- 著者
- Zhuo Zhao Jing Bai Chang Liu Yansong Wang Shuang Wang Furong Zhao Qiufang Gu
- 出版者
- SOCIETY FOR FREE RADICAL RESEARCH JAPAN
- 雑誌
- Journal of Clinical Biochemistry and Nutrition (ISSN:09120009)
- 巻号頁・発行日
- vol.73, no.2, pp.161-171, 2023 (Released:2023-09-01)
- 参考文献数
- 51
Metabolic differences between colorectal cancer (CRC) and NI (NI) play an important role in early diagnoses and in-time treatments. We investigated the metabolic alterations between CRC patients and NI, and identified some potential biomarkers, and these biomarkers might be used as indicators for diagnosis of CRC. In this study, there were 79 NI, 50 CRC I patients, 52 CRC II patients, 56 CRC III patients, and 52 CRC IV patients. MS-MS was used to measure the metabolic alterations. Univariate and multivariate data analysis and metabolic pathway analysis were applied to analyze metabolic data and determine differential metabolites. These indicators revealed that amino acid and fatty acids could separate these groups. Several metabolites indicated an excellent variables capability in the separation of CRC patients and NI. Ornithine, arginine, octadecanoyl carnitine, palmitoyl carnitine, adipoyl carnitine, and butyryl carnitine/propanoyl carnitine were selected to distinguish the CRC patients and NI. And methionine and propanoyl carnitine, were directly linked to different stages of CRC. Receiver operating characteristics curves and variables importance in projection both represented an excellent performance of these metabolites. In conclusion, we assessed the difference between CRC patients and NI, which supports guidelines for an early diagnosis and effective treatment.
- 著者
- Shuang WANG Hui CHEN Lei DING He SUI Jianli DING
- 出版者
- The Institute of Electronics, Information and Communication Engineers
- 雑誌
- IEICE TRANSACTIONS on Information and Systems (ISSN:09168532)
- 巻号頁・発行日
- vol.E106-D, no.7, pp.1209-1218, 2023-07-01
- 被引用文献数
- 1
The issue of a low minority class identification rate caused by data imbalance in anomaly detection tasks is addressed by the proposal of a GAN-SR-based intrusion detection model for industrial control systems. First, to correct the imbalance of minority classes in the dataset, a generative adversarial network (GAN) processes the dataset to reconstruct new minority class training samples accordingly. Second, high-dimensional feature extraction is completed using stacked asymmetric depth self-encoder to address the issues of low reconstruction error and lengthy training times. After that, a random forest (RF) decision tree is built, and intrusion detection is carried out using the features that SNDAE retrieved. According to experimental validation on the UNSW-NB15, SWaT and Gas Pipeline datasets, the GAN-SR model outperforms SNDAE-SVM and SNDAE-KNN in terms of detection performance and stability.