Affiliations: [a] Department of Military Preventive Medicine, The Fourth Military Medical University, Xi’an, Shaanxi, China
| [b] Department of Medical Record, Lanzhou University Second Hospital, Lanzhou, Gansu, China
| [c] Department of Intensive Care Unit/Medical Record2, Lanzhou University Second Hospital, Lanzhou, Gansu, China
Correspondence:
[*]
Corresponding author. Duowen Xie, Department of Intensive Care Unit/Medical Record2, Lanzhou University Second Hospital, Lanzhou, Gansu, China. E-mail: [email protected].
Abstract: Medical image processing is an interdisciplinary subject of integrated medical imaging, mathematics, computer science and other disciplines. With high spatial resolution, high signal-to-noise ratio and high resolution of soft tissue, the technology can accurately locate the target areas of interest in medical images, thus providing useful information for clinicians to formulate disease treatment plans. These techniques include digital subtraction angiography, magnetic resonance imaging, computed tomography, ultrasound imaging and positron emission tomography. The purpose of this paper is to study the application of fuzzy C-means clustering in image analysis of critical medicine. This paper discusses the classification effect, clustering process, iteration times and running time of different algorithms, and the segmentation effect of different algorithms. By designing parameters and carrying out simulation experiments, the traditional clustering algorithm and improved local adaptive method are compared, and the problem of long coding time of traditional image compression algorithm is solved. The simulation results under the same working environment show that the coding speed of the algorithm is about five times faster than that of the traditional image compression algorithm without affecting the signal-to-noise ratio and compression rate, which proves the superiority of the algorithm.
Keywords: Medical image processing, fuzzy C-means algorithm, clustering algorithm, medical image analysis
