Search Results (167)

Monocular 3D object detection (Mono3D) is essential for autonomous driving and augmented reality, yet its performance degrades significantly at night due to the scarcity of annotated nighttime data. In this paper, we investigate the use of style transfer for nighttime data augmentation and evaluate its effect on individual components of 3D detection. Using CycleGAN, we generated synthetic night images from daytime scenes in the nuScenes dataset and trained a modular Mono3D detector under different configurations. Our results show that training solely on style-transferred images improves certain metrics, such as AP@0.95 (from 0.0299 to 0.0778, a 160% increase) and depth error (11% reduction), compared to daytime-only baselines. However, performance on orientation and dimension estimation deteriorates. When real nighttime data is included, style transfer provides complementary benefits: for cars, depth error decreases from 0.0414 to 0.021, and AP@0.95 remains stable at 0.66; for pedestrians, AP@0.95 improves by 13% (0.297 to 0.336) with a 35% reduction in depth error. Cyclist detection remains unreliable due to limited samples. We conclude that style transfer cannot replace authentic nighttime data, but when combined with it, it reduces false positives and improves depth estimation, leading to more robust detection under low-light conditions. This study highlights both the potential and the limitations of style transfer for augmenting Mono3D training, and it points to future research on more advanced generative models and broader object categories. Full article

Jiangsu Province is an important economic province on the eastern coast of China, revealing the spatial–temporal characteristics, dynamic degree, and transition direction of land use/cover change, and its main driving factors are significant for the effective use of land resources and the promotion of regional human–land coordinated development. Based on land use data of Jiangsu Province from 2000 to 2020, this study investigates the spatiotemporal evolution characteristics of land use/cover using the dynamics model and the transfer matrix model, and examines the influence and interaction of the driving factors between human activities and the natural environment based on 10-factor data using Geodetector. The results showed that (1) In the past 20 years, the type of land use/cover in Jiangsu Province primarily comprises cropland, water, and impervious, with the land use/cover change mode mainly consisting of a dramatic change in cropland and impervious and relatively little change in forest, grassland, water, and barren. (2) From the perspective of the dynamic rate of land use/cover change, the single land use dynamic degree showed that impervious is the only land type whose dynamics have positively increased from 2000 to 2010 and 2010 to 2020, with values of 3.67% and 3.03%, respectively. According to the classification of comprehensive motivation, the comprehensive land use motivation in Jiangsu Province in each time period from 2000 to 2010 and 2010 to 2020 is 0.46% and 0.43%, respectively, which belongs to the extremely slow change type. (3) From the perspective of land use/cover transfer, Jiangsu Province is mainly characterized by a large area of cropland transfer (−7954.30 km²) and a large area of impervious transfer (8759.58 km²). The increase in impervious is mainly attributed to the transformation of cropland and water, accounting for 4066.07 km² and 513.73 km² from 2010 to 2020, which indicates that the non-agricultural phenomenon of cropland in Jiangsu Province, i.e., the process of transforming cropland into non-agricultural construction land, is significant. (4) From the perspective of driving factors, population density (q = 0.154) and night light brightness (q = 0.156) have always been important drivers of land use/cover change in Jiangsu Province. The interaction detection indicates that the land use/cover change is driven by both socio-economic factors and natural geographic factors. (5) In response to the dual pressures of climate change and rapid urbanization, coordinating the multiple objectives of socio-economic development, food security, and ecological protection is the fundamental path to achieving sustainable land use in Jiangsu Province and similar developed coastal areas. By revealing the characteristics and driving factors of land use/cover change in Jiangsu Province, this study provides qualitative and quantitative theoretical support for the coordinated decision-making of economic development and land use planning in Jiangsu Province, specifically contributing to sustainable land planning, climate adaptation policy-making, and the enhancement of community well-being through optimized land use. Full article

