A deep learning approach for the forensic evaluation of sexual assault

Despite the existence of patterns able to discriminate between consensual and non-consensual intercourse, the relevance of genital lesions in the corroboration of a legal rape complaint is currently under debate in many countries. The testimony of the physicians when assessing these lesions has been questioned in court due to several factors (e.g., a lack of comprehensive knowledge of lesions, wide spectrum of background area, among others). Therefore, it is relevant to provide automated tools to support the decision process in an objective manner. In this work, we evaluate the performance of state-of-the-art deep learning architectures for the forensic assessment of sexual assault. We propose a deep architecture and learning strategy to tackle the class imbalance on deep learning using ranking. The proposed methodologies achieved the best results when compared with handcrafted feature engineering and with other deep architectures.     

Circular Economy in Tailings Management

Mine Water and the Environment - Tailings size distribution can be modified to greatly reduce surface storage area and the risk of catastrophic tailings dam failures. Total flotation and backfill...     

Jerk analysis of a six-degrees-of-freedom three-legged parallel manipulator

In this work the jerk analysis of a 3-RRPS parallel manipulator to realize six degrees of freedom is approached by means of the theory of screws. The input/output equations of velocity, acceleration and jerk of the moving platform with respect to the fixed platform are obtained systematically by resorting to reciprocal-screw theory. A numerical example is included in order to show the application of the method of kinematic analysis. Furthermore, the numerical results obtained via screw theory are satisfactorily compared with simulations generated with the aid of commercially available software.     

Comprehensive characterization of the functional activities of pressurized liquid and ultrasound-assisted extracts from Chlorella vulgaris

Chlorella vulgaris has been referred as a potential source of bioactive compounds (carotenoids and fatty acids). In this work, the ability of an environmentally friendly extraction technique such as Pressurized Liquid Extraction (PLE) and a traditional extraction technique such as Ultrasound-Assisted Extraction (UAE) to obtain functional compounds from C. vulgaris, at analytical scale, has been demonstrated. Seeked bioactivities were antioxidant and antimicrobial, for their interest in the food industry. Therefore, a methodology including analytical extraction, in-vitro assays and chemical characterization via HPLC-DAD and GC–MS has been used to determine the interest of Chlorella as a source of functional food ingredients. Results demonstrated that PLE provide higher yields than UAE while similar bioactivities were obtained. Important concentration of carotenoids (lutein, neoxanthin, β-carotene, etc.), chlorophylls, sterols, phytols, and fatty acids (among others) have been found in Chlorella extracts that could be correlated to the observed biological activity.     

Rejoinder to “Multiple versions of the h-index: cautionary use for formal academic purposes”

Scientometrics - In a Letter to the Editor, the authors of this paper show, to highlight the practical risks of using the h-index, how academics’ data and bibliometric information can be...     

EETP-MAC: energy efficient traffic prioritization for medium access control in wireless body area networks

Wireless body area network (WBAN) has witnessed significant attentions in the healthcare domain using biomedical sensor-based monitoring of heterogeneous nature of vital signs of a patient’s body. The design of frequency band, MAC superframe structure, and slots allocation to the heterogeneous nature of the patient’s packets have become the challenging problems in WBAN due to the diverse QoS requirements. In this context, this paper proposes an Energy Efficient Traffic Prioritization for Medium Access Control (EETP-MAC) protocol, which provides sufficient slots with higher bandwidth and guard bands to avoid channels interference causing longer delay. Specifically, the design of EETP-MAC is broadly divided in to four folds. Firstly, patient data traffic prioritization is presented with broad categorization including Non-Constrained Data (NCD), Delay-Constrained Data (DCD), Reliability-Constrained Data (RCD) and Critical Data (CD). Secondly, a modified superframe structure design is proposed for effectively handling the traffic prioritization. Thirdly, threshold based slot allocation technique is developed to reduce contention by effectively quantifying criticality on patient data. Forth, an energy efficient frame design is presented focusing on beacon interval, superframe duration, and packet size and inactive period. Simulations are performed to comparatively evaluate the performance of the proposed EETP-MAC with the state-of-the-art MAC protocols. The comparative evaluation attests the benefit of EETP-MAC in terms of efficient slot allocation resulting in lower delay and energy consumption.

     

A simple approach to solving the kinematics of the 4-UPS/PS (3R1T) parallel manipulator

Journal of Mechanical Science and Technology - This work reports on the position, velocity and acceleration analyses of a four-degrees-of-freedom parallel manipulator, 4-DoF-PM for brevity, which...     

Comparison of Peleg and Azuara et al. models in the modeling mass transfer during pile salting of goat sheets

Application of the Peleg and Azuara et al. models for describing mass transfer during pile salting of goat sheets using different mixtures of NaCl, KCl, CaCl₂ and MgCl₂ was investigated The high coefficient of determination (R² > 0.97 and R² > 0.95 for Peleg and Azuara et al. models, respectively) and the modulus of mean relative errors (MRE < 10% for both models) indicated the acceptability of both Peleg and Azuara et al. models for predicting both moisture loss and salt uptake. The equilibrium moisture and salt contents were estimated using Peleg rate constants and Azuara et al. model parameters. Equilibrium moisture (x_w) and salt contents (x_s) varied from 0.522 to 0.860 g water/g db and from 0.311 to 0.352 g NaCl/g db respectively. Multiple comparison of means showed that x_w and x_s predicted by the Peleg model were lower than those estimated by the Azuara et al. model. The values differed by 1.86–13.6% for x_w and by 1.13–10.57% for x_s.     

Occupational therapy for people with physical disability using interactive environments

Universal Access in the Information Society - This paper preconizes the use of interactive environments as an accessible and inexpensive solution that allows supporting the recovery process of...