UAV Visual and Thermographic Power Line Detection Using Deep Learning.

Inspecting and maintaining power lines is essential for ensuring the safety, reliability, and efficiency of electrical infrastructure. This process involves regular assessment to identify hazards such as damaged wires, corrosion, or vegetation encroachment, followed by timely maintenance to prevent accidents and power outages. By conducting routine inspections and maintenance, utilities can comply with regulations, enhance operational efficiency, and extend the lifespan of power lines and equipment. Unmanned Aerial Vehicles (UAVs) can play a relevant role in this process by increasing efficiency through rapid coverage of large areas and access to difficult-to-reach locations, enhanced safety by minimizing risks to personnel in hazardous environments, and cost-effectiveness compared to traditional methods. UAVs equipped with sensors such as visual and thermographic cameras enable the accurate collection of high-resolution data, facilitating early detection of defects and other potential issues. To ensure the safety of the autonomous inspection process, UAVs must be capable of performing onboard processing, particularly for detection of power lines and obstacles. In this paper, we address the development of a deep learning approach with YOLOv8 for power line detection based on visual and thermographic images. The developed solution was validated with a UAV during a power line inspection mission, obtaining mAP@0.5 results of over 90.5% on visible images and over 96.9% on thermographic images.

Malignant transformation of a long-term ulcer in a patient with leprosy sequelae: a case report.

Marjolin's ulcer (MU) is a rare condition defined as a malignant skin tumor arising from chronic wounds and inflammation. The most common histological findings in MUs are squamous cell carcinomas (SCC) (or spinocellular carcinomas [SPC]), such as basal cell carcinomas (BCC) and malignant melanomas (MM). The aim of this study is to report the case of a patient with sequelae of leprosy who presented malignant transformation in a long-standing ulcer on the right leg, thereby contributing to understanding of the progression and significance of early diagnosis of MU. The MU was diagnosed by incisional biopsy of the lesion and upon obtaining a positive result the patient was referred to an oncology service. Treatment of MU is multidisciplinary and surgical excision is the first therapeutic option. Proper management and surveillance of chronic ulcers by the healthcare team are necessary for early recognition of MU.

An Overview of Innovative Surface-Modification Routes for Pool Boiling Enhancement.

This overview intends to provide a comprehensive assessment of the novel fluids and the current techniques for surface modification for pool boiling enhancement. The surface modification at macro-, micro-, and nanoscales is assessed concerning the underlying fluid routing and capability to eliminate the incipient boiling hysteresis and ameliorate the pool boiling heat-transfer ability, particularly when employed together with self-rewetting fluids and nanofluids with enriched thermophysical properties. Considering the nanofluids, it is viable to take the profit of their high thermal conductivity and their specific heat simultaneously and to produce a film of deposited nanoparticles onto the heating surface, which possesses enhanced surface roughness and an increased density of nucleation sites. Whilst the diverse improvement scales are found to achieve distinct levels of success regarding the nucleate boiling heat-transfer capability enhancement, it is also shown that the micro-nanoscale boiling surface features are susceptible to blockage, leading to the degradation of the improvement with time. Furthermore, topics relating to the heat transfer thermal behavior, ease of manufacture, cost-effectiveness, reliability, and durability are reviewed whenever available and challenges and recommendations for further research are highlighted.

An Overview of the Recent Advances in Pool Boiling Enhancement Materials, Structrure, and Devices.

This review attempts to provide a comprehensive assessment of recent methodologies, structures, and devices for pool boiling heat transfer enhancement. Several enhancement approaches relating to the underlying fluid route and the capability to eliminate incipient boiling hysteresis, augment the nucleate boiling heat transfer coefficient, and improve the critical heat flux are assessed. Hence, this study addresses the most relevant issues related to active and passive enhancement techniques and compound enhancement schemes. Passive heat transfer enhancement techniques encompass multiscale surface modification of the heating surface, such as modification with nanoparticles, tunnels, grooves, porous coatings, and enhanced nanostructured surfaces. Also, there are already studies on the employment of a wide range of passive enhancement techniques, like displaced enhancement, swirl flow aids, and bi-thermally conductive surfaces. Moreover, the combined usage of two or more enhancement techniques, commonly known as compound enhancement approaches, is also addressed in this survey. Additionally, the present work highlights the existing scarcity of sufficiently large available databases for a given enhancement methodology regarding the influencing factors derived from the implementation of innovative thermal management systems for temperature-sensitive electronic and power devices, for instance, material, morphology, relative positioning and orientation of the boiling surface, and nucleate boiling heat transfer enhancement pattern and scale. Such scarcity means the available findings are not totally accurate and suitable for the design and implementation of new thermal management systems. The analysis of more than 100 studies in this field shows that all such improvement methodologies aim to enhance the nucleate boiling heat transfer parameters of the critical heat flux and nucleate heat transfer coefficient in pool boiling scenarios. Finally, diverse challenges and prospects for further studies are also pointed out, aimed at developing important in-depth knowledge of the underlying enhancement mechanisms of such techniques.

