Deep learning is a subfield of artificial intelligence technology that has progressed to the point where many different AI models that are based on deep learning have been used in the investigation of musculoskeletal conditions. It has also been shown that deep learning is able to aid in the prediction of the outcome of a patient's condition. In the meantime, deep learning has also been the focus of a substantial amount of study in the treatment of a range of illnesses, including breast, prostate, and lung cancers. Deep learning has become more applicable in the study of Soft Tissue cancers over the course of the last several years On the basis of radiological (such as X-ray, CT, MRI, and SPECT) and pathological pictures, an increasing number of deep learning models have demonstrated promising outcomes in the detection, segmentation, classification, volume calculation, grading, and assessment of tumour necrosis rate in primary and metastatic cancers. This shows that deep learning may have the ability to help in the detection of Soft Tissue tumours and to forecast the patient's prognosis in these circumstances. In this research, we started by presenting a description of the processes of deep learning techniques in medical photos. This was followed by a discussion of the contemporary applications of deep learning-based AI for diagnosis and the prediction of prognosis in Soft Tissue cancers.