Search Results (86)

Non-contact, automated health monitoring is a cornerstone of modern precision livestock farming, crucial for enhancing animal welfare and productivity. Infrared thermography (IRT) offers a powerful, non-invasive means to assess physiological status. However, its practical use on farms is limited by a key challenge: accurately locating regions of interest (ROIs), like the eyes and face, in the blurry, low-resolution thermal images common in farm settings. To solve this, we developed a new framework called ID-APM, which is designed for robust ROI registration in agriculture. Our method uses a trinocular system and our RAP-CPD algorithm to robustly match features and accurately calculate the target’s 3D position. This 3D information then enables the precise projection of the ROI’s location onto the ambiguous thermal image through inverse disparity estimation, effectively overcoming errors caused by image blur and spectral inconsistencies. Validated on a self-built dataset of farmed sika deer, the ID-APM framework demonstrated exceptional performance. It achieved a remarkable overall accuracy of 96.95% and a Correct Matching Ratio (CMR) of 99.93%. This research provides a robust and automated solution that effectively bypasses the limitations of low-resolution thermal sensors, offering a promising and practical tool for precision health monitoring, early disease detection, and enhanced management of semi-wild farmed animals like sika deer. Full article

Background/Objectives: Diagnosis of Persistent Spinal Pain Syndrome Type 2 (PSPS-T2) currently lacks objective biomarkers. Therefore, this retrospective study aimed to investigate differences in glucose metabolism in the axial musculoskeletal system in PSPS-T2 patients by means of [¹⁸F]FDG PET-CT imaging. Methods: Nine PSPS-T2 patients (five females, four males; mean age of 53 ± 4.82 years) and nine age- and gender-matched healthy controls (five females, four males; mean age of 53 ± 3.91 years) were included. For each participant, 24 regions of interest (ROIs) were manually drawn, including areas of the vertebral endplates, the intervertebral discs, and the psoas muscles. For each ROI, the mean standardized uptake values (SUVs) were assessed. Group differences were evaluated using repeated measures ANOVA with Bonferroni-adjusted post-hoc pairwise comparisons. Additionally, Pearson correlation analyses examined associations between SUV_mean values and the Numerical Rating Scale (NRS) pain scores. Results: Results demonstrated significantly higher SUV_mean values in healthy controls compared to PSPS-T2 patients, particularly at the superior endplates of L4 and S1, the intervertebral discs at L4-L5 and L5-S1, and the posterior endplates of L4 and L5. Although PSPS-T2 patients exhibited higher SUV_mean values than controls in the psoas muscle, these differences were not statistically significant. Additionally, no significant correlations were found between SUV_mean values and NRS pain scores, suggesting that metabolic activity alone does not directly reflect pain severity. Conclusions: Despite the limited sample size of this pilot study, the metabolic fingerprint of the axial musculoskeletal system was shown to be distinctly different in PSPS-T2 patients compared to healthy controls. This could lead to an improved understanding of PSPS-T2 pathophysiology and might open new doors for better diagnosis and treatment strategies. Full article

Lumbar spine Magnetic Resonance Imaging (MRI) is commonly used for intervertebral disc (IVD) and vertebral body (VB) evaluation during low back pain. Segmentation of these tissues can provide useful quantitative information such as shape and volume. The objective of the study was to determine the performances of Segment Anything Model (SAM) and medical SAM (MedSAM), two “zero-shot” deep learning models, in segmenting lumbar IVD and VB from MRI images and compare against the nnU-Net model. This cadaveric study used 82 donor spines. Manual segmentation was performed to serve as ground truth. Two readers processed the spine MRI using SAM and MedSAM by placing points or drawing bounding boxes around regions of interest (ROI). The outputs were compared against ground truths to determine Dice score, sensitivity, and specificity. Qualitatively, results varied but overall, MedSAM produced more consistent results than SAM, but neither matched the performance of nnU-Net. Mean Dice scores for MedSAM were 0.79 for IVDs and 0.88 for VBs, and significantly higher (each p < 0.001) than those for SAM (0.64 for IVDs, 0.83 for VBs). Both were lower compared to nnU-Net (0.99 for IVD and VB). Sensitivity values also favored MedSAM. These results demonstrated the feasibility of “zero-shot” DL models to segment lumbar spine MRI. While performance falls short of recent models, these zero-shot models offer key advantages in not needing training data and faster adaptation to other anatomies and tasks. Validation of a generalizable segmentation model for lumbar spine MRI can lead to more precise diagnostics, follow-up, and enhanced back pain research, with potential cost savings from automated analyses while supporting the broader use of AI and machine learning in healthcare. Full article

