Objective：This study aims to investigate the value of detecting driver gene mutations in non-small cell lung cancer（NSCLC） patients using next-generation sequencing（NGS） technology. Methods：Somatic mutations in 150 NSCLC patients including EGFR，KRAS，BRAF，NRAS，Her-2 and PIK3CA were detected by Ion torrent personal genome machine（PGM）. Sanger sequencing and ddPCR was used to test and verify the results of EGFR gene mutations. Results：According to NGS results，mutations were detected in EGFR（51.33%，77/150），KRAS（7.33%，11/150），BRAF（1.33%，2/150），NRAS（1.33%，2/150），Her-2（2%，3/150）and PIK3CA（4.67%，7/150）. There were 57 samples without any somatic mutations in all genes，84 samples had one or more mutations in single gene，while 9 samples harboured mutations in two or more genes. The overall detection rate of EGFR mutation by NGS and Sanger sequencing was 51.33%（77/150）and 38.67%（58/150）. The difference between the two methods was statistically significant（χ2=4.862，P=0.027）. Compared to Sanger sequencing，the sensitivity and specificity of NGS assays was 98.28% and 78.26%，respectively. The overall detection rate of EGFR mutation by NGS and ddPCR was 51.33%（77/150）and 50%（75/150）. There was no significant difference between the two methods（χ2=0.053，P=0.818）. Compared to ddPCR，the positive concordance rate was 82.67%，the negative consistency rate was 80%，and the total concordance rate was 81.33%. Conclusion：Next-generation sequencing technology is more suitable for the diagnosis and treatment of NSCLC. Sanger sequencing and ddPCR can be useful supplements.