Epidemiology, Management, and Survival Outcomes of Germ Cell Cancer in Southern Portugal: A Population-Based Study (2008-2012).

INTRODUCTION: Building on previous suboptimal survival results, we aimed to perform a study of the epidemiological status, management, and outcomes of germ cell tumors (GCT) in the Portuguese population. MATERIALS AND METHODS: Retrospective populational study of GCT cases diagnosed between 2008 and 2012 in southern Portugal. Joinpoint regression was used to compute average annual percentage change (AAPC) in incidence rate. ESMO/EAU guidelines served as references to evaluate compliance. Association between compliance with guidelines and hospital GCT case load was performed by generalized estimating equation. Survival was calculated by Kaplan-Meier and prognostic factors by Cox models. RESULTS: The study included 401 GCT male cases. The AAPC was 5.4% (IC 95% 3.3-7.4, P < .001) from 1999 (an earlier cohort published) to 2012. The median time to diagnosis was 63 days (Q25 = 33 days; Q75 = 114 days; IQR = 81 days). For stage II/III the median time to start chemotherapy was 34 days (Q25 = 22 days; Q75 = 56 days; IQR = 22 days). In 86% cases there was noncompliance with guidelines for the orchiectomy report, 6% for staging, 38% for tumor markers evaluation, 20% for treatment and 25% for chemotherapy dose intensity. The 5-year overall survival was 93.8% (95% CI, 91.3%-96.4%). Hospitals that managed ≤ 3 GCT cases/ year had higher odds for noncompliance with guidelines of blood markers, treatment and dose intensity. None of GCT healthcare access and management factors studied were associated with prognosis. CONCLUSIONS: The burden of GCT is rising in Portugal. Although survival has improved, efforts must be made to nationally enhance training and expertise in GCT and support region adapted models of centralization of care.

Classification of Continuous ECG Segments - Performance Analysis of a Deep Learning Model.

Classification of electrocardiogram (ECG) signals plays an important role in the diagnosis of heart diseases. It is a complex and non-linear signal, which is the first option to preliminary identify specific pathologies/conditions (e.g., arrhythmias). Currently, the scientific community has proposed a multitude of intelligent systems to automatically process the ECG signal, through deep learning techniques, as well as machine learning, where this present high performance, showing state-of-the-art results. However, most of these models are designed to analyze the ECG signal individually, i.e., segment by segment. The scientific community states that to diagnose a pathology in the ECG signal, it is not enough to analyze a signal segment corresponding to the cardiac cycle, but rather an analysis of successive segments of cardiac cycles, to identify a pathological pattern.In this paper, an intelligent method based on a Convolutional Neural Network 1D paired with a Multilayer Perceptron (CNN 1D+MLP) was evaluated to automatically diagnose a set of pathological conditions, from the analysis of the individual segment of the cardiac cycle. In particular, we intend to study the robustness of the referred method in the analysis of several simultaneous ECG signal segments. Two ECG signal databases were selected, namely: MIT-BIH Arrhythmia Database (D1) and European ST-T Database (D2). The data was processed to create datasets with two, three and five segments in a row, to train and test the performance of the method. The method was evaluated in terms of classification metrics, such as: precision, recall, f1-score, and accuracy, as well as through the calculation of confusion matrices.Overall, the method demonstrated high robustness in the analysis of successive ECG signal segments, which we can conclude that it has the potential to detect anomalous patterns in the ECG signal. In the future, we will use this method to analyze the ECG signal coming in real-time, acquired by a wearable device, through a cloud system.Clinical Relevance-This study evaluates the potential of a deep learning method to classify one or several segments of the cardiac cycle and diagnose pathologies in ECG signals.