Objectives: The purpose of this study was to analyze image quality and radiation exposure of thoracic computed tomography angiography (CTA) in children with congenital heart diseases (CHDs) using either manual contrast medium (CM) injection or automated power injectors with bolus tracking. Methods: A total of 137 thoracic CTAs of 120 consecutive pediatric patients were included in this retrospective study. We analyzed the method of CM administration (power injection with bolus tracking (PI) or manual injection (MI)), injection routes, volumes and flow rates of CM. For the evaluation of objective image quality, attenuation values in the heart chambers and great thoracic vessels were determined by region-of-interest (ROI) analysis and signal-to-noise (SNR) and contrast-to-noise (CNR) ratios calculated thereof. Visual image quality was assessed by two blinded readers (four-point Likert-scale) analyzing the presence of artifacts and the depiction of relevant anatomical structures. Effective radiation doses were calculated with dose length products and specific conversion factors. Results: CM administration was performed using PI in 119/137 CTAs, whereas MI was conducted in 18/137. The smallest size of peripheral venous cannulas was 24 gauge in 36/137 (26.3%) cases. Overall mean CM volume was 17 mL ± 16 mL (mean ± SD). In PI, the mean flow rate of CM was 1.52 ± 0.90 mL/s with a range between 0.5 and 5.0 mL/s. When comparing the overall PI population and an age-, size- and weight-matched PI subpopulation (18 cases) with the MI population, attenuation values in Hounsfield units (HU) and CNR values were significantly higher in the PI groups than in the MI group for each relevant cardiac structure (left ventricle, right ventricle, ascending aorta and pulmonary trunk, p = 0.02–0.001). Overall image quality and depiction of cardiac structures were rated significantly better in CTAs with PI (interquartile ranges: “good” to “excellent” (Likert 3–4)) in PI compared with CTAs acquired with MI (interquartile ranges: “fair” to “good” (2–3)) in MI by both readers (p < 0.001). The inter-observer reliability was strong, with a Kendall’s Tau-b correlation coefficient of τ = 0.802 (p < 0.001). The mean effective radiation dose (E) did not differ significantly when comparing the stratified samples (i.e., the matched PI subgroup and the MI group; 0.5 (±0.3) mSv in both, p = 0.76). There were no complications associated with the CM injections for both application approaches. Conclusions: Automated contrast agent applications with power injectors and bolus tracking ensure better image quality in pediatric CTA, even when low volumes and flow rates need to be applied. There is a slight increase in radiation associated with bolus tracking. This approach represents a suitable imaging technique for the work-up of congenital heart disease. Full article

