Data Transparency and Fairness Analysis of the NYPD Stop-and-Frisk Program

Given the increased concern of racial disparities in the stop-and-frisk programs, the New York Police Department ( NYPD ) requires publicly displaying detailed data for all the stops conducted by police authorities, including the suspected offense and race of the suspects. By adopting a public data transparency policy, it becomes possible to investigate racial biases in stop-and-frisk data and demonstrate the benefit of data transparency to approve or disapprove social beliefs and police practices. Thus, data transparency becomes a crucial need in the era of Artificial Intelligence ( AI ), where police and justice increasingly use different AI techniques not only to understand police practices but also to predict recidivism, crimes, and terrorism. In this study, we develop a predictive analytics method, including bias metrics and bias mitigation techniques to analyze the NYPD Stop-and-Frisk datasets and discover whether underline bias patterns are responsible for stops and arrests. In addition, we perform a fairness analysis on two protected attributes, namely, the race and the gender, and investigate their impacts on arrest decisions. We also apply bias mitigation techniques. The experimental results show that the NYPD Stop-and-Frisk dataset is not biased toward colored and Hispanic individuals and thus law enforcement authorities can apply the bias predictive analytics method to inculcate more fair decisions before making any arrests.

Time-to-event prediction using survival analysis methods for Alzheimer's disease progression

INTRODUCTION

Many research studies have well investigated Alzheimer's disease (AD) detection and progression. However, the continuous-time survival prediction of AD is not yet fully explored to support medical practitioners with predictive analytics. In this study, we develop a survival analysis approach to examine interactions between patients' inherent temporal and medical patterns and predict the probability of the AD next stage progression during a time period. The likelihood of reaching the following AD stage is unique to a patient, helping the medical practitioner analyze the patient's condition and provide personalized treatment recommendations ahead of time.

METHODOLOGIES

We simulate the disease progression based on patient profiles using non-linear survival methods-non-linear Cox proportional hazard model (Cox-PH) and neural multi-task logistic regression (N-MTLR). In addition, we evaluate the concordance index (C-index) and Integrated Brier Score (IBS) to describe the evolution to the next stage of AD. For personalized forecasting of disease, we also developed deep neural network models using the dataset provided by the National Alzheimer's Coordinating Center with their multiple-visit details between 2005 and 2017.

RESULTS

The experiment results show that our N-MTLR based survival models outperform the CoxPH models, the best of which gives Concordance-Index of 0.79 and IBS of 0.09. We obtained 50 critical features out of 92 by applying recursive feature elimination and random forest techniques on the clinical data; the top ones include normal cognition and behavior, criteria for dementia, community affairs, etc. Our study demonstrates that selecting critical features can improve the effectiveness of probabilities at each time interval.

CONCLUSIONS

The proposed deep learning-based survival method and model can be used by medical practitioners to predict the patients' AD shift efficiently and recommend personalized treatment to mitigate or postpone the effects of AD. More generally, our proposed survival analysis approach for predicting disease stage shift can be used for other progressive diseases such as cancer, Huntington's disease, and scleroderma, just to mention a few, using the corresponding clinical data.

Analysis of highly efficient quad-crescent-shaped Si nanowires solar cell

Nanostructured semiconductor nanowires (NWs) present a smart solution for broadband absorption solar cells (SCs) with high efficiency and low-cost. In this paper, a novel design of quad crescent-shaped silicon NW SC is introduced and numerically studied. The suggested NW has quad crescent shapes which create a cavity between any adjacent NWs. Such a cavity will permit multiple light scattering with improved absorption. Additionally, new modes will be excited along the NWs, which are highly coupled with the incident light. Further, the surface reflection from the crescent NWs is decreased due to the reduced surface filling ratio. The finite difference time domain method is utilized to analyze the optical characteristics of the reported structure. The proposed NW offers short circuit current density (Jsc) of 27.8 mA/cm² and ultimate efficiency (η_ul) of 34% with an enhancement of 14% and volume reduction of 40% compared to the conventional NWs. The J_sc and η_ul are improved to 35.8 mA/cm² and 43.7% by adding a Si substrate and back reflector to the suggested crescent NWs.

