Journal of Capital Medical University ›› 2019, Vol. 40 ›› Issue (6): 875-880.doi: 10.3969/j.issn.1006-7795.2019.06.013

Previous Articles     Next Articles

A functional laboratory biochemical model for the prediction of pulmonary fibrosis risk

Leng Dong1, Li Guiqin2, Wang Ying1, Miao Ran3, Chen Duo4, Huang Xiaoxi3   

  1. 1. Clinical Laboratory, Beijing Chaoyang Hospital, Capital Medical University, Beijing 100020, China;
    2. Office for Disease Prevention and Control, Chaoyang Hospital, Capital Medical University, Beijing 100020, China;
    3. Medical Research Center, Beijing Chaoyang Hospital, Capital Medical University, Beijing 100020, China;
    4. Department of Respiratory and Critical Care Medicine, Beijing Chaoyang Hospital, Capital Medical University, Beijing 100020, China
  • Received:2019-04-24 Online:2019-11-21 Published:2019-12-18
  • Supported by:
    This study was supported by National Natural Science Foundation of China (81700061,81300049), Beijing Municipal Administration of Hospitals Clinical Medicine Development of Special Funding Support (ZYLX201811), Key Subject Construction Project of China.

Abstract: Objective To determine promising indices for early diagnosis of pulmonary fibrosis (PF) by data extraction and analysis. Methods Biochemical data from patients with PF (n=29) and healthy controls (n=55) were collected and normalized by Z-score indexation and Log2 transformation,followed by principal component and Bayesian probit regression analyses. Signature parameters for PF were identified and used for discriminative function modeling and receiver operating characteristic analysis. Results The α-hydroxybutyric dehydrogenase (HBDH),lactic dehydrogenase,albumin,albumin:globulin ratio,prealbumin,and calcium parameters were significantly different between PF and healthy control samples (P<0.05),and discriminant functions of PF and health were constructed. HBDH was found to be the efficient parameter that could indicate the potential risk of PF. Conclusion This study is attempt to investigate PF serum biochemical characteristics by mining of clinical biochemical data. Discriminant functional model for predicting PF with signature biochemical parameters were constructed successfully.

Key words: pulmonary fibrosis, clinical biochemistry, data modeling

CLC Number: