Design of self-starting multivariate control chart for monitoring patients suspected to bone marrow metastasis

Abstract

Nowadays, statistical process control methods are widely used in healthcare, especially for cancer patients, to help doctors in interpreting and diagnosing medical data. Since adequate data from patients with bone marrow metastasis are often unavailable, the researchers use a self-starting control chart. In this regard, using the self-starting multivariate control chart, we monitor the status of people suspected of bone marrow metastasis in the pelvic region. For this, using a two-dimensional discrete wavelet transformation, we extracted some features from the ADC and T1 magnetic resonance images of ten bone marrow metastasis samples. Out of these features, we selected six ones for final analysis. Then, using the self-starting SSMEWMA and SST2, we performed the simulation studies, as well as a numerical example, on the extracted features and evaluated the performance of the proposed methods in terms of average run length measure. The simulation results verified the appropriate performance of the proposed methods in diagnosing patients with bone marrow metastasis than non-metastasis ones.