Metastatic spine cancer can cause pain and neurological issues, making it challenging to distinguish from spinal compression fractures using magnetic resonance imaging (MRI). To improve diagnostic accuracy, this study developed artificial intelligence (AI) models to differentiate between metastatic spine cancer and spinal compression fractures in MRI images. MRI data from Gyeongsang National University Hospital, collected from January 2019 to April 2022, were processed using Otsu’s binarization and Canny edge detection algorithms. Using these preprocessed datasets, convolutional neural network (CNN) and support vector machine (SVM) models were built. The T1-weighted image-based CNN model demonstrated high sensitivity (1.00) and accuracy (0.98) in identifying metastatic spine cancer, particularly with data processed by Otsu’s binarization and Canny edge detection, achieving exceptional performance in detecting cancerous cases. This approach highlights the potential of preprocessed MRI data for AI-assisted diagnosis, supporting clinical applications in distinguishing metastatic spine cancer from spinal compression fractures.

This study aimed to analyze the epidemiological characteristics of Vibrio vulnificus (V. vulnificus ) infections in the western Gyeongsangnam-do region and highlight their clinical implications. A retrospective observational study was conducted among 70 patients with culture-confirmed V. vulnificus infection between 2006 and 2023. Demographic and temporal characteristics were analyzed in the entire cohort, whereas detailed clinical characteristics, laboratory findings, treatment variables, and mortality outcomes were analyzed in 47 patients treated from 2010 onward, for whom detailed electronic medical records were available. Of the 70 patients, 53 (75.7%) were male, and 3 (4.3%) were < 40 years. The mean monthly number of cases was 0.278, 0.402, and 0.244 cases/month during El Niño, La Niña, and neutral conditions, respectively. Monthly infection counts tended to increase as MEI.v2 decreased toward La Niña conditions (IRR, 0.71 per 1-unit increase; 95% CI, 0.50–1.02; P = .064), although the association was not statistically significant. Among the 47 patients with electronic medical records, 32 (68.1%) had underlying diseases. The case fatality rate (CFR) among the 46 patients was 50.0% (n = 23). Patients with an initial systolic blood pressure (SBP) ≤ 90 mm Hg had a significantly higher CFR than those with SBP > 90 mm Hg (81.3% vs 33.3%, P = .005). Mortality differed significantly among the three WBC categories of < 4000/μL, 4000–10,000/μL, and > 10,000/μL (61.5% vs 66.7% vs 20.0%, P = .022). Patients with elevated creatinine levels (≥2.0 mg/dL) had significantly higher CFR than those with lower creatinine levels (<2.0 mg/dL) (73.7% vs 33.3%, P = .016). Among the 34 patients who presented with skin lesions and/or extremity pain, those who did not undergo surgery had a significantly higher CFR than those who underwent surgery (88.9% [8/9] vs 36.0% [9/25], P = .017). This study identified clinical factors associated with mortality and an exploratory relationship between ENSO variability and V. vulnificus occurrence.