Addressing the challenge of manual dependency and the difficulty in automating the online detection of height discrepancies following engine oil seal assembly, this paper proposes a composite vision-based method for the post-assembly size inspection of engine oil seals. The proposed method enables non-contact, online three-dimensional measurement of oil seals already installed on the engine. To achieve accurate positioning of the inner and outer ring regions of the oil seals, the process begins with obtaining the center point and the major and minor axes through ellipse fitting, which is performed using progressive template matching and the least squares method. After scaling the ellipse along its axes, the preprocessed image is segmented using the peak–valley thresholding method to generate an annular ROI (region of interest) mask, thereby reducing the complexity of the image. By integrating three-frequency four-step phase-shifting profilometry with an improved RANSAC (random sample consensus)-based plane fitting algorithm, the height difference between the inner and outer rings as well as the press-in depth are accurately calculated, effectively eliminating interference from non-target regions. Experimental results demonstrate that the proposed method significantly outperforms traditional manual measurement in terms of speed, with the relative deviations of the height difference and press-in depth confined within 0.33% and 1.45%, respectively, and a detection success rate of 96.35% over 1415 samples. Compared with existing methods, the proposed approach not only enhances detection accuracy and efficiency but also provides a practical and reliable solution for real-time monitoring of engine oil seal assembly dimensions, highlighting its substantial industrial application potential. Full article

Background/Objectives: Cochlear implantation is the primary treatment for severe-to-profound hearing loss, yet outcomes vary significantly among recipients. While visual–auditory cross-modal reorganization has been identified as a contributing factor to this variability, its impact in early-implanted children remains unclear. To address this knowledge gap, we investigated visual processing and its relationship with auditory outcomes in children who received early bilateral cochlear implants. Methods: To examine potential cross-modal reorganization, we recorded visual evoked potentials (VEPs) in response to pattern-reversal stimuli in 25 children with cochlear implants (CIs) (mean implantation age: 1.44 years) and 28 age-matched normal-hearing (NH) controls. Analysis focused on both the occipital region of interest (ROI: O1, OZ, and O2 electrode sites) and right temporal ROI, examining VEP components and their correlation with speech perception outcomes. Results: Unlike previous studies in later-implanted children, the overall occipital ROI showed no significant differences between groups. However, the left occipital electrode (O1) revealed reduced P1 amplitudes and delayed N1 latencies in CI users. Importantly, O1 N1 latency negatively correlated with speech-in-noise performance (r = −0.318; p = 0.02). The right temporal region showed no significant differences in VEP N1 between groups and no correlation with speech performance in CI users. Conclusions: Early bilateral cochlear implantation appears to preserve global visual processing, suggesting minimal maladaptive reorganization. However, subtle alterations in left occipital visual processing may influence auditory outcomes, highlighting the importance of early intervention and the complex nature of sensory integration in this population. Full article

An important goal in the reform of architectural design education is to instruct students in ways of acquiring relevant site information quickly and efficiently during a design project, and then integrating that information into their architectural designs. This study focuses on a teaching experiment conducted within the “Urban Village Renovation Design” course for third-year undergraduates at Zhejiang University. This study aims to improve teaching efficiency by combining active and passive information acquisition methods during the site information acquisition stage. A teaching experiment on “Urban Village Renovation Design” was conducted with third-year undergraduates at Zhejiang University, comparing two experimental groups based on whether the teacher provides site information reports (i.e., passive information acquisition). The study explores efficient methods for acquiring different types of site information in architectural design teaching and develops a matching framework. It evaluates the impact of active vs. passive methods on students’ cognitive levels, using Bloom’s taxonomy, and quantitatively tests cognitive efficiency differences through the ROI model. Results show that combining both methods yields the highest teaching efficiency, with specific types of information corresponding to effective active or passive acquisition methods. This study explores which research methods can yield beneficial site information more efficiently and clarifies the role of previously overlooked passive information acquisition methods in site cognition, providing theoretical support for the design of teaching plans during the research phase. From a practical standpoint, it is suggested that instructors provide certain site information directly rather than have students acquire it independently, to shorten the research phase of teaching and simultaneously enhance site cognition efficiency. Full article

