Dysferlinopathy is a muscle disease characterized by a variable clinical presentation and is caused by mutations in the DYSF gene. The Jain Clinical Outcome Study for Dysferlinopathy (COS) followed the largest cohort of patients (n=187) with genetically confirmed dysferlinopathy throughout a three-year natural history study, in which the patients underwent muscle function tests and muscle magnetic resonance imaging (MRI). We previously described the pattern of muscle pathology in this population and established a series of imaging criteria for diagnosis. In this paper, we describe the muscle imaging and clinical features of a subgroup of COS participants whose muscle imaging results did not completely meet the diagnostic criteria. We reviewed 184 T1-weighted (T1w) muscle MRI scans obtained at the baseline visit of the COS study, of which 106 were pelvic and lower limb only and 78 were whole-body scans. We identified 116 of the 184 patients (63%) who did not meet at least one of the established imaging criteria. The highest number found of unmet criteria was four per patient. We identified 24 patients (13%) who did not meet three or more of the nine established criteria and considered them as “outliers”. The most common unmet criterion (27.3% of cases) was the adductor magnus being equally or more affected than the adductor longus. We compared the genetic, demographic, clinical and muscle function data of the outlier patients with those who met the established criteria and observed that the outlier patients had an age of disease onset that was significantly older than the whole group (29.3 vs 20.5 years, p=0.0001). This study expands the phenotypic muscle imaging spectrum of patients with dysferlinopathy and can help to guide the diagnostic process in patients with limb girdle weakness of unknown origin.