An Overview of the Nano-Enhanced Phase Change Materials for Energy Harvesting and Conversion.

This review offers a critical survey of the published studies concerning nano-enhanced phase change materials to be applied in energy harvesting and conversion. Also, the main thermophysical characteristics of nano-enhanced phase change materials are discussed in detail. In addition, we carried out an analysis of the thermophysical properties of these types of materials as well as of some specific characteristics like the phase change duration and the phase change temperature. Moreover, the fundamental improving techniques for the phase change materials for solar thermal applications are described in detail, including the use of nano-enhanced phase change materials, foam skeleton-reinforced phase change materials, phase change materials with extended surfaces, and the inclusion of high-thermal-conductivity nanoparticles in nano-enhanced phase change materials, among others. Those improvement techniques can increase the thermal conductivity of the systems by up to 100%. Furthermore, it is also reported that the exploration of phase change materials enhances the overall efficiency of solar thermal energy storage systems and photovoltaic-nano-enhanced phase change materials systems. Finally, the main limitations and guidelines for future research in the field of nano-enhanced phase change materials are summarized.

Workaholism Scales: Some Challenges Ahead.

Although extensively used in the academic literature, workaholism as a concept has been explained in different ways, which has influenced the development and use of some measurement tools. As such, this article aims to address the subject through a systematic study review focusing on articles where the main objective was to develop, adapt, or analyze the psychometric properties of a workaholism scale. The main purpose is to describe the state of the art concerning workaholism measurement tools, highlighting trends and research perspectives for further research. In essence, this study may serve as a summary and starting point for scholars interested in measuring workaholism. It was observed that the discrepancy concerning the definition of workaholism has resulted in scales that attempt to evaluate diverging conceptualizations. Moreover, each scale has been readapted when tested in different countries. For further investigations, it is important to converge the concept of workaholism and validate the scales across differing contexts, regarding the industry, culture, and country of the sample.

Building the Bridge to a Participatory Citizenship: Curricular Integration of Communal Environmental Issues in School Projects Supported by the Internet of Things.

Generally, there is much to praise about the rise in acknowledging the need for young citizens to exercise their rights and duties, but the belief remains that this is not yet entrenched in young citizens' overall democratic involvement. A lack of citizenship and engagement in community issues was revealed by a recent study conducted by the authors in a secondary school from the outskirts of Aveiro, Portugal, during the 2019/2020 school year. Under the umbrella of a Design-Based Research methodological framework, citizen science strategies were implemented in the context of teaching, learning, and assessment, and at the service of the educational project of the target school, in a STEAM approach, and under Domains of Curricular Autonomy activities. The study's findings suggest that to build the bridge for participatory citizenship, teachers should engage students in collecting and analyzing data regarding communal environmental issues in a Citizen Science approach supported by the Internet of Things. The new pedagogies addressing the lack of citizenship and engagement in community issues promoted students' involvement at school and in the community, contributed to inform municipal education policies, and promoted dialogue and communication between local actors.

Cardiologist's exposure to radiation in cath lab measured with InstadoseTMdosimeter.

INTRODUCTION: complex fluoroscopy-guided interventional procedures in cardiology are known to result in higher radiation doses for patients and staff. PURPOSE: to estimate the equivalent dose received in different regions of the cardiologist's body in catheterism (CATH) and percutaneous coronary intervention (PCI) procedures, as well as to evaluate the effectiveness of monitoring the doses in the catheritization laboratory (cath lab) using a direct ion storage dosimeter. MATERIALS AND METHODS: the InstadoseTMand the thermoluminescent dosimeters (TLD-100) were fixed simultaneously in the following regions of the cardiologist's body: near the eyes (left and right), the trunk region (over the lead apron) and the left ankle. Occupational doses were recorded during 86 procedures (60% CATH). RESULTS: catheterization procedures showed third quartile dose values near to the left eye region equal to 0.10 mSv (TLD-100) and 0.12 (InstadoseTM) and for intervention 0.15 mSv (TLD-100 and InstadoseTM). The doses measured in the trunk region, over the lead apron, were about 13% higher for catheterization procedures and 20% higher for intervention procedures compared to left eye region measurements. The Wilcoxon-Mann-Whitney test was applied for unpaired data for all body regions, comparing the data obtained between the TLD-100 and InstadoseTMdosimeters. For CATH and PCI, the responses of the TLD-100 and InstadoseTMdosimeters are considered equal for all analysed regions (p> 0.05) with the exception of the right eye region. CONCLUSION: the InstadoseTMpassive dosimeter can be useful as a complementary assessment in the monitoring of a cardiologist's personal occupational doses in the cath lab.