Background/Objectives: Recent studies indicate extensive shared white matter (WM) abnormalities between bipolar disorder (BD) and schizophrenia (SZ). However, the heterogeneity of WM in BD in terms of the presence of psychosis remains a critical issue for exploring the boundaries between BD and SZ. Previous studies comparing WM microstructures in psychotic and nonpsychotic BDs (PBD and NPBD) have resulted in limited findings, probably due to subtle changes, emphasizing the need for further investigation. Methods: Diffusion tensor imaging measures were obtained from 8 individuals with PBD, 8 with NPBD, and 22 healthy controls (HC), matched for age, gender, handedness, and educational years. Group comparisons were conducted using tract-based spatial statistics (TBSS). The most significant voxels showing differences between PBD and HC in the TBSS analyses were defined as a TBSS-ROI and subsequently analyzed. Results: Increased radial diffusivity (RD) in PBD compared to NPBD (p < 0.006; d = 1.706) was observed in TBSS-ROI, distributed in the confined regions of some WM tracts, including the body of the corpus callosum (bCC), the left genu of the CC (gCC), and the anterior and superior corona radiata (ACR and SCR). Additionally, NPBD exhibited significant age-associated RD increases (R² = 0.822, p < 0.001), whereas the greater RD observed in PBD compared to NPBD remained consistent across middle age. Conclusions: Preliminary findings from this small sample suggest severe frontal WM disconnection in the anterior interhemispheric communication, left fronto-limbic circuits, and cortico-striatal-thalamic loop in PBD compared to NPBD. While these results require replication and validation in larger and controlled samples, they provide insights into the pathophysiology of PBD, which is diagnostically located at the boundary between BD and SZ. Full article

Using deep learning or artificial intelligence (AI) in research with animals is a new interdisciplinary area of research. In this study, we have explored the potential of thermal imaging and AI in pig research. Thermal cameras play a vital role in obtaining and collecting a large amount of data, and AI has the capabilities of processing and extracting valuable information from these data. The amount of data collected using thermal imaging is huge, and automation techniques are therefore crucial to find a meaningful interpretation of the changes in temperature. In this paper, we present a complete pipeline to extract temperature automatically from a selected Region of Interest (ROI). This system consists of three stages: the first one checks whether the ROI is completely visible to observe the thermal temperature, and then the second stage uses an encoder–decoder structure of a convolution neural network to segment the ROI, if the condition was met at stage one. In the last stage, the maximum temperature is extracted and saved in an external file. The segmentation model showed good performance, with a mean Pixel Class accuracy of 92.3%, and a mean Intersection over Union of 87.1%. The extracted temperature observed by the model entirely matched the manually observed temperature. The system showed reliable results to be used independently without human intervention to determine the temperature in the selected ROI in pigs. Full article

This study evaluates the feasibility of using Haralick texture analysis on low-field, T2-weighted MRI images for detecting prostate cancer, extending current research from high-field MRI to the more accessible and cost-effective low-field MRI. A total of twenty-one patients with biopsy-proven prostate cancer (Gleason score 4+3 or higher) were included. Before transperineal biopsy guided by low-field (58–74mT) MRI, a radiologist annotated suspicious regions of interest (ROIs) on high-field (3T) MRI. Rigid image registration was performed to align corresponding regions on both high- and low-field images, ensuring an accurate propagation of annotations to the co-registered low-field images for texture feature calculations. For each cancerous ROI, a matching ROI of identical size was drawn in a non-suspicious region presumed to be normal tissue. Four Haralick texture features (Energy, Correlation, Contrast, and Homogeneity) were extracted and compared between cancerous and non-suspicious ROIs. Two extraction methods were used: the direct computation of texture measures within the ROIs and a sliding window technique generating texture maps across the prostate from which average values were derived. The results demonstrated statistically significant differences in texture features between cancerous and non-suspicious regions. Specifically, Energy and Homogeneity were elevated (p-values: <0.00001–0.004), while Contrast and Correlation were reduced (p-values: <0.00001–0.03) in cancerous ROIs. These findings suggest that Haralick texture features are both feasible and informative for differentiating abnormalities, offering promise in assisting prostate cancer detection on low-field MRI. Full article

