Human exhaled gas excreting through alveolar exchange is considered as the top air of blood, which reflects the endogenous metabolism of human body to some extent. In recent years, with the development of non-invasive medical diagnosis, breath analysis, as one of the least invasive and painless technique, attracts more and more attention when applying to clinical early diagnosis and monitoring of diseases. Modern analytical methods (gas chromatography, mass spectrometry, optical spectrum, sensor arrays) make diagnosis of diseases based on human exhaled gases possible, with sufficient accuracy and precision for qualitative and quantitative analysis of thousands of trace components in breath. Researches on human exhaled breath samples indicate that there are correlations existing between volatile organic components (VOCs) in breath and the metabolic status of diseases, for instance, breath acetone diabetes, aldehydes and breast cancer, alkanes (ethane, pentane) and oxidative stress. This paper gives an overview of present techniques used for human exhaled breath sampling and pre-concentration. The qualitative and quantitative analytical techniques using GC-MS, SIFT-MS (selected ion flow tube mass spectrometry), PTR-MS (proton transfer reaction mass spectrometry), and optical spectrum and chemical sensors are also summarized. In this review, the potential biomarkers in human exhaled breath for diabetes, breast cancer, and lung cancer are discussed in the second part. Furthermore, the present situation of breath analysis in China and key issues existing in breath analysis are also given some attentions in this paper.

Contents
1 Introduction
2 Analytical techniques
2.1 Sampling
2.2 Pre-concentration
2.3 Qualitative and quantitative analysis
3 Potential applications in non-invasive clinical diagnosis
3.1 Diabetes
3.2 Breast cancer
3.3 Lung diseases
3.4 Oxidative stress
4 Conclusion and outlook