In the field of drone-to-drone detection tasks, the issue of fusing temporal information with infrared and visible light data for detection has been rarely studied. This paper presents the first temporal misaligned rgb-thermal dataset for drone-to-drone target detection, named TMRGBT-D2D. The dataset covers various lighting conditions (i.e., high-light scenes captured during the day, medium-light and low-light scenes captured at night, with night scenes accounting for 38.8% of all data), different scenes (sky, forests, buildings, construction sites, playgrounds, roads, etc.), different seasons, and different locations, consisting of a total of 42,624 images organized into sequential frames extracted from 19 RGB-T video pairs. Each frame in the dataset has been meticulously annotated, with a total of 94,323 annotations. Except for drones that cannot be identified under extreme conditions, infrared and visible light annotations are one-to-one corresponding. This dataset presents various challenges, including small object detection (the average size of objects in visible light images is approximately 0.02% of the image area), motion blur caused by fast movement, and detection issues arising from imaging differences between different modalities. To our knowledge, this is the first temporal misaligned rgb-thermal dataset for drone-to-drone target detection, providing convenience for research into rgb-thermal image fusion and the development of drone target detection. Full article

Background/Objective: The overnight 1-mg dexamethasone suppression test (DST) represents the conventional/standard tool for endogenous hypercortisolemia screening, typically in relationship with adrenal and pituitary masses. Nevertheless, an associated spectrum of challenges and pitfalls is found in daily practice. This analysis aimed to evaluate: (I.) the diagnosis relevance of 1-mg DST in patients with adrenal incidentalomas (AIs) with/without mild autonomous cortisol secretion (MACS) exploring different cutoffs of the second-day plasma cortisol after dexamethasone administration (cs-DST) with respect to cardio-metabolic outcomes; (II.) the potential utility of adding other biomarkers to DST [plasma morning adrenocorticotropic hormone (ACTH), 24-h urinary free cortisol (UFC), late-night salivary cortisol (LNSC), dehydroepiandrosterone sulfate (DHEAS)]; and (III.) DST variability in time. Methods: This narrative analysis was based on searching full-text, English articles in PubMed (between January 2023 and April 2025) via using different term combinations: “dexamethasone suppression test” (n = 239), “diagnosis test for autonomous cortisol secretion” (n = 22), “diagnosis test for mild autonomous cortisol secretion” (n = 13) and “diagnosis test for Cushing Syndrome” (n = 61). We manually checked the title and abstract and finally included only the studies that provided hormonal testing results in adults with non-functional adenomas (NFAs) ± MACS. We excluded: reviews, meta-analyses, editorials, conference abstracts, case reports, and case series; non-human research; studies that did not provide clear criteria for distinguishing between Cushing syndrome and MACS; primary aldosteronism. Results: The sample-focused analysis (n = 13 studies) involved various designs: cross-sectional (n = 4), prospective (n = 1), retrospective (n = 7), and cohort (n = 1); a total of 4203 patients (female-to-male ratio = 1.45), mean age of 59.92 years. I. Cs-DST cutoffs varied among the studies (n = 6), specifically, 0.87, 0.9, 1.2, and 1.4 µg/dL in relationship with the cardio-metabolic outcomes. After adjusting for age (n = 1), only the prevalence of cardiovascular disease remained significantly higher in >0.9 µg/dL vs. ≤0.9 group (OR = 2.23). Multivariate analysis (n = 1) found cs-DST between 1.2 and 1.79 µg/dL was independently associated with hypertension (OR = 1.55, 95%CI: 1.08–2.23, p = 0.018), diabetes (OR = 1.60, 95%CI: 1.01–2.57, p = 0.045), and their combination (OR = 1.96, 95%CI:1.12–3.41, p = 0.018) after adjusting for age, gender, obesity, and dyslipidemia. A higher cs-DST was associated with a lower estimated glomerular filtration rate (eGFR), independently of traditional cardiovascular risk factors. Post-adrenalectomy eGFR improvement was more pronounced in younger individuals, those with lower eGFR before surgery, and with a longer post-operative follow-up. Cs-DST (n = 1) was strongly associated with AIs size and weakly associated with age, body mass index and eGFR. Cortisol level increased by 9% (95% CI: 6–11%) for each 10 mL/min/1.73 m² decrease in eGFR. A lower cs-DST was associated with a faster post-adrenalectomy function recovery; the co-diagnosis of diabetes reduced the likelihood of this recovery (OR = 24.55, p = 0.036). II. Additional biomarkers assays (n = 5) showed effectiveness only for lower DHEAS to pinpoint MACS amid AIs (n = 2, cutoffs of <49.31 µg/dL, respectively, <75 µg/dL), and lower ACTH (n = 1, <12.6 pmol/L). III. Longitudinal analysis of DST’s results (n = 3): 22% of NFAS switch to MACS after a median of 35.7 months (n = 1), respectively, 29% (n = 1) after 48.6 ± 12.5 months, 11.8% (n = 1) after 40.4 ± 51.17 months. A multifactorial model of prediction showed the lowest risk of switch (2.4%) in individuals < 50 years with unilateral tumor and cs-DST < 0.45 µg/dL. In the subgroup of subjects without cardio-metabolic comorbidities at presentation, 25.6% developed ≥1 comorbidities during surveillance. Conclusions: The importance of exploring the domain of AIs/NFAs/MACS relates to an increasing detection in aging population, hence, the importance of their optimum hormonal characterization and identifying/forestalling cardio-metabolic consequences. The spectrum of additional biomarkers in MACS (other than DST) remains heterogeneous and still controversial, noting the importance of their cost-effectiveness, and availability in daily practice. Cs-DST serves as an independent predictor of cardio-metabolic outcomes, kidney dysfunction, while adrenalectomy may correct them in both MACS and NFAs, especially in younger population. Moreover, it serves as a predictor of switching the NFA into MACS category during surveillance. Changing the hormonal behavior over time implies awareness, since it increases the overall disease burden. Full article

Background/Objectives: Primary nocturnal enuresis (PNE) is a common condition that adversely affects the quality of life of both children and their families. It is known to have a multifactorial pathogenesis. This study aimed to evaluate serum levels of 25-hydroxyvitamin D (25OHD), vitamin B12, folic acid, and ferritin in children diagnosed with PNE and to investigate the impact of correcting detected deficiencies on the number of wet nights. Methods: A total of 150 pediatric patients diagnosed with monosymptomatic primary nocturnal enuresis (PNE) who had previously undergone standard urotherapy without clinical improvement were included in this study. Serum levels of vitamin B12 and 25-hydroxyvitamin D (25OHD) were assessed, and patients with deficiencies were identified. Vitamin supplementation was administered to those with deficient/insufficient levels. The number of wet nights was recorded at monthly follow-up visits to monitor clinical response. Results: Only 14% of the 150 patients had no detectable vitamin deficiencies. A deficiency in serum vitamin B12 levels was observed in 78.6% of patients, while 41.3% had reduced 25-hydroxyvitamin D (25OHD) levels. Concurrent deficiencies in both 25OHD and vitamin B12 were detected in 34% of the patients. No folate deficiency was observed in any patient. Notably, vitamin supplementation alone resulted in successful enuresis management in 77.6% of the patients. Conclusions: A high prevalence of vitamin B12 and 25-hydroxyvitamin D (25OHD) deficiencies was identified among patients diagnosed with primary nocturnal enuresis (PNE). Significant improvements in nocturnal dryness were achieved solely through correction of these deficiencies, without the use of desmopressin therapy. These findings suggest that targeted vitamin supplementation may play a crucial role in enhancing the success rate of standard urotherapy in the management of PNE. Full article

Vitamin A deficiency (VAD) remains a significant cause of preventable blindness worldwide, with ocular surface changes representing early manifestations that require prompt recognition and treatment. Conventional examination methods are capable of detecting advanced changes; however, subtle conjunctival abnormalities may be overlooked, potentially delaying the administration of appropriate interventions. We herein present the case of a 5-year-old Japanese boy with severe VAD due to selective eating patterns. This case demonstrates the utility of infrared photography as a novel diagnostic approach for detecting and monitoring conjunctival surface abnormalities. The patient exhibited symptoms including corneal ulcers, night blindness, and reduced visual acuity. Furthermore, blood tests revealed undetectable levels of vitamin A (5 IU/dL), despite relatively normal physical growth parameters. Conventional slit-lamp examination revealed characteristic sandpaper-like conjunctival changes. However, infrared photography (700–900 nm wavelength) revealed distinct abnormal patterns of conjunctival surface folds and keratinization that were not fully appreciated on a routine examination. Following high-dose vitamin A supplementation (4000 IU/day), complete resolution of ocular abnormalities was achieved within 2 months, with infrared imaging objectively documenting treatment response and normalization of conjunctival surface patterns. This case underscores the potential for severe VAD in developed countries, particularly in the context of dietary restrictions, thereby underscoring the significance of a comprehensive dietary history and a meticulous ocular examination. Infrared photography provides a number of advantages, including the capacity for non-invasive assessment, enhanced visualization of subtle changes, objective monitoring of treatment response, and cost-effectiveness due to the use of readily available equipment. This technique represents an underutilized diagnostic modality with particular promise for screening programs and clinical monitoring of VAD-related ocular manifestations, potentially preventing irreversible visual loss through early detection and intervention. Full article