The positioning of lithium battery tabs in electric vehicles is a crucial aspect of the power battery assembly process. During the pre-tightening process of the lithium battery stack assembly, cells and foams undergo different deformations, leading to varying displacements of cells at different levels. Consequently, determining tab positions poses numerous challenges during the pre-tightening process of the stack assembly. To address these challenges, this paper proposes a method for detecting feature points and calculating the displacement of lithium battery stack tabs based on the MicKey method. This research focuses on the cell tab, utilizing the hue, saturation, and value (HSV) color space for image segmentation to adaptively extract the cell tab region and further obtain the ROI of the cell tab. In order to enhance the accuracy of tab displacement calculation, a novel method for feature point detection and displacement calculation of lithium battery stacks based on the MicKey (Metric Keypoints) method is introduced. MicKey can predict the coordinates of corresponding keypoints in the 3D camera space through keypoint matching based on neural networks, and it can acquire feature point pairs of the subject to be measured through its unique depth reduction characteristics. Results demonstrate that the average displacement error and root mean square error of this method are 0.03 mm and 0.04 mm, respectively. Compared to other feature matching algorithms, this method can more consistently and accurately detect feature points and calculate displacements, meeting the positioning accuracy requirements for the stack pole ear in the actual assembly process. It provides a theoretical foundation for subsequent procedures. Full article

The integration of Internet of Medical Things (IoMT) technology has revolutionized healthcare, allowing rapid access to medical images and enhancing remote diagnostics in telemedicine. However, this advancement raises serious cybersecurity concerns, particularly regarding unauthorized access and data integrity. This paper presents a novel, user-friendly, visible watermarking scheme for medical images—Visual and User-Friendly Medical Image Watermarking Scheme (VUF-MIWS)—designed to secure medical image ownership while maintaining usability for diagnostic purposes. VUF-MIWS employs a unique combination of inpainting and data hiding techniques to embed hospital logos as visible watermarks, which can be removed seamlessly once image authenticity is verified, restoring the image to its original state. Experimental results demonstrate the scheme’s robust performance, with the watermarking process preserving critical diagnostic information with high fidelity. The method achieved Peak Signal-to-Noise Ratios (PSNR) above 70 dB and Structural Similarity Index Measures (SSIM) of 0.99 for inpainted images, indicating minimal loss of image quality. Additionally, VUF-MIWS effectively restored the ROI region of medical images post-watermark removal, as verified through test cases with restored watermarked regions matching the original images. These findings affirm VUF-MIWS’s suitability for secure telemedicine applications. Full article

Ultrasound imaging provides real-time guidance during needle interventions, but localizing the needle in ultrasound videos remains a challenging task. This paper introduces a novel machine learning-based method to localize the needle in ultrasound videos. The method comprises three phases for analyzing the image frames of the ultrasound video and localizing the needle in each image frame. The first phase aims to extract features that quantify the speckle variations associated with needle insertion, the edges that match the needle orientation, and the pixel intensity statistics of the ultrasound image. The features are analyzed using a machine learning classifier to generate a quantitative image that characterizes the pixels associated with the needle. In the second phase, the quantitative image is processed to identify the region of interest (ROI) that contains the needle. In the third phase, the ROI is processed using a custom-made Ranklet transform to accurately estimate the needle trajectory. Moreover, the needle tip is identified using a sliding window approach that analyzes the speckle variations along the needle trajectory. The performance of the proposed method was evaluated by localizing the needle in ex vivo and in vivo ultrasound videos. The results show that the proposed method was able to localize the needle with failure rates of 0%. The angular, axis, and tip errors computed for the ex vivo ultrasound videos are within the ranges of 0.3–0.7°, 0.2–0.7 mm, and 0.4–0.8 mm, respectively. Additionally, the angular, axis, and tip errors computed for the in vivo ultrasound videos are within the ranges of 0.2–1.0°, 0.3–1.0 mm, and 0.3–1.1 mm, respectively. A key advantage of the proposed method is the ability to achieve accurate localization of the needle without altering the clinical workflow of the intervention. Full article