Adjuvant Chemotherapy De-Escalation with Genomic Assay Protocol in Patients with Early Breast Cancer: A Single-Centre Prospective Cohort Study.

INTRODUCTION: Genomic assays are useful tools for tailoring adjuvant treatment in early breast cancer. We aimed to analyse the role of an institutional protocol of a genomic assay for chemotherapy de-escalation. MATERIAL AND METHODS: Prospective cohort study of all consecutive women diagnosed with hormone receptor-positive and human epidermal growth factor receptor 2-negative early breast cancer, tested with the 21-gene Recurrence Score (RS) assay from August 2015 to July 2018 at a Portuguese cancer centre. For being tested, patients should meet at least one of the pre-defined inclusion criteria: i) luminal A-like, pT2pN0; ii) luminal A-like, 1 - 3 positive nodes and comorbidities with higher risk of chemotherapy-induced toxicity; iii) pT1-2pN0, progesterone receptor ≤ 20% and/or Ki67 14% - 40%. Adjuvant treatment was de-escalated to isolated endocrine therapy if RS was less than 18. We measured the reduction in chemotherapy prescribing and its clinical impact, the RS association with pathologic features, and the protocol feasibility. RESULTS: We tested 154 women with a median age of 61 years old (range: 25 - 79), 69% postmenopausal. Tumours were mainly pT1 (55%), pN0 (82%), invasive ductal (73%), G2 (86%), luminal B-like (69%) and stage IA (85%). We obtained a RS less than 18 in 60% of women, with an overall adjuvant chemotherapy reduction of 65%. Seven (95% confidence interval: 5 - 10) patients needed to be screened with the 21-gene RS assay to prevent one clinically relevant adverse event during the first six months of adjuvant treatment. Considering the currently used RS cut-off, only 9% of node-negative and 11% of node-positive patients had RS over 25. We found no relevant associations between RS and pathologic features. The protocol was feasible and did not compromise the adequate timing for adjuvant treatment. CONCLUSION: These criteria allowed the de-escalation of adjuvant systemic treatment in at least six out of ten women.

The Pool-Boiling-Induced Deposition of Nanoparticles as the Transient Game Changer-A Review.

It is widely known by the scientific community that the suspended nanoparticles of nanofluids can enhance the thermophysical properties of base fluids and maximize pool-boiling heat transfer. However, the nanoparticles may undergo extended boiling times and deposit onto the heating surfaces under pool-boiling conditions, thus altering their intrinsic characteristics such as wettability and roughness over time. The present study reviews the fundamental mechanisms and characteristics of nanoparticle deposition, and its impact on surface roughness and wettability, density of vaporized core points, and thermal resistance, among other factors. Moreover, the effect of the nanoparticle layer in long-term thermal boiling performance parameters such as the heat transfer coefficient and critical heat flux is also discussed. This work attempts to highlight, in a comprehensive manner, the pros and cons of nanoparticle deposition after extended pool-boiling periods, leading the scientific community toward further investigation studies of pool-boiling heat-transfer enhancement using nanofluids. This review also attempts to clarify the inconsistent results of studies on heat transfer parameters using nanofluids.

The Effect of Auriculotherapy on Situational Anxiety Trigged by Examinations: A Randomized Pilot Trial.

BACKGROUND: Auriculotherapy may activate the parasympathetic nerve system and reduce anxiety levels. Short-term auriculotherapy's effects and safety on university students' anxiety levels was assessed prior to exams. METHODS: A randomized, controlled pilot trial was conducted. The day before the exam, university students were randomly allocated to the auriculotherapy group (AA, n = 13) or the waiting-list group (WG, n = 13). Baseline measures were taken 4 weeks before the exam at Time point (TP 0); at 7.30 a.m. on the day before the exam (TP I); at 11 a.m. before auriculotherapy (TP II); 30 min after AA (TP III); and at 7.30 a.m. before the exam (TP IV). The outcomes were the State-Trait-Anxiety Inventory (STAI); quality of night-sleep, Visual Analogue scale (VAS) for anxiety, and salivary cortisol. Adverse events were also recorded. RESULTS: A total of 26 students participated in this study and became more anxious as assessed by STAI in TPII (p = 0.002) and TPIV (p = 0.000) than TP0. AA reduced the STAI in TPIII (p = 0.045) and PIV (p = 0.001) and the VAS (p = 0.012) in TPIV. Cortisol was reduced in TPIII (p = 0.004), and the AA slept better (p = 0.014) at TPIV. Discomfort at the auricular site was reported in only one AA participant. CONCLUSIONS: Auriculotherapy appeared safe and effective in reducing anxiety levels before university exams.