Lung cancer survival period prediction and understanding: Deep learning approaches

INTRODUCTION

Survival period prediction through early diagnosis of cancer has many benefits. It allows both patients and caregivers to plan resources, time and intensity of care to provide the best possible treatment path for the patients. In this paper, by focusing on lung cancer patients, we build several survival prediction models using deep learning techniques to tackle both cancer survival classification and regression problems. We also conduct feature importance analysis to understand how lung cancer patients' relevant factors impact their survival periods. We contribute to identifying an approach to estimate survivability that are commonly and practically appropriate for medical use.

METHODOLOGIES

We have compared the performance across three of the most popular deep learning architectures - Artificial Neural Networks (ANN), Convolutional Neural Networks (CNN), and Recurrent Neural Networks (RNN) while comparing the performing of deep learning models against traditional machine learning models. The data was obtained from the lung cancer section of Surveillance, Epidemiology, and End Results (SEER) cancer registry.

RESULTS

The deep learning models outperformed traditional machine learning models across both classification and regression approaches. We obtained a best of 71.18 % accuracy for the classification approach when patients' survival periods are segmented into classes of '<=6 months',' 0.5 - 2 years' and '>2 years' and Root Mean Squared Error (RMSE) of 13.5 % andR² value of 0.5 for the regression approach for the deep learning models while the traditional machine learning models saturated at 61.12 % classification accuracy and 14.87 % RMSE in regression.

CONCLUSIONS

This approach can be a baseline for early prediction with predictions that can be further improved with more temporal treatment information collected from treated patients. In addition, we evaluated the feature importance to investigate the model interpretability, gaining further insight into the survival analysis models and the factors that are important in cancer survival period prediction.

Identifying Mouse Autoimmune Uveitis from Fundus Photographs Using Deep Learning

Purpose

To develop a deep learning model for objective evaluation of experimental autoimmune uveitis (EAU), the animal model of posterior uveitis that reveals its essential pathological features via fundus photographs.

Methods

We developed a deep learning construct to identify uveitis using reference mouse fundus images and further categorized the severity levels of disease into mild and severe EAU. We evaluated the performance of the model using the area under the receiver operating characteristic curve (AUC) and confusion matrices. We further assessed the clinical relevance of the model by visualizing the principal components of features at different layers and through the use of gradient-weighted class activation maps, which presented retinal regions having the most significant influence on the model.

Results

Our model was trained, validated, and tested on 1500 fundus images (training, 1200; validation, 150; testing, 150) and achieved an average AUC of 0.98 for identifying the normal, trace (small and local lesions), and disease classes (large and spreading lesions). The AUCs of the model using an independent subset with 180 images were 1.00 (95% confidence interval [CI], 0.99-1.00), 0.97 (95% CI, 0.94-0.99), and 0.96 (95% CI, 0.90-1.00) for the normal, trace and disease classes, respectively.

Conclusions

The proposed deep learning model is able to identify three severity levels of EAU with high accuracy. The model also achieved high accuracy on independent validation subsets, reflecting a substantial degree of generalizability.

Translational Relevance

The proposed model represents an important new tool for use in animal medical research and provides a step toward clinical uveitis identification in clinical practice.

Evaluation of a real-time low-power cardiorespiratory sensor for the IoT

A wide variety of sensors have been developed in the biomedical engineering community for telemedicine and personalized healthcare applications. However, they usually focus on sensor connectivity and embedded signal processing, at the expense of the sensing part. This observation lead to the development and exhaustive evaluation of a new ECGbased cardiorespiratory IoT sensor. In order to improve the robustness of our IoT-based sensor, we discuss in detail the influence of electrodes placement and nature. Performance assessment of our sensor resulted in a best-case sensitivity of 99.95% and a precision of 99.89% for an abdominal positioning of wet electrodes, while a sensitivity of 99.47% and a precision of 99.31% were observed using a commercialgrade dry electrodes belt. Consequently, we prove that our sensor is fit for the comfortable medical-grade monitoring of the cardiorespiratory activity in order to provide insights of patients health in a telemedicine context.

