Medical Image Segmentation with Adjustable Computational Complexity Using Data Density Functionals

Chien-Chang Chen,Ming-Ze Kao,Henry Horng-Shing Lu,Meng-Yuan Tsai

doi:10.3390/app9081718

Abstract

Techniques of automatic medical image segmentation are the most important methods for clinical investigation, anatomic research, and modern medicine. Various image structures constructed from imaging apparatus achieve a diversity of medical applications. However, the diversified structures are also a burden of contemporary techniques. Performing an image segmentation with a tremendously small size (<25 pixels by 25 pixels) or tremendously large size (>1024 pixels by 1024 pixels) becomes a challenge in perspectives of both technical feasibility and theoretical development. Noise and pixel pollution caused by the imaging apparatus even aggravate the difficulty of image segmentation. To simultaneously overcome the mentioned predicaments, we propose a new method of medical image segmentation with adjustable computational complexity by introducing data density functionals. Under this theoretical framework, several kernels can be assigned to conquer specific predicaments. A square-root potential kernel is used to smoothen the featured components of employed images, while a Yukawa potential kernel is applied to enhance local featured properties. Besides, the characteristic of global density functional estimation also allows image compression without losing the main image feature structures. Experiments on image segmentation showed successful results with various compression ratios. The computational complexity was significantly improved, and the score of accuracy estimated by the Jaccard index had a great outcome. Moreover, noise and regions of light pollution were mostly filtered out in the procedure of image compression.

Highlights

Automatic identification and segmentation of medical imageries benefit the planning and guidance of modern surgery [1,2,3,4,5], clinical investigations [6,7,8,9,10], rehabilitation [11,12,13], and so forth.High-quality reconstructive anatomical morphology provides convenience on surgery planning and the understanding between organ functionalities and pathological diagnosis
Effects of sharpened kinetic energy density functionals were introduced into the datasets
Since there is no ground truth for each employed image, each segmentation result without compressing was utilized as the ground truth and a common measurement of similarity, Jaccard index [48], was used for similarity comparisons

Summary

Introduction

Automatic identification and segmentation of medical imageries benefit the planning and guidance of modern surgery [1,2,3,4,5], clinical investigations [6,7,8,9,10], rehabilitation [11,12,13], and so forth.High-quality reconstructive anatomical morphology provides convenience on surgery planning and the understanding between organ functionalities and pathological diagnosis. Sci. 2019, 9, 1718 research progress of connectome [14] and clinical diagnoses related to Alzheimer’s and Parkinson’s diseases [15] Among these applications, robust automatic methods of segmentation for large-scale medical imageries can efficiently save extensive and tedious manual interventions while dealing with humongous tissue labeling and clinical process setting. Among those developed techniques, state-of-the-art methodologies on the field of magnetic resonance image (MRI) processing successfully combined several merits from interdisciplinary methods [16,17,18] and exhibit an opportunity to track brain regions related to relevant diseases [19,20]. Precise identification and segmentation of subthalamic nucleus from three-dimensional medical imageries facilitate the automatic planning of deep brain stimulation, and the clinical result has shown clinical potential for relieving the motor symptoms of advanced Parkinson’s disease [8,21,22]

Results

Discussion

Conclusion