ABSTRACT
Accurately detecting defects while reconstructing a high-quality normal background in surface defect detection using unsupervised methods remains a significant challenge. This study proposes an unsupervised method that effectively addresses this challenge by achieving both accurate defect detection and a high-quality normal background reconstruction without noise. We propose an adaptive weighted structural similarity (AW-SSIM) loss for focused feature learning. AW-SSIM improves structural similarity (SSIM) loss by assigning different weights to its sub-functions of luminance, contrast, and structure based on their relative importance for a specific training sample. Moreover, it dynamically adjusts the Gaussian window's standard deviation (σ) during loss calculation to balance noise reduction and detail preservation. An artificial defect generation algorithm (ADGA) is proposed to generate an artificial defect closely resembling real ones. We use a two-stage training strategy. In the first stage, the model trains only on normal samples using AW-SSIM loss, allowing it to learn robust representations of normal features. In the second stage of training, the weights obtained from the first stage are used to train the model on both normal and artificially defective training samples. Additionally, the second stage employs a combined learned Perceptual Image Patch Similarity (LPIPS) and AW-SSIM loss. The combined loss helps the model in achieving high-quality normal background reconstruction while maintaining accurate defect detection. Extensive experimental results demonstrate that our proposed method achieves a state-of-the-art defect detection accuracy. The proposed method achieved an average area under the receiver operating characteristic curve (AuROC) of 97.69% on six samples from the MVTec anomaly detection dataset.
ABSTRACT
The patient, a 20-year-old nulligravida, presented with a progressive swelling on the right side of her vulva over the past three years. She also reported dull, aching pain, particularly during long walks. Upon examination, the patient appeared comfortable, and her vital signs were within normal limits. Genitourinary examination revealed a right vulvar mass, 12 cm by 7 cm, which was movable, soft in consistency, and non-tender. Subsequent inguinovulvar ultrasonography confirmed the presence of a compressible soft-tissue mass measuring 10 × 7 × 4 cm, consistent with a diagnosis of lipoma. Surgical excision resulted in a significant improvement in the patient's condition and she was discharged from hospital.
ABSTRACT
BACKGROUND: Anemia is still one of the major public health problems in many developing countries including Ethiopia. Thus, this study aimed to assess individual and contextual-level factors associated with iron-folic acid supplement intake during pregnancy in Ethiopia. METHODS: A secondary analysis was done on the 2019 mini-Ethiopian Demographic and Health Survey (EDHS) dataset. A total of 3,927 pregnant women who gave birth five years before the survey were included in the analysis. Multi-level mixed-effect logistic regression analysis was done by STATA/SE version 14.0 to identify individual and contextual-level factors. Adjusted Odds Ratio (AOR) with 95% Confidence Interval (CI) was used to show the strength and direction of the association. The level of statistical significance was declared at a P value less than 0.05. RESULTS: Those primary educated [AOR = 1.83, 95% CI: (1.24, 2.74)], secondary educated [AOR = 2.75, 95% CI: (1.57, 4.824)], women who had greater than 5 living children [AOR = 2.02, 95% CI: (1.25, 3.27)], women who had ANC visit [AOR = 21.26, 95% CI: (13.56, 33.32)] and women who lived in a cluster with high proportion of women had ANC visit [AOR = 1.72, 95% CI: (1.17, 2.54)] and women who lived in Somali [AOR = 0.44 0.73, 95% CI: (0.22, 0.87)] were significantly associated with iron-folic acid intake during pregnancy. CONCLUSIONS: Both individual and contextual-level factors were significantly associated with iron-folic acid intake during pregnancy. From individual-level factors: education status of women, the total numbers of living children, and ANC follow-up are significant and from contextual-level factors: region and living in a high proportion of women who had ANC follow-up were found to have a statistically significant association. Promoting women's education and maternal health services like ANC and intervention targeting the Somali region would be the recalled area of the government.