Comparative Neuropathogenesis of Equine Herpesvirus 9 and its Mutant Clone (SP21) Inoculated Intranasally in a Hamster Model

The neuropathogenesis of equine herpesvirus 9 (EHV-9), a neurotropic herpesvirus, and its mutant clone (SP21) was studied experimentally in a hamster model. EHV-9-infected hamsters showed clinical signs of infection at 3 days post infection (dpi), while infection with SP21 resulted in clinical signs at 4 dpi. Clinical signs were more severe in the EHV-9-infected group than in the SP21-infected group. There was a significant difference in the time of anterograde transmission of EHV-9 and SP21 inside the brain. Viraemia was detected in the EHV-9-infected group at 4-5 dpi, while no viraemia was detected in the SP21-infected group. The serum concentration of tumour necrosis factor-α was significantly higher in EHV-9-infected animals than in those infected by SP21 group at 4-5 dpi, but there was no difference in the serum concentration of interferon-γ. The spatiotemporal profiles of viral replication and virus-associated histopathology were remarkably similar, were high in the olfactory bulb and cerebral hemispheres, and decreased progressively towards the medulla oblongata. The mean group scores of the histopathological changes for the entire brain were significantly higher in the EHV-9 group than in the SP21 group at all time points, starting from 3 dpi. These results suggest that the gene products of the open reading frame (ORF)19 and ORF14 play essential roles in the neuropathogenesis of EHV-9, as the two point-mutations detected in SP21 significantly altered the neuropathogenesis of the virus.

Effect of annealing on La0.8Sr0.2MnO3 thin films prepared by pulsed laser deposition

La _0.8 Sr_0.2MnO₃ (LSM) polycrystalline powder was synthesized using hydrothermal method at 900 °C. High Resolution Transmission Electron Microscope (HR-TEM) showed that the particles were uniform with average particle size of 657 nm. X-ray Diffraction (XRD) and lattice fringes indicated rhombohedral structure of LSM particles. Thin LSM films were successfully grown on cleaned Si (100) substrate by pulsed laser deposition (PLD). Annealing of LSM films in air affected structure, morphology and electrical properties that films where crystallization of the LSM films was started at 600 °C and enhanced by further annealing as indicated by XRD. Field Emission Scanning Electron Microscope (FESEM) revealed that the grain size increases by increasing annealing temperature and at temperature of 1000 °C cracks were seen. Average roughness and root mean square roughness decreased with further annealing (up to 800 °C) then increased at 1000 °C that were verified by atomic force microscope (AFM). Moreover, Raman scattering was enhanced and all major bands were revealed at 800 °C. Resistivity of LSM films decreases with increasing temperature (from 25 °C to 200 °C) and all films underwent a semiconductor behavior in the most of applied temperatures. The lowest resistivity of LSM films was reached at annealing temperature of 800 °C with low activation energy value (E_a) of about 0.1 eV.

Identity Management Systems for the Internet of Things: A Survey Towards Blockchain Solutions

The Internet of Things aims at connecting everything, ranging from individuals, organizations, and companies to things in the physical and virtual world. The digital identity has always been considered as the keystone for all online services and the foundation for building security mechanisms such as authentication and authorization. However, the current literature still lacks a comprehensive study on the digital identity management for the Internet of Things (IoT). In this paper, we firstly identify the requirements of building identity management systems for IoT, which comprises scalability, interoperability, mobility, security and privacy. Then, we trace the identity problem back to the origin in philosophy, analyze the Internet digital identity management solutions in the context of IoT and investigate recent surging blockchain sovereign identity solutions. Finally, we point out the promising future research trends in building IoT identity management systems and elaborate challenges of building a complete identity management system for the IoT, including access control, privacy preserving, trust and performance respectively.

Hybrid Group Recommendation Using Modified Termite Colony Algorithm: A Context Towards Big Data

Since the introduction of Web 2.0, group recommendation systems become an effective tool for consulting and recommending items according to the choices of group of likeminded users. However, the population of dataset consisting of the large number of choices increases the size of storage. As a result, identification of the combination for specific recommendation becomes complex. Hence, the existing group recommendation system should support methodology for handling large data volume with varsity. In this paper, we propose a content-boosted modified termite colony optimisation-based rating prediction algorithm (CMTRP) for group recommendation system. CMTRP employs a hybrid recommendation framework with respect to the big data paradigm to deal with the trend of large data. The framework utilises the communal ratings that help to overcome the scalability problem. The experimental results reveal that CMTRP provides less error in the rating prediction and higher recommendation precision compared with the existing algorithms.