A sensorized needle guide for ultrasound assisted breast biopsy.

One in every eight women will get breast cancer during their lifetime. Therefore, the early diagnosis of the lesions is fundamental to improve the chances of recovery. To find breast cancers, breast screening using techniques such as mammography and ultrasound (US) imaging scans are often used. When a lesion is found, a breast biopsy is performed to extract a tissue sample for analysis. The breast biopsy is usually assisted by an US to help find the lesion and guide the needle to its location. However, the identification of the needle tip in US image is challenging, possibly resulting in puncture failures. In this paper, we intend to study the potential of a sensorized needle guide system that provides information about the needle angle and displacement in respect to the US probe. Laboratory tests were initially conducted to evaluate the accuracy of each sensor in controlled conditions. After, a practical experiment with the US probe, working as a proof of concept, was performed. The angle sensor showed a root mean square error (RMSE) of 0.48 degrees and the displacement sensor showed a RMSE of 0.26mm after being calibrated. For the US probe tests, the displacement sensor shows high errors in the range of 1.19mm to 2.05mm due to mechanical reasons. Overall, the proposed system showed its potential to be used to accurately estimate needle tip localization throughout breast biopsies guided by US, corroborating its potential clinical application. Clinical relevance - Potential for clinical application where precise needle localization in ultrasound image is required.

Augmented Reality for Human-Robot Collaboration and Cooperation in Industrial Applications: A Systematic Literature Review.

With the continuously growing usage of collaborative robots in industry, the need for achieving a seamless human-robot interaction has also increased, considering that it is a key factor towards reaching a more flexible, effective, and efficient production line. As a prominent and prospective tool to support the human operator to understand and interact with robots, Augmented Reality (AR) has been employed in numerous human-robot collaborative and cooperative industrial applications. Therefore, this systematic literature review critically appraises 32 papers' published between 2016 and 2021 to identify the main employed AR technologies, outline the current state of the art of augmented reality for human-robot collaboration and cooperation, and point out future developments for this research field. Results suggest that this is still an expanding research field, especially with the advent of recent advancements regarding head-mounted displays (HMDs). Moreover, projector-based and HMDs developed approaches are showing promising positive influences over operator-related aspects such as performance, task awareness, and safety feeling, even though HMDs need further maturation in ergonomic aspects. Further research should focus on large-scale assessment of the proposed solutions in industrial environments, involving the solution's target audience, and on establishing standards and guidelines for developing AR assistance systems.

Benign Metastasizing Leiomyomatosis to the Skin and Lungs, Intravenous Leiomyomatosis, and Leiomyomatosis Peritonealis Disseminata: A Series of Five Cases.

Benign metastasizing leiomyomatosis (BML) is a rare disease that typically occurs in women with a history of uterine leiomyomatosis. Benign metastasizing leiomyomatosis occurs more frequently in the lungs but may also develop in other organs and tissues. Other unusual variants of extra-uterine leiomyomatosis include intravenous leiomyomatosis (IVL) and leiomyomatosis peritonealis disseminata (LPD). In this article, three cases of BML are presented. One case, in a premenopausal woman, presented cutaneous metastases. We also present a case of IVL and a case of LPD, which occurred in postmenopausal women. Given the rarity of BML, IVL, and LPD, the authors reviewed the literature and herein discuss the implications for treatment in all five cases. Evidence for treating BML, IVL, and LPD is still scarce, and data available from our series and other small series seem to point to the patient's hormonal status playing a fundamental part in the treatment plan. Furthermore, a collecting bag when performing excision of uterine leiomyomas may help avoid the potential spreading of leiomyomatosis. Hysterectomized patients with chronic cough, frequent respiratory infections, abdominal discomfort, right heart failure, or non-specific symptoms should be actively screened for BML, IVL, and LPD. Treatment should be individualized according to each patient's hormonal status and desires.

