Cited by Deep Learning Based Encoder-Decoder Model for Skin Cancer Classification

A deep learning-powered diagnostic model for acute pancreatitis DOI

Chi Zhang, Peng Jin, Lu Wang

et al.

BMC Medical Imaging, Journal Year: 2024, Volume and Issue: 24(1)

Published: June 20, 2024

Abstract Background Acute pancreatitis is one of the most common diseases requiring emergency surgery. Rapid and accurate recognition acute can help improve clinical outcomes. This study aimed to develop a deep learning-powered diagnostic model for pancreatitis. Materials methods In this investigation, we enrolled cohort 190 patients with who were admitted Sichuan Provincial People’s Hospital between January 2020 December 2021. Abdominal computed tomography (CT) scans obtained from both healthy individuals. Our was constructed using two modules: (1) classifier module; (2) lesion segmentation module. Each model’s performance assessed based on precision, recall rate, F1-score, Area Under Curve (AUC), loss frequency-weighted accuracy (fwavacc), Mean Intersection over Union (MIOU). Results Upon admission, significant variations observed mild severe in inflammatory indexes, liver, kidney function indicators, as well coagulation parameters. The module exhibited commendable efficacy, showing an impressive AUC 0.993 (95%CI: 0.978–0.999) test set (comprising examination vs. those pancreatitis, P < 0.001) 0.850 0.790–0.898) external validation (healthy 0.001). Furthermore, demonstrated exceptional set. For pancreas segmentation, peripancreatic exudation, effusion, abscess necrosis, MIOU values 86.02 (84.52, 87.20), 61.81 (56.25, 64.83), 57.73 (49.90, 68.23), 66.36 (55.08, 72.12), respectively. These findings underscore robustness reliability developed models accurately characterizing assessing Conclusion driven by learning, exhibits excellent efficacy evaluating severity condition. Trial Registration retrospective study.

Language: Английский

Citations

Editorial for Special Issue “Image Analysis and Machine Learning in Cancers” DOI