Analysis of the pen pressure and grip force signal during basic drawing tasks: The timing and speed changes impact drawing characteristics

Writing is a complex fine and trained motor skill, involving complex biomechanical and cognitive processes. In this paper, we propose the study of writing kinetics using three angles: the pen-tip normal force, the total grip force signal and eventually writing quality assessment. In order to collect writing kinetics data, we designed a sensor collecting these characteristics simultaneously. Ten healthy right-handed adults were recruited and were asked to perform four tasks: first, they were instructed to draw circles at a speed they considered comfortable; they then were instructed to draw circles at a speed they regarded as fast; afterwards, they repeated the comfortable task compelled to follow the rhythm of a metronome; and eventually they performed the fast task under the same timing constraints. Statistical differences between the tasks were computed, and while pen-tip normal force and total grip force signal were not impacted by the changes introduced in each task, writing quality features were affected by both the speed changes and timing constraint changes. This verifies the already-studied speed-accuracy trade-off and suggest the existence of a timing constraints-accuracy trade-off.

Collaborative Business and Enterprise Urbanization

Information system urbanization consists of studying the various functional sectors of an enterprise in order to carry out cartographical surveys, then, similarly study its information system (Sassoon, 1998). However, the studied information system fails to respond to on-demand production and dynamic collaboration. In fact, a company must adapt to market changes by production structures and the organization of its processes and not the flow of its information alone. In this article, a novel urbanization approach conjointly ensures the evolution of the production and the information systems by providing an incremental, process-oriented strategy considering horizontal and vertical organizations of firms.

Photocatalytic degradation of methyl red dye by silica nanoparticles

Silica nanoparticles (SiO2 NPs) were found to be photocatalytically active for degradation of methyl red dye (MR). The SiO2 NPs and SiO2 NPs doped with silver (and or) gold nanoparticles were prepared. From the transmission electron microscopy (TEM) images the particle size and particle morphology of catalysts were monitored. Moreover, SiO2 NPs doped with silver and gold ions were used as a photocatalyst for degradation of MR. The rate of photocatalytic degradation of MR was found to be increased in the order of SiO2 NPs, SiO2 NPs coated with gold nanoparticles (Au NPs) and silver nanoparticles (Ag NPs), SiO2 NPs coated with Ag NPs, SiO2 NPs coated with Au NPs, Ag+-doped SiO2 NPs, and Au3+-doped SiO2 NPs. The kinetic and mechanism of photocatalytic reaction were studied and accorded well with experimental results.

Biosynthesis of gold nanoparticles using Pseudomonas aeruginosa

Pseudomonas aeruginosa were used for extra-cellular biosynthesis of gold nanoparticles (Au NPs). Consequently, Au NPs were formed due to reduction of gold ion by bacterial cell supernatant of P. aeruginosa ATCC 90271, P. aeruginosa (2) and P. aeruginosa (1). The UV-vis and fluorescence spectra of the bacterial as well as chemical prepared Au NPs were recorded. Transmission electron microscopy (TEM) micrograph showed the formation of well-dispersed gold nanoparticles in the range of 15-30 nm. The process of reduction being extra-cellular and may lead to the development of an easy bioprocess for synthesis of Au NPs.

Effect of PVA surrounding medium on ZnSe nanoparticles: size, optical, and electrical properties

Polyvinyl alcohol (PVA) matrix was used to confine the particle size of ZnSe nanocrystallites as well as the variation of zinc (Zn) to selenium (Se) ion ratio which showed a remarkable decrease on the particle size as this ratio increased. The particle size decrease was monitored from the UV-vis absorption measurement as well as photoluminescence which suffered a blue shift with particle size decrease. The particle size was characterized with the aid of X-ray diffraction (XRD). The Raman spectra showed that, as the particle size increases, the peak position of the line centers (LO) mode were found to be red shifted from 239 to 234 cm(-1), accompanied by an increase in the full-width at half-maximum (FWHM). The electrical measurements and FT-IR spectra (overtone and normal) band vibration were used to study the effect of ZnSe NPs size on the PVA matrix.

Effect of silver nanowires on the surface-enhanced Raman spectra (SERS) of the RNA bases

RNA bases have a great importance in the biological and genetics applications. The surface-enhanced Raman scattering (SERS) was used to study the orientation and adsorption structure of RNA bases adsorbed on the surface of silver nanowires (Ag NWs). The Ag NWs were prepared and its UV-vis spectra were recorded. The RNA bases oriented perpendicularly on the surface of Ag NWs, as the coverage area decreases. Consequently, the in-plane bands were enhanced according to the SERS selection rule. Many bands were red shifted due to the chemisorption of RNA bases on the Ag NWs surface. New bands corresponding to the base-surface bond were appeared in the SERS spectra.