Computed tomography (CT) quality control (QC) is regularly performed with standard phantoms, to bar faulty equipment from medical use. Its accuracy may be improved by replacing qualitative methods based on good visual distinction with pixel value-based quantitative methods. We hypothesized that statistical texture analysis (TA) that covers the entire phantom image would be a more appropriate tool. Therefore, our study devised a novel QC method based on the TA for contrast resolution (CR) and spatial resolution (SR) and proposed new, quantitative CT QC criteria. TA of CR and SR images on an American Association of Physicists in Medicine (AAPM) CT Performance Phantom were performed with nine CT scanner models. Six texture descriptors derived from first-order statistics of grayscale image histograms were analyzed. Principal component analysis was used to reveal descriptors with high utility. For CR evaluation, contrast and softness were the most accurate descriptors. For SR evaluation, contrast, softness, and skewness were the most useful descriptors. We propose the following ranges: contrast for CR, 29.5 ± 15%, for SR, 29 ± 10%; softness for CR, <0.015, for SR, −1.85. Our novel TA method may improve the assessment of CR and SR of AAPM phantom images.

Normal body tissue or lesion characteristics in T1 images have been evaluated; however, how external parameters effect the change in signal intensity by gadolinium-based contrast agent remains unknown. We investigated how contrast enhancement changed according to echo time (TE) in 3.0T magnetic resonance (MR) T1 imaging and determined the optimal settings for TE in contrast-enhanced T1 imaging. Since there are no guidelines regarding parameters for T1 enhancement when using MR-contrast agents, we analyzed results from varying TEs (between 25 and 7 msec) in both a phantom and clinical study. We obtained the following results: contrast percentage of fat to saline increased from 740.0–1003.6%, response start point increased from 30–90 mmol, max peak signal intensity increased from 1771–2425 a.u., max peak point increased from 2–4 mmol, enhancement percentage of the max peak signal intensity (MPSI) to saline increased from 1671.0–2065.2%, the average of SI on each mol as TE increased from 600.8–996.6 a.u., the average of SI as TE on each molar concentration increased from 378–845 a.u., the AEPSS increased from 44.3–140.3%, and the AEPSC increased from 224.3–647.8%. We confirmed that TE can affect contrast enhancement, and the lowest TE has faster and higher effects on contrast enhancement.

Computed tomography (CT) is a rapid and precise medical imaging modality, but it poses the challenge of high radiation exposure to patients. To control this issue, stringent quality control (QC) evaluations are imperative for CT. One crucial aspect of CT QC involves the evaluation of phantom images, utilizing specifically designed phantoms for accuracy management and subsequent objective evaluation. However, CT QC has qualitative evaluation methods, particularly for evaluating spatial and contrast resolutions. To solve this problem, we propose a quality control method based on deep-learning object detection for quantitatively evaluating spatial and contrast resolutions, CT Attention You Only Look Once v8 (CTA-YOLOv8). First, we utilized the YOLOv8 network as the foundational model, optimizing it for enhanced accuracy. Second, we enhanced the network’s capabilities by integrating the Convolutional Block Attention Module (CBAM) and Swin Transformers, tailored for phantom image evaluations. The CBAM module was employed internally to pinpoint the optimal position for achieving peak performance in CT QC data. Similarly, we fine-tuned the code and patch size of the Swin Transformer module to align it with YOLOv8, culminating in the identification of the optimal configuration. Our proposed CTA-YOLOv8 network showed superior agreement with qualitative evaluation methods, achieving accuracies of 92.03% and 97.56% for spatial and contrast resolution evaluations, respectively. Thus, we suggest that our method offers nearly equivalent performance to qualitative methods. The utilization of the CTA-YOLOv8 network in evaluating CT phantom images holds potential for setting a new standard in quantitative assessment methodologies.

KSII Transactions on Internet and Information Systems TIIS Paper Details

An on-load tap changer (OLTC) is a critical component of power transformers, and the vibration signals generated during its operation provide valuable information for forecasting equipment conditions and anomalies. In this study, we propose a novel mixture-of-experts-based Kolmogorov–Arnold network (KAN) model, referred to as MoEKAN, to enhance the accuracy of time series forecasting for the vibration signals of the OLTC. The proposed MoEKAN incorporates reversible instance normalization (RevIN) to flexibly adapt to changes in data distribution and employs a transformer-based gating mechanism to dynamically integrate forecasts from various KAN expert models. In addition, multi-scale signal processing is performed to effectively capture the complex periodicity and patterns present in the vibration data. Experiments using real OLTC operational data demonstrate that the MoEKAN model achieves superior forecasting performance, recording forecasting errors with MSE, MAE, and MAPE values of 133.4579, 7.2801, and 4.4272, respectively, outperforming all comparison models. These results validate the practicality and contribution of the proposed model and confirm its potential as a highly reliable diagnostic tool for condition monitoring and predictive maintenance of OLTCs.

The On-Load Tap Changer (OLTC) in power transformers is a critical component responsible for regulating the output voltage, and the early detection of OLTC faults is essential for maintaining power grid stability. In this paper, we propose a one-class deep learning anomaly detection model named “U-AttentionFlow” based on acoustic signals from the OLTC operation. The proposed model is trained exclusively on normal operating data to accurately model normal patterns and identify anomalies when new signals deviate from the learned patterns. To enhance the ability of the model to focus on significant features, we integrate the squeeze-and-excitation (SE) block and Convolutional Block Attention Module (CBAM) into the network architecture. Furthermore, static positional encoding and multihead self-attention (MHSA) are employed to effectively learn the temporal characteristics of time-series acoustic signals. We also adopted a U-Flow-style invertible multiscale coupling structure, which integrates features across multiple scales while ensuring the invertibility of the model. Experimental validation was conducted using acoustic data collected under realistic voltage and load conditions from actual ECOTAP VPD OLTC equipment, resulting in an anomaly detection accuracy of 99.15%. These results demonstrate the outstanding performance and practical applicability of the U-AttentionFlow model for OLTC anomaly detection.