Background: Recent studies have described unique aspects of default mode network connectivity in patients with idiopathic generalized epilepsy (IGE). A complete background in this field could be gained by combining this research with spectral analysis. Objectives: An important objective of this study was to compare linear connectivity and power spectral densities across different activity bands of patients with juvenile absence epilepsy (JAE), juvenile myoclonic epilepsy (JME), generalized tonic–clonic seizures alone (EGTCSA), and drug-resistant IGE (DR-IGE) with healthy, age-matched controls. Methods: This was an observational case–control study. We performed EEG spectral analysis in MATLAB and connectivity analysis with LORETA for 39 patients with IGE and 12 drug-resistant IGE (DR-IGE) and healthy, age-matched subjects. We defined regions of interest (ROIs) from the default mode network (DMN) and performed connectivity statistics using time-varying spectra for paired samples. Using the same EEG data, we compared mean power spectral density (PSD) with epilepsy subgroups and controls across different activity bands. Results: We obtained a modified value for the mean power spectral density in the beta band for the JME group as follows. The connectivity analysis showed that, in general, there was increased linear connectivity in the DMN for the JAE, JME, and EGCTSA groups compared to the healthy controls. Reduced linear connectivity between regions of the DMN was found for DR-IGE. Conclusions: Spectral analysis of electroencephalography (EEG) for generalized epilepsy syndromes seems to be less informative than connectivity analysis for DMN. DMN connectivity analysis, especially for DR-IGE, opens up the possibility of finding biomarkers related to drug response in IGE. Full article

Bisphophonates (BPs) are widely used in Osteogenesis imperfecta (OI). Cone Beam Computed Tomography (CBCT) shows clinical usefulness in evaluating impacted teeth and adjacent structure relationships, extraction socket healing, bone mineral density (BMD) and BP-related jaw osteonecrosis (BRONJ). The aim of the study was to compare alveolar sockets and the adjacent bone area before and after third molar extraction in OI type I (OI-I) adolescents treated with BPs and age-matched healthy subjects (HSs) by CBCT. Methods: Five adolescents with genetically proven OI-I treated with BPs (three males and two females, mean age: 15.2 ± 1.78 years) and four age-matched healthy subjects (two males and two females, mean age: 15.5 ± 1.29 years) were included in this study. Eight Regions of Interest (ROIs) were evaluated: between 3.7 and 3.8 (ROI-1) and 4.7 and 4.8 (ROI-2); after 3.8 (ROI-3) and 4.8 (ROI-4); alveolar sockets 3.8 (ROI-5) and 4.8 (ROI-6); left (ROI-7) and right (ROI-8) cortical bone. Results: ROIs were evaluated on both sides of the mandible for all the subjects except one OI patient in which CBCTs were performed pre- and-post third molar extraction only on the right side. CBCT was performed 12.8 ± 4.60 and 11.5 ± 2.51 days before and 8.0 ± 1.41 and 7.7 ± 0.5 months after extraction in OI-I and HSs, respectively. BPs were discontinued 62.0 ± 36.5 months before extraction. None of the OI-I adolescents developed BRONJ. Statistically significant greater values were observed in OI-I for ROI-1 and -2 (p = 0.0464), ROI-3 and -4 (p = 0.0037) and ROI-7 and -8 (p = 0.0079) after extraction. Conclusions: This descriptive study confirms that, in OI-I adolescents treated with BPs, third molar extraction is safe, and socket healing occurs properly. In addition, it demonstrates that, if the same device and imaging conditions are used and comparisons to predetermined standard values are avoided, CBCT can be used to monitor BMD changes. The significant greater BMD observed for different ROIs in OI-I could reflect the increased secondary mineralization related to the BP-dependent reduction in bone turnover. Full article