Single-leadSECG monitoring is to facilitate being miniaturized, being used home and used as a Holter. Some of the abnormal electrode patterns, such as electrode loosened, different electrodesSbeing used, etc, whichScan heavily affect the ECG signalSquality, evenSresulting in wrong diagnosis. Artificial identification of the electrode patterns depends on theSperson"s experiences. In this paper, theSautomatic identificationSmethod for four kinds ofSelectrodeSpatterns in single-leadSECG monitoring is studied:S(1) first, 8 digital indices for distinguishing different electrodeSpatterns are defined, (2) second, theSLDA-basedSfeatureSdimension reductionSand theSnearest neighbor classifier are adopted to complete the identification process.SExperiments showSthat the proposedSautomatic identification methodSof different electrode patterns in single leadSECGSmonitoringSwas effective, with 100%Ssuccessful identifying rate for electrode being loosened.