A novel technique has been proposed for the detection of ischemia in ECG signals based on statistical features obtained from ST deviations in ECG signals. Wavelet transform has been used for preprocessing and ECG features delineation. Then mean thresholds for ST segment deviations are used to classify the ischemic beats from normal beats. The ischemia recognition from enduring ST segments is made through the coefficient of variation (COV), kurtosis and form factor. The algorithms are implemented in MATLAB 2012a and IBM SPSS 22 is used for statistical analysis. The results show average sensitivity  98.76% and positive predictivity (+P) 98.51% for 82,357 ST segments of 40 arbitrarily chosen records of annotated European ST-T database (EDB) after validation. These results are significantly better than the available methods in the literature.