Lossless image coding is an essential requirement for medical imaging appli
cations. Lossless image compression techniques usually have two major compo
nents: adaptive prediction and adaptive entropy coding. The paper is concer
ned with adaptive prediction. Recently, several researchers have studied pr
ediction schemes in which the final prediction is formed by a combination o
f a group of subpredictors. The authors present an overview of this new typ
e of prediction technique. They show that the basic principle of adaptive p
redictor combination has been extensively studied and applied to many scien
ce and engineering problems. They then describe their own combination schem
e, which is based on the estimation of the local prediction error variance.
Experimental results show that the compression performance of the algorith
ms that employ this new type of predictor is consistently better than that
of state-of-the-art algorithms.