The one I found interesting is this one:
Pain Profile Discovery In Urologic Chronic Pelvic Pain Syndrome (Ucpps): Consensus Clustering Findings From The Mapp Research Network
INTRODUCTION AND OBJECTIVES:
Defining clinically relevant patient phenotypes with differential outcomes is essential for advancement of precision medicine approaches to UCPPS treatment. We performed consensus clustering (CC) using Brief Pain Inventory (BPI) body map data collected in the MAPP Research Network Epidemiology and Phenotyping Study (EPS).
METHODS:
Baseline body map data from 424 participants (233 women; 191 men) with UCPPS (i.e., interstitial cystitis/ bladder pain syndrome or chronic prostatitis/ chronic pelvic pain syndrome) in the MAPP EPS were analyzed using CC. Item clusters of 45 body map sites were utilized to generate patient clusters with distinctly different pain profiles.
RESULTS:
At baseline, the most commonly reported body sites with pain were the pelvis (85%), lower back (39%), lower abdomen (33%), front of the head (27%) and back of the neck (22%) (Fig 1). Using 1,000 replications of k-means within CC, K=4 clusters emerged (Fig 2). Cluster 1 patients (n=108) reported only pelvic region pain (PP). Cluster 2 patients (n=140) reported PP, minimal sites beyond PP, but 0% endorsed lower back pain. Cluster 3 patients (n=134) reported PP, minimal sites beyond PP, but 100% endorsed lower back pain. Cluster 4 patients (n=42) reported PP & widespread pain beyond PP. Chronic Overlapping Pain Conditions (IBS, FM, CFS, Migraine, TMJ) ranged from 30% (Cluster 1) to 91% (Cluster 4). Furthermore, Cluster 1 (PP only) patients were twice as likely to improve (p<0.01) in severity of non-urological pain, compared to Cluster 2-4 patients.
CONCLUSIONS:
These results hold great promise for scaling CC methods to expanded symptom domains to produce stable clusters that shed new insights into UCPPS patient subtypes.