Objective: Ultrasonography (US) is a useful tool for evaluating the activity of rheumatoid arthritis (RA) patients. As the systemic evaluation of many joints is time-consuming, a method to evaluate this activity with a smaller number of joints is needed. The aim of this study was to clarify whether the number of joints assessed may be reduced using patient-oriented joint selection. Methods: A total of 492 RA patients were recruited at Kyoto University Hospital. Bilateral metacarpophalangeal (MCP), (proximal) interphalangeal (PIP/IP), and wrist joints were evaluated by US. Gray scale and power Doppler imaging findings were scored by a 0-3 semi-quantitative method. Clinical assessments were performed by physicians who were blind to US results, and a questionnaire on subjective symptoms was collected from each patient. Results: The correlation between the US score of all 22 joints (US22) and patient-oriented painful joints (PtUS) or physician-oriented tender and/or swollen joints were moderate (Spearman's ρ = 0.435) and weak (ρ = 0.383), respectively. These correlations were weaker than that between the total US score of 5 preselected joints (unilateral 2MCP, 3MCP, 2PIP, 3PIP, and the wrist) and US22 (ρ = 0.813). However, when focusing on patients whose painful joints were 5 and more, the correlation between PtUS and US22 was markedly stronger (ρ = 0.757). Conclusion: Patient-oriented joint selection reflected actual joint inflammation to some extent. However, excessive reductions in the number of joints assessed need to be avoided even if patients do not have arthralgia because of the potential for underestimations.