaluation of Non-Contact Photoplethysmography-Based Methods for Remote Respiratory Rate Estimation.

The respiration rate (RR) is one of the physiological signals deserving monitoring for assessing human health and emotional states. However, traditional devices, such as the respiration belt to be worn around the chest, are not always a feasible solution (e.g., telemedicine, device discomfort). Recently, novel approaches have been proposed aiming at estimating RR in a less invasive yet reliable way, requiring the acquisition and processing of contact or remote Photoplethysmography (contact reference and remote-PPG, respectively). The aim of this paper is to address the lack of systematic evaluation of proposed methods on publicly available datasets, which currently impedes a fair comparison among them. In particular, we evaluate two prominent families of PPG processing methods estimating Respiratory Induced Variations (RIVs): the first encompasses methods based on the direct extraction of morphological features concerning the RR; and the second group includes methods modeling respiratory artifacts adopting, in the most promising cases, single-channel blind source separation. Extensive experiments have been carried out on the public BP4D+ dataset, showing that the morphological estimation of RIVs is more reliable than those produced by a single-channel blind source separation method (both in contact and remote testing phases), as well as in comparison with a representative state-of-the-art Deep Learning-based approach for remote respiratory information estimation.

Robust Single-Sample Face Recognition by Sparsity-Driven Sub-Dictionary Learning Using Deep Features.

Face recognition using a single reference image per subject is challenging, above all when referring to a large gallery of subjects. Furthermore, the problem hardness seriously increases when the images are acquired in unconstrained conditions. In this paper we address the challenging Single Sample Per Person (SSPP) problem considering large datasets of images acquired in the wild, thus possibly featuring illumination, pose, face expression, partial occlusions, and low-resolution hurdles. The proposed technique alternates a sparse dictionary learning technique based on the method of optimal direction and the iterative ℓ 0 -norm minimization algorithm called k-LiMapS. It works on robust deep-learned features, provided that the image variability is extended by standard augmentation techniques. Experiments show the effectiveness of our method against the hardness introduced above: first, we report extensive experiments on the unconstrained LFW dataset when referring to large galleries up to 1680 subjects; second, we present experiments on very low-resolution test images up to 8 × 8 pixels; third, tests on the AR dataset are analyzed against specific disguises such as partial occlusions, facial expressions, and illumination problems. In all the three scenarios our method outperforms the state-of-the-art approaches adopting similar configurations.

Sum signal dosimetry: A new approach for high dose quality assurance with Gafchromic EBT3.

Gafchromic EBT3 film dosimetry in radiosurgery (RS) and hypofractionated radiotherapy (HRT) is complicated by the limited film accuracy at high fractional doses. The aim of this study is to develop and evaluate sum signal (SS) film dosimetry to increase dose resolution at high fractional doses, thus allowing for use of EBT3 for dose distribution verification of RS/HRT treatments. To characterize EBT3 dose-response, a calibration was performed in the dose range 0.44-26.43 Gy. Red (RC) and green (GC) channel net optical densities were linearly added to produce the SS. Dose resolution and overall accuracy of the dosimetric protocol were estimated and compared for SS,RC, and GC. A homemade Matlab software was developed to compare, in terms of gamma analysis, dose distributions delivered by a Cyberknife on EBT3 films to dose distributions calculated by the treatment planning system. The new SS and conventional single channel (SC) methods were compared, using 3%/1 and 4%/1 mm acceptance criteria, for 20 patient plans. Our analysis shows that the SS dose-response curve is characterized by a steeper trend in comparison with SC, with SS providing a higher dose resolution in the whole dose range investigated. Gamma analysis confirms that the percentage of points satisfying the agreement criteria is significantly higher for SS compared to SC: 95.03% vs. 88.41% (P = 0.014) for 3%/1 mm acceptance criteria and 97.24% vs. 93.58% (P = 0.048) for 4%/1 mm acceptance criteria. This study demonstrates that the SS approach is a new and effective method to improve dosimetric accuracy in the framework of the RS-HRT patient-specific quality assurance protocol.

pyVHR: a Python framework for remote photoplethysmography.

Remote photoplethysmography (rPPG) aspires to automatically estimate heart rate (HR) variability from videos in realistic environments. A number of effective methods relying on data-driven, model-based and statistical approaches have emerged in the past two decades. They exhibit increasing ability to estimate the blood volume pulse (BVP) signal upon which BPMs (Beats per Minute) can be estimated. Furthermore, learning-based rPPG methods have been recently proposed. The present pyVHR framework represents a multi-stage pipeline covering the whole process for extracting and analyzing HR fluctuations. It is designed for both theoretical studies and practical applications in contexts where wearable sensors are inconvenient to use. Namely, pyVHR supports either the development, assessment and statistical analysis of novel rPPG methods, either traditional or learning-based, or simply the sound comparison of well-established methods on multiple datasets. It is built up on accelerated Python libraries for video and signal processing as well as equipped with parallel/accelerated ad-hoc procedures paving the way to online processing on a GPU. The whole accelerated process can be safely run in real-time for 30 fps HD videos with an average speedup of around 5. This paper is shaped in the form of a gentle tutorial presentation of the framework.

Orthogonal Procrustes Analysis for Dictionary Learning in Sparse Linear Representation.

In the sparse representation model, the design of overcomplete dictionaries plays a key role for the effectiveness and applicability in different domains. Recent research has produced several dictionary learning approaches, being proven that dictionaries learnt by data examples significantly outperform structured ones, e.g. wavelet transforms. In this context, learning consists in adapting the dictionary atoms to a set of training signals in order to promote a sparse representation that minimizes the reconstruction error. Finding the best fitting dictionary remains a very difficult task, leaving the question still open. A well-established heuristic method for tackling this problem is an iterative alternating scheme, adopted for instance in the well-known K-SVD algorithm. Essentially, it consists in repeating two stages; the former promotes sparse coding of the training set and the latter adapts the dictionary to reduce the error. In this paper we present R-SVD, a new method that, while maintaining the alternating scheme, adopts the Orthogonal Procrustes analysis to update the dictionary atoms suitably arranged into groups. Comparative experiments on synthetic data prove the effectiveness of R-SVD with respect to well known dictionary learning algorithms such as K-SVD, ILS-DLA and the online method OSDL. Moreover, experiments on natural data such as ECG compression, EEG sparse representation, and image modeling confirm R-SVD's robustness and wide applicability.

Improving automatic face recognition with user interaction.

Face recognition systems aim to recognize the identity of a person depicted in a photograph by comparing it against a gallery of prerecorded images. Current systems perform quite well in controlled scenarios, but they allow for none or little interaction in case of mistakes due to the low quality of images or to algorithmic limitations. Following the needs and suggestions of investigators, we present a guided user interface that allows to adjust from a fully automatic to a fully assisted modality of execution, according to the difficulty of the task and to amount of available information (gender, age, etc.): the user can generally rely on automatic execution and intervene only on a limited number of examples when a failure is automatically detected or when the quality of intermediate results is deemed unsatisfactory. The interface runs on top of a preexistent automatic face recognition algorithm in such a way to guarantee full control over the execution flow and to exploit the peculiarities of the underlying image processing techniques. The viability of the proposed solution is tested on a classic face identification task run on a standard publicly available database (the XM2VTS), assessing the improvement to user interaction over the automatic system performance.