Renal structural image processing techniques: a systematic review

Background and objective: Renal disease, such as nephritis and nephropathy, is very harmful to human health. Accordingly, how to achieve early diagnosis and enhance treatment for kidney disorders would be the important lesion. Nevertheless, the clues from the clinical data, such as biochemistry exam...

Full description

Saved in:
Bibliographic Details
Published in:Renal failure 2019-01, Vol.41 (1), p.57-68
Main Authors: Asadzadeh, Shiva, Khosroshahi, Hamid Tayebi, Abedi, Behzad, Ghasemi, Yaghoob, Meshgini, Saeed
Format: Article
Language:English
Subjects:
Biopsy
Blood vessels
Diagnosis
Glomeruli segmentation
image analysis
Image processing
Kidneys
Nephritis
Nephropathy
pathology
renal
Renal cell carcinoma
Segmentation
State-of-the-Art Review
Tubules
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Background and objective: Renal disease, such as nephritis and nephropathy, is very harmful to human health. Accordingly, how to achieve early diagnosis and enhance treatment for kidney disorders would be the important lesion. Nevertheless, the clues from the clinical data, such as biochemistry examination, serological examination, and radiological studies are quite indirect and limited. It is no doubt that pathological examination of kidney will supply the direct evidence. There is a requirement for greater understanding of image processing techniques for renal diagnosis to optimize treatment and patient care. Methods: This study aims to systematically review the literature on publications that has been used image processing methods on pathological microscopic image for renal diagnosis. Results: Nine included studies revealed image analysis techniques for the diagnosis of renal abnormalities on pathological microscopic image, renal image studies are clustered as follows: Glomeruli Segmentation and analysis of the Glomerular basement membrane (55/55%), Blood vessels and tubules classification and detection (22/22%) and The Grading of renal cell carcinomas (22/22%). Conclusions: A medical image analysis method should provide an auto-adaptive and no external-human action dependency. In addition, since medical systems should have special characteristics such as high accuracy and reliability then clinical validation is highly recommended. New high-quality studies based on Moore neighborhood contour tracking method for glomeruli segmentation and using powerful texture analysis techniques such as the local binary pattern are recommended.
ISSN:0886-022X
1525-6049
DOI:10.1080/0886022X.2019.1572016