We propose HELEN, a new code-based public-key cryptosystem whose security is based on the hardness of the Learning from Parity with Noise problem~(LPN) and the decisional minimum distance problem. We show that the resulting cryptosystem achieves indistinguishability under chosen plaintext attacks (IND-CPA security). Using the Fujisaki-Okamoto generic construction, HELEN achieves IND-CCA security in the random oracle model. We further propose concrete parameters.