Real-time wildlife monitoring on edge devices poses significant challenges due to limited power, constrained bandwidth, and unreliable connectivity, especially in remote natural habitats. Conventional object detection systems often transmit redundant data of the same animals detected across multiple consecutive frames as a part of a single event, resulting in increased power consumption and inefficient bandwidth usage. Furthermore, maintaining consistent animal identities in the wild is difficult due to occlusions, variable lighting, and complex environments. In this study, we propose a lightweight hybrid tracking framework built on the YOLOv8m deep neural network, combining motion-based Kalman filtering with Local Binary Pattern (LBP) similarity for appearance-based re-identification using texture and color features. To handle ambiguous cases, we further incorporate Hue-Saturation-Value (HSV) color space similarity. This approach enhances identity consistency across frames while reducing redundant transmissions. The framework is optimized for real-time deployment on edge platforms such as NVIDIA Jetson Orin Nano and Raspberry Pi 5. We evaluate our method against state-of-the-art trackers using event-based metrics such as MOTA, HOTA, and IDF1, with a focus on detected animals occlusion handling, trajectory analysis, and counting during both day and night. Our approach significantly enhances tracking robustness, reduces ID switches, and provides more accurate detection and counting compared to existing methods. When transmitting time-series data and detected frames, it achieves up to 99.87% bandwidth savings and 99.67% power reduction, making it highly suitable for edge-based wildlife monitoring in resource-constrained environments. Full article

This study evaluated the performance of the operational GK2A (GEO-KOMPSAT-2A) fog detection algorithm (GK2A_FDA) using ground-based visibility meter data from 176 stations across South Korea from 2021 to 2023. According to the verification method using the nearest pixel and 3 × 3 neighborhood pixel approaches to the visibility meter, the 3-year average probability of detection (POD) is 0.59 and 0.70, the false alarm ratio (FAR) is 0.86 and 0.81, and the bias is 4.25 and 3.73, respectively. POD is highest during daytime (0.72; bias: 7.34), decreases at night (0.57; bias: 3.89), and is lowest at twilight (0.52; bias: 2.36). The seasonal mean POD is 0.65 in winter, 0.61 in spring and autumn, and 0.47 in summer, with August reaching the minimum value, 0.33. While POD is higher in coastal areas than inland areas, inland regions show lower FAR, indicating more stable performance. Over-detections occurred regardless of geographic location and time, mainly due to the misclassification of low-level clouds and cloud edges as fog. Especially after sunrise, the fog dissipated and transformed into low-level clouds. These findings suggest that there are limitations to improving fog detection levels using satellite data alone, especially when the surface is obscured by clouds, indicating the need to utilize other data sources, such as objective ground-based analysis data. Full article

Underground cast-in-place pipes (CIPP, Diameter of 2’–5’) are used to transport water for the Phoenix, AZ area. These pipes have developed leaks due to their age and changes in the environment, resulting in a significant waste of water. Currently, leaks can only be identified when water pools above ground occur and are then manually confirmed through the inside of the pipe, requiring the shutdown of the water system. However, many leaks may not develop a puddle of water, making them even harder to identify. The primary objective of this research was to develop an inspection method utilizing drone-based infrared imagery to remotely and non-invasively sense thermal signatures of abnormal soil moisture underneath urban surface treatments caused by the leakage of water pipelines during the regular operation of water transportation. During the field tests, five known leak sites were evaluated using an intensive experimental procedure that involved conducting multiple flights at each test site and a stringent filtration process for the measured temperature data. A detectable thermal signal was observed at four of the five known leak sites, and these abnormal thermal signals directly overlapped with the location of the known leaks provided by the utility company. A strong correlation between ground temperature and shading before sunset was observed in the temperature data collected at night. Thus, a shadow and solar energy model was implemented to estimate the position of shadows and energy flux at given times based on the elevation of the surrounding structures. Data fusion between the metrics of shadow time, solar energy, and the temperature profile was utilized to filter the existing points of interest further. When shadows and solar energy were considered, the final detection rate of drone-based infrared imaging was determined to be 60%. Full article

