Fig. 5: Clustering of all detected swallows for a single patient. | Communications Medicine

Fig. 5: Clustering of all detected swallows for a single patient.

From: A deep learning-based approach to enhance accuracy and feasibility of long-term high-resolution manometry examinations

Fig. 5

a The swallows are first clustered into a certain number of classes, depending on the lowest mean intra-cluster distance. We consider all clusters containing 15% or more of all samples to be the main categories of swallows; the others are considered to be special classes. b As clinicians are mainly interested in special cases when looking for intermittently occurring motility disorders, the remaining samples are clustered a second time, in a more granular fashion.

Back to article page