Semi-supervised ensemble studying (SEL) is a practical approach for clinical picture segmentation that combines the benefits of supervised and unsupervised getting-to-know methods. It can be used for clinical image segmentation duties, including improving segmentation accuracy with small annotated datasets. Transfer learning is a technique that leverages information from a related domain to a unique one. In medical image segmentation, transfer getting-to-know methods allow higher segmentation by using suitable function illustrations and parameters of a pertained model. This paper proposes a semi-supervised ensemble getting to know for clinical image segmentation with transfer gaining knowledge. We utilize two supervised techniques and two unsupervised strategies for every man or woman model. All fashions are then combined using an ensemble getting-to-know-the technique. We examine the proposed approach on 3 photo segmentation datasets—the facts from the CHESS mission, ISBI cervical cellular segmentation project, and the ISBI. The outcomes display that our proposed technique outperforms the prevailing strategies with an average cube score of zero.92 ± zero.04 on the CHESS venture, zero. Eighty three ± zero.05 on the ISBI cervical mobile segmentation mission and zero.90 ± 0.05 at the ISBI breast most.