Factors related to human-vector contact that modify the likelihood of malaria transmission during a contained Plasmodium falciparum outbreak in Praia, Cabo Verde.

Background: Determining the reproductive rate and how it varies over time and space (RT) provides important insight to understand transmission of a given disease and inform optimal strategies for controlling or eliminating it. Estimating RT for malaria is difficult partly due to the widespread use of interventions and immunity to disease masking incident infections. A malaria outbreak in Praia, Cabo Verde in 2017 provided a unique opportunity to estimate RT directly, providing a proxy for the intensity of vector-human contact and measure the impact of vector control measures. Methods: Out of 442 confirmed malaria cases reported in 2017 in Praia, 321 (73%) were geolocated and informed this analysis. RT was calculated using the joint likelihood of transmission between two cases, based on the time (serial interval) and physical distance (spatial interval) between them. Log-linear regression was used to estimate factors associated with changes in RT, including the impact of vector control interventions. A geostatistical model was developed to highlight areas receptive to transmission where vector control activities could be focused in future to prevent or interrupt transmission. Results: The RT from individual cases ranged between 0 and 11 with a median serial- and spatial-interval of 34 days [interquartile range (IQR): 17-52] and 1,347 m (IQR: 832-1,985 m), respectively. The number of households receiving indoor residual spraying (IRS) 4 weeks prior was associated with a reduction in RT by 0.84 [95% confidence interval (CI) 0.80-0.89; p-value <0.001] in the peak-and post-epidemic compared to the pre-epidemic period. Conclusions: Identifying the effect of reduced human-vector contact through IRS is essential to determining optimal intervention strategies that modify the likelihood of malaria transmission and can inform optimal intervention strategies to accelerate time to elimination. The distance within which two cases are plausibly linked is important for the potential scale of any reactive interventions as well as classifying infections as imported or introduced and confirming malaria elimination.

Programming Robots by Demonstration Using Augmented Reality.

The world is living the fourth industrial revolution, marked by the increasing intelligence and automation of manufacturing systems. Nevertheless, there are types of tasks that are too complex or too expensive to be fully automated, it would be more efficient if the machines were able to work with the human, not only by sharing the same workspace but also as useful collaborators. A possible solution to that problem is on human-robot interaction systems, understanding the applications where they can be helpful to implement and what are the challenges they face. This work proposes the development of an industrial prototype of a human-machine interaction system through Augmented Reality, in which the objective is to enable an industrial operator without any programming experience to program a robot. The system itself is divided into two different parts: the tracking system, which records the operator's hand movement, and the translator system, which writes the program to be sent to the robot that will execute the task. To demonstrate the concept, the user drew geometric figures, and the robot was able to replicate the operator's path recorded.

Evaluating the Single-Shot MultiBox Detector and YOLO Deep Learning Models for the Detection of Tomatoes in a Greenhouse.

The development of robotic solutions for agriculture requires advanced perception capabilities that can work reliably in any crop stage. For example, to automatise the tomato harvesting process in greenhouses, the visual perception system needs to detect the tomato in any life cycle stage (flower to the ripe tomato). The state-of-the-art for visual tomato detection focuses mainly on ripe tomato, which has a distinctive colour from the background. This paper contributes with an annotated visual dataset of green and reddish tomatoes. This kind of dataset is uncommon and not available for research purposes. This will enable further developments in edge artificial intelligence for in situ and in real-time visual tomato detection required for the development of harvesting robots. Considering this dataset, five deep learning models were selected, trained and benchmarked to detect green and reddish tomatoes grown in greenhouses. Considering our robotic platform specifications, only the Single-Shot MultiBox Detector (SSD) and YOLO architectures were considered. The results proved that the system can detect green and reddish tomatoes, even those occluded by leaves. SSD MobileNet v2 had the best performance when compared against SSD Inception v2, SSD ResNet 50, SSD ResNet 101 and YOLOv4 Tiny, reaching an F1-score of 66.15%, an mAP of 51.46% and an inference time of 16.44ms with the NVIDIA Turing Architecture platform, an NVIDIA Tesla T4, with 12 GB. YOLOv4 Tiny also had impressive results, mainly concerning inferring times of about 5 ms.