This paper explores the potential of public webcams as a source of data for transport research. Eight different open-source object detection models were tested on three publicly accessible webcams located in the city of Brunswick, Germany. Fifteen images at different lighting conditions (bright light, dusk, and night) were selected from each webcam and manually labelled with regard to the following six categories: cars, persons, bicycles, trucks, trams, and buses. The manual counts in these six categories were then compared to the number of counts found by the object detection models. The results show that public webcams constitute a useful source of data for transport research. In bright light conditions, applying out-of-the-box object detection models can yield reliable counts of cars or persons in public squares, streets, and junctions. However, the detection of cars and persons was not reliably accurate at dusk or night. Thus, different object detection models might have to be used to generate accurate counts in different lighting conditions. Furthermore, the object detection models worked less well for identifying trams, buses, bicycles, and trucks. Hence fine-tuning and adapting the models to the specific webcams might be needed to achieve satisfactory results for these four types of traffic participants. Full article

Background/Objectives: This study aimed to explore the functional implications of occlusal changes during clear aligner treatment (CAT) to (a) analyze occlusal changes throughout CAT and the extent of post-treatment occlusal recovery; (b) assess the relationship between post-treatment occlusion and masticatory performance; (c) investigate whether case complexity, facial biotype, and type of malocclusion influence occlusal adaptation and functional outcomes; and (d) evaluate the presence and progression of signs or symptoms of TMDs in patients undergoing CAT. Methods: This longitudinal cohort pilot study included 42 individuals who underwent CAT. Occlusion was evaluated at three timepoints: before treatment (T0), at treatment completion (T1), and three months after with night-only aligner use (T2). Masticatory performance was assessed using a two-colored chewing gum test analyzed through colorimetric software. TMD signs/symptoms were assessed using the Diagnostic Criteria for TMD [DC/TMD]. Statistical analysis used non-parametric tests. Results: A significant decrease in occlusal contact area was observed during active CAT [p = 0.016], which partially recovered at follow-up. Individuals with normal facial proportions (normodivergent) showed more anterior contacts at T1 compared to hyperdivergent individuals [p = 0.013]. Masticatory performance remained stable between T1 and T2 [p = 0.528]. A weak negative correlation was found between posterior contact number and performance score at T1 [r = −0.378, p < 0.05], suggesting that more contacts may be linked to better chewing. No TMD signs or symptoms were detected at any timepoint. Conclusions: Although CAT temporarily reduces occlusal contact area, it does not negatively impact chewing efficiency or trigger TMD symptoms. These findings support the functional safety of CAT when treatment is properly planned and monitored. Full article

Dark ships, vessels deliberately disabling their AIS signals, constitute a grave maritime safety hazard, with detection efforts hindered by issues like over-reliance on AIS, inadequate surveillance coverage, and significant mismatch rates. This paper proposes an improved multi-feature association method that integrates satellite remote sensing and AIS data, with a focus on oriented bounding box course estimation, to improve the detection of dark ships and enhance maritime surveillance. Firstly, the oriented bounding box object detection model (YOLOv11n-OBB) is trained to break through the limitations of horizontal bounding box orientation representation. Secondly, by integrating position, dimensions (length and width), and course characteristics, we devise a joint cost function to evaluate the combined significance of multiple features. Subsequently, an advanced JVC global optimization algorithm is employed to ensure high-precision association in dense scenes. Finally, by integrating data from Gaofen-6 (optical) and Gaofen-3B (SAR) satellites, a day-and-night collaborative monitoring framework is constructed to address the blind spots of single-sensor monitoring during night-time or adverse weather conditions. Our results indicate that the detection model demonstrates a high average precision (AP50) of 0.986 on the optical dataset and 0.903 on the SAR dataset. The association accuracy of the multi-feature association algorithm is 91.74% in optical image and AIS data matching, and 91.33% in SAR image and AIS data matching. The association rate reaches 96.03% (optical) and 74.24% (SAR), respectively. This study provides an efficient technical tool for maritime safety regulation through multi-source data fusion and algorithm innovation. Full article

