An Improved YOLOv5 Algorithm for Vulnerable Road User Detection

The vulnerable road users (VRUs), being small and exhibiting random movements, increase the difficulty of object detection of the autonomous emergency braking system for vulnerable road users AEBS-VRUs, with their behaviors highly random. To overcome existing problems of AEBS-VRU object detection, an enhanced YOLOv5 algorithm is proposed. While the Complete Intersection over Union-Loss (CIoU-Loss) and Distance Intersection over Union-Non-Maximum Suppression (DIoU-NMS) are fused to improve the model's convergent speed, the algorithm also incorporates a minor object detection layer to increase the performance of VRU detection. A dataset for complex AEBS-VRUS scenarios is established based on existing datasets such as Caltech, nuScenes, and Penn-Fudan, and the model is trained using migration learning based on the PyTorch framework. A number of comparative experiments using models such as YOLOv6, YOLOv7, YOLOv8 and YOLOx are carried out. The results of the comparative evaluation show that the proposed improved YOLO5 algorithm has the best overall performance in terms of efficiency, accuracy and timeliness of target detection.

A novel tunable metal-clad planar waveguide with 0.62PMN-0.38PT material for detection of cancer cells

A dynamically tunable metal clad planar waveguide having 0.62PMN-0.38PT material is simulated and optimized for detection of cancer cells. Angular interrogation of the TE0 mode of waveguide shows that critical angle increases greater than the resonance angle with increasing of cover refractive index, which limits the detection range of waveguide. To overcome this limitation, proposed waveguide applies a potential on the PMN-PT adlayer. Although a sensitivity of 105.42 degree/RIU was achieved at 70 Volts in testing the proposed waveguide, it was found that the optimal performance parameters were obtained at 60 Volts. At this voltage, the waveguide demonstrated detection range 1.3330-1.5030, a detection accuracy 2393.33, and a figure of merit 2243.59 RIU-1 , which enabled the detection of the entire range of the targeted cancer cells. Therefore, it is recommended to apply a potential of 60 Volts to achieve the best performance from the proposed waveguide.

Brown fat detection by infrared thermography-An invaluable research methodology with noteworthy uncertainties confirmed by a mathematical proof

Brown adipose tissue (BAT) represents a pivotal scientific renaissance worthy as a strategy for obesity and diabetes since its re-discovery in adults over a decade ago. Equally compelling is the adoption of infrared thermography (IRT) in recent times as a precise and viable alternative methodology over the 'gold standard' PET-CT scan, given constraints of the latter's high ionizing radiation doses and costs. Unravelling BAT metabolic physiology in live humans has been challenging until recent rigorous validation of IRT against PET. Nevertheless, IRT remains a nascent technique with pitfalls unbeknownst to many researchers. Factors impacting its accuracy merit an in-depth scientific scrutiny. This article discusses the strengths and pitfalls of IRT as an emergent BAT detection technique and provides a mathematical proof of its limitations that BAT researchers should be cognizant of. Understanding these limitations of IRT can prompt extra efforts to control these uncertainties with greater rigour. In conclusion, this warrants further investigations of improving IRT quality via advanced auto-segmentation, powerful image processing of thermograms and protocol standardization along the lines of BARCIST 1.0 to minimize errors and enhance the confidence of the global BAT research community in IRT as a robust and reliable BAT research tool.

A longitudinal study of the accuracy of positive pressure therapy machine-detected apnea-hypopnea events

STUDY OBJECTIVES

During positive airway pressure (PAP) therapy for sleep apnea syndromes, the machine detected respiratory event index (REI_FLOW) is an important method for clinicians to evaluate the beneficial effects of PAP. There are concerns about the accuracy of this detection, which also confounds a related question-how common and severe are residual events on PAP.

METHODS

Subjects with OSA who underwent a split night polysomnography were recruited prospectively. Those treated with PAP and tracked by the EncoreAnywhere system were analyzed. The ones who stopped PAP within one month were excluded for this analysis. Compliance, therapy data and waveform data were analyzed. Machine detected versus manually scored events were compared at the 1st, 3rd, 6th and 12th month from PAP initiation. Logistic regression was used to determine factors associated with a high REI_FLOW difference.

RESULTS

One hundred and seventy-nine patients with a mean age 59.06 ± 13.97 years old, median body mass index 33.60 (29.75-38.75) kg/m², and median baseline AHI 46.30 (31.50-65.90) times/hour were included. The difference between the machine detected REI_FLOW and manually scored REI_FLOW was 10.72 ±8.43 in the first month and remained stable for up to 12 months. Male sex and large leak ≥ 1.5% were more frequent in patients who had an REI_FLOW difference of ≥ 5 / hour of use. A titration arousal index ≥ 15/ hour of sleep, and higher ratio of unstable to stable breathing were also associated with an REI_FLOW difference ≥ 5 times/hour of use.

CONCLUSIONS

There is a substantial and sustained difference between manual and automated event estimates during PAP therapy, and some associated factors were identified.

