Use of statistical process control to study the patient-specific quality assurance in head and neck cancer

Figure 1 Individual-moving range for the first 20 patients. A-D: Individual-moving range charts for the first 20 patients evaluated using global gamma criteria of 3%/3 mm, 3%/2 mm, 2%/3 mm, and 2%/2 mm, respectively. I-MR: Individual-moving range; UCL: Upper control limit; LCL: Lower control limit.

Figure 2 Exponentially weighted moving average. A-D: Exponentially weighted moving average charts for the first 20 patients evaluated using global gamma criteria of 3%/3 mm, 3%/2 mm, 2%/3 mm, and 2%/2 mm, respectively. EWMA: Exponentially Weighted Moving Average; UCL: Upper control limit; LCL: Lower control limit.

Figure 3 Individual-moving range for the 350 patients. A-D: Individual-moving range charts for the 350 patients evaluated using global gamma criteria of 3%/3 mm, 3%/2 mm, 2%/3 mm, and 2%/2 mm, respectively. I-MR: Individual-moving range; UCL: Upper control limit; LCL: Lower control limit.

Figure 4 Exponentially weighted moving average. A-D: Exponentially weighted moving average charts for the 350 patients evaluated using global gamma criteria of 3%/3 mm, 3%/2 mm, 2%/3 mm, and 2%/2 mm, respectively. EWMA: Exponentially Weighted Moving Average; UCL: Upper control limit; LCL: Lower control limit.

Figure 5 Schematic summary figure showing how the statistical process control parameters (control limits, tolerance limits, and action limits) progressively decline across stricter gamma criteria (3%/3 mm → 2%/2 mm). SPC: Statistical process control; CL: Control limits.