Background: Traumatic brain injury (TBI) is a major cause of morbimortality in the world, and it can cause potential intracranial hemorrhage (ICH), a life-threatening condition that requires rapid diagnosis with computed tomography (CT). Artificial intelligence tools for ICH detection are now commercially available. Objectives: Investigate the real-world performance of qER.ai, an artificial intelligence-based CT hemorrhage detection tool, in a post-traumatic population. Methods: Retrospective monocentric observational study of a dataset of consecutively acquired head CT scans at the emergency radiology unit to explore brain trauma. AI performance was compared to ground truth determined by expert consensus. A subset of night shift cases with the radiological report of a junior resident was compared to the AI results and ground truth. Results: A total of 682 head CT scans were analyzed. AI demonstrated a sensitivity of 88.8% and a specificity of 92.1% overall, with a positive predictive value of 65.4% and a negative predictive value of 98%. AI’s performance was comparable to that of junior residents in detecting ICH, with the latter showing a sensitivity of 85.7% and a high specificity of 99.3%. Interestingly, the AI detected two out of three ICH cases missed by the junior residents. When AI assistance was integrated, the combined sensitivity improved to 95.2%, and the overall accuracy reached 98.8%. Conclusions: This study shows better performance from AI and radiologist residents working together than each one alone. These results are encouraging for rethinking the radiological workflow and the future of triage of this large population of brain traumatized patients in the emergency unit. Full article

Nighttime object detection poses significant challenges due to low illumination, noise, and reduced contrast, which can severely impact the performance of standard detection models. In this paper, we present NF-DETR (Night-Frequency Detection Transformer), a novel framework that leverages frequency domain information to enhance object detection in challenging nighttime environments. Our approach integrates physics-prior enhancement to improve the visibility of objects in low-light conditions, frequency domain feature extraction to capture structural information potentially lost in the spatial domain, and window cross-attention fusion that efficiently combines complementary features while reducing computational complexity, significantly improving detection performance without increasing the parameter count. Extensive experiments on two challenging nighttime detection benchmarks, BDD100K-Night and City-Night3K, demonstrate the effectiveness of our approach. Compared to strong baselines such as YOLOv8-M, YOLOv12-X, and RT-DETRv2-50, NF-DETR-L achieves improvements of up to +3.5% AP@50 and +3.7% AP@50:95 on BDD100K-Night, and +2.7% AP@50 and +1.9% AP@50:95 on City-Night3K, while maintaining competitive inference speeds. Ablation studies confirm that each proposed component contributes positively to detection performance, with their combination yielding the best results. NF-DETR offers a more robust solution for nighttime perception systems in autonomous driving and surveillance applications, effectively addressing the unique challenges of low-light object detection. Full article

Background/Objectives: Gender differences influence the clinical manifestations, progression, and treatment response in obstructive sleep apnea (OSA) syndrome, suggesting the existence of distinct gender-related phenotypes potentially driven by anatomical, physiological, and hormonal factors. However, the impact of gender on OSA-related cognitive profiles remains unknown. This study aimed to investigate the neuropsychological and polysomnographic (PSG) differences between OSA females and males in order to detect the impact of gender on clinical manifestation and PSG features. Methods: Data were collected from 28 OSA patients (14 females and 14 males matched for age, education, and disease severity). All patients performed a complete neuropsychological evaluation, Epworth sleepiness scale, and whole-night PSG. To evaluate the relationship between specific sleep profiles and cognitive performance, PSG parameters were correlated to scores obtained on neuropsychological tests. Results: Both male and female groups performed within the normal range across all administered neuropsychological tests, according to Italian normative values. Compared with OSA males, female patients showed significantly lower values on the Rey–Osterrieth Complex Figure (ROCF) Recall Test. By contrast, no significant statistical clinical difference emerged between the two OSA groups in terms of clinical manifestation and sleep parameters. Conclusions: This study improves the knowledge on gender-related cognitive impairment in OSA patients. Our preliminary findings demonstrate that the ROCF Recall Test may be altered in OSA females, but not in males. Further longitudinal studies are needed to investigate whether OSA female patients will develop a frank dementia over time. Full article