Using a recent convex formulation of IBM Model 2, we propose a new initialization scheme which has some favorable comparisons to the standard method of initializing IBM Model 2 with IBM Model 1. Additionally, we derive the Viterbi alignment for the convex relaxation of IBM Model 2 and show that it leads to better F-Measure scores than those of IBM Model 2.

BACKGROUND Inclusion body myositis (IBM) is a slowly progressive inflammatory myopathy of the elderly that does not show significant clinical improvement in response to steroid therapy. Distinguishing IBM from polymyositis (PM) is clinically important since PM is steroid-responsive; however, the two conditions can show substantial histologic overlap. RESULTS We performed quantitative immunohi...

OBJECTIVE To correlate muscle biopsy findings with prebiopsy and postbiopsy clinical course and response to therapy in polymyositis (PM) and sporadic inclusion body myositis (IBM). BACKGROUND Existence of pure PM has recently been questioned; subsequently, the definition and criteria for diagnosing PM were debated. METHODS Patient records, follow-up information, and muscle biopsies were ana...