I am working on voice pathology detection using the Saarbrücken Voice Database (SVD) and exploring the possibility of classifying different pathology types into stages (e.g., beginning, mid, severe). I would like to know if there is any established pathological research or medical criteria for staging voice disorders based on acoustic or clinical parameters.

Are there any studies, frameworks, or feature-based classifications (e.g., jitter, shimmer, HNR, MFCC, deep learning-based embeddings) that could be referenced for stage-wise prediction? Any insights or recommendations on how to approach this problem from a voice pathology research perspective would be greatly appreciated!

I did some internet surfing to find the base for determining the stages of pathology but I did not find any research or any medical base to it.