Do Not Tell Me More; You Are Honest: A Preconceived Honesty Bias

According to the previous literature, only a few papers found better accuracy than a chance to detect dishonesty, even when more information and verbal cues (VCs) improve precision in detecting dishonesty. A new classification of dishonesty profiles has recently been published, allowing us to study if this low success rate happens for all people or if some people have higher predictive ability. This paper aims to examine if (dis)honest people can detect better/worse (un)ethical behavior of others. With this in mind, we designed one experiment using videos from one of the most popular TV shows in the UK where contestants make a (dis)honesty decision upon gaining or sharing a certain amount of money. Our participants from an online MTurk sample (N = 1,582) had to determine under different conditions whether the contestants would act in an (dis)honest way. Three significant results emerged from these two experiments. First, accuracy in detecting (dis)honesty is not different than chance, but submaximizers (compared to maximizers) and radical dishonest people (compare to non-radicals) are better at detecting honesty, while there is no difference in detecting dishonesty. Second, more information and VCs improve precision in detecting dishonesty, but honesty is better detected using only non-verbal cues (NVCs). Finally, a preconceived honesty bias improves specificity (honesty detection accuracy) and worsens sensitivity (dishonesty detection accuracy).

A Review on Lung Cancer Diagnosis Using Data Mining Algorithms

Due to the serious consequences of lung cancer, medical associations use computer-aided diagnostic procedures to diagnose this disease more accurately. Despite the damaging effects of lung cancer on the body, the lifetime of cancer patients can be extended by early diagnosis. Data mining techniques are practical in diagnosing lung cancer in its first stages. This paper surveys a number of leading data mining-based cancer diagnosis approaches. Moreover, this review draws a comparison between data mining approaches in terms of selection criteria and presents the advantages and disadvantages of each method.

High Three-Dimensional Detection Accuracy in Piezoelectric-Based Touch Panel in Interactive Displays by Optimized Artificial Neural Networks

High detection accuracy in piezoelectric-based force sensing in interactive displays has gained global attention. To achieve this, artificial neural networks (ANN)-successful and widely used machine learning algorithms-have been demonstrated to be potentially powerful tools, providing acceptable location detection accuracy of 95.2% and force level recognition of 93.3% in a previous study. While these values might be acceptable for conventional operations, e.g., opening a folder, they must be boosted for applications where intensive operations are performed. Furthermore, the relatively high computational cost reported prevents the popularity of ANN-based techniques in conventional artificial intelligence (AI) chip-free end-terminals. In this article, an ANN is designed and optimized for piezoelectric-based touch panels in interactive displays for the first time. The presented technique experimentally allows a conventional smart device to work smoothly with a high detection accuracy of above 97% for both location and force level detection with a low computational cost, thereby advancing the user experience, and serviced by piezoelectric-based touch interfaces in displays.

Accuracy, Confidence, and Experiential Criteria for Lie Detection Through a Videotaped Interview

An individual's ability to discriminate lies from truth is far from accurate, and is poorly related to an individual's confidence in his/her detection. Both law enforcement and non-professional interviewers base their evaluations of truthfulness on experiential criteria, including emotional and expressive features, cognitive complexity, and paraverbal aspects of interviewees' reports. The current experimental study adopted two perspectives of investigation: the first is aimed at assessing the ability of naïve judges to detect lies/truth by watching a videotaped interview; the second takes into account the interviewee's detectability as a liar or as telling the truth by a sample of judges. Additionally, this study is intended to evaluate the criteria adopted to support lie/truth detection and relate them with accuracy and confidence of detection. Results showed that judges' detection ability was moderately accurate and associated with a moderate individual sense of confidence, with a slightly better accuracy for truth detection than for lie detection. Detection accuracy appeared to be negatively associated with detection confidence when the interviewee was a liar, and positively associated when the interviewee was a truth-teller. Furthermore, judges were found to support lie detection through criteria concerning emotional features, and to sustain truth detection by taking into account the cognitive complexity and the paucity of expressive manifestations related with the interviewee's report. The present findings have implications for the judicial decision on witnesses' credibility.

Optimization of Detection Accuracy of Closed-Loop Optical Voltage Sensors Based on Pockels Effect

The influence of optical parameters on the performance of closed-loop optical voltage sensors (OVSs) based on Pockels effect is analyzed and a control algorithm is proposed to suppress the nonlinearity caused by the unideal parameters of optical devices for optimizing the detection precision of OVSs. First, a quantified model of the feedback phase demonstrates how the optical parameters of optical devices (including light source, polarizer, 45° fusion point, Faraday rotator and half-wave plate) result in the nonlinearity of closed-loop OVSs. Then, the parameter indexes of different optical devices are put forward to instruct the manufacturing process of the optical system. Furthermore, a closed-loop control algorithm is investigated to improve the measurement accuracy of nonlinear OVSs considering the unideal parameters. The experiment results indicate that additional bias caused by undesirable optical parameters is obviously decreased so that the measurement accuracy of OVSs satisfies the demand of IEC60044-3 for 0.1 level measurement accuracy, which verifies the effectiveness and correctness of the methods for suppressing the impact of unideal optical parameters on OVSs.