Kidney Boundary Detection Algorithm Based on Extended Maxima Transformations for Computed Tomography Diagnosis

This article describes the automated computed tomography (CT) image processing technique supporting kidney detection. The main goal of the study is a fully automatic generation of a kidney boundary for each slice in the set of slices obtained in the computed tomography examination. This work describ...

Full description

Bibliographic Details
Main Authors: Tomasz Les, Tomasz Markiewicz, Miroslaw Dziekiewicz, Malgorzata Lorent
Format: Article
Language:English
Published: MDPI AG 2020-10-01
Series:Applied Sciences
Subjects:
Online Access:https://www.mdpi.com/2076-3417/10/21/7512
id doaj-cf0c310ca33342e4bdb6ff2182bdb3ed
record_format Article
spelling doaj-cf0c310ca33342e4bdb6ff2182bdb3ed2020-11-25T03:41:51ZengMDPI AGApplied Sciences2076-34172020-10-01107512751210.3390/app10217512Kidney Boundary Detection Algorithm Based on Extended Maxima Transformations for Computed Tomography DiagnosisTomasz Les0Tomasz Markiewicz1Miroslaw Dziekiewicz2Malgorzata Lorent3Faculty of Electrical Engineering, Warsaw University of Technology, 00-661 Warsaw, PolandFaculty of Electrical Engineering, Warsaw University of Technology, 00-661 Warsaw, PolandMilitary Institute of Medicine, 04-349 Warsaw, PolandMilitary Institute of Medicine, 04-349 Warsaw, PolandThis article describes the automated computed tomography (CT) image processing technique supporting kidney detection. The main goal of the study is a fully automatic generation of a kidney boundary for each slice in the set of slices obtained in the computed tomography examination. This work describes three main tasks in the process of automatic kidney identification: the initial location of the kidneys using the U-Net convolutional neural network, the generation of an accurate kidney boundary using extended maxima transformation, and the application of the slice scanning algorithm supporting the process of generating the result for the next slice, using the result of the previous one. To assess the quality of the proposed technique of medical image analysis, automatic numerical tests were performed. In the test section, we presented numerical results, calculating the F1-score of kidney boundary detection by an automatic system, compared to the kidneys boundaries manually generated by a human expert from a medical center. The influence of the use of U-Net support in the initial detection of the kidney on the final F1-score of generating the kidney outline was also evaluated. The F1-score achieved by the automated system is <inline-formula><math display="inline"><semantics><mrow><mn>84</mn><mo>%</mo><mo> </mo><mo>±</mo><mo> </mo><mn>10</mn><mo>%</mo></mrow></semantics></math></inline-formula> for the system without U-Net support and <inline-formula><math display="inline"><semantics><mrow><mn>89</mn><mo>%</mo><mo> </mo><mo>±</mo><mo> </mo><mn>9</mn><mo>%</mo></mrow></semantics></math></inline-formula> for the system with U-Net support. Performance tests show that the presented technique can generate the kidney boundary up to 3 times faster than raw U-Net-based network. The proposed kidney recognition system can be successfully used in systems that require a very fast image processing time. The measurable effect of the developed techniques is a practical help for doctors, specialists from medical centers dealing with the analysis and description of medical image data.https://www.mdpi.com/2076-3417/10/21/7512computer-aided diagnosisimage segmentationartificial intelligencekidney disease diagnosis
collection DOAJ
language English
format Article
sources DOAJ
author Tomasz Les
Tomasz Markiewicz
Miroslaw Dziekiewicz
Malgorzata Lorent
spellingShingle Tomasz Les
Tomasz Markiewicz
Miroslaw Dziekiewicz
Malgorzata Lorent
Kidney Boundary Detection Algorithm Based on Extended Maxima Transformations for Computed Tomography Diagnosis
Applied Sciences
computer-aided diagnosis
image segmentation
artificial intelligence
kidney disease diagnosis
author_facet Tomasz Les
Tomasz Markiewicz
Miroslaw Dziekiewicz
Malgorzata Lorent
author_sort Tomasz Les
title Kidney Boundary Detection Algorithm Based on Extended Maxima Transformations for Computed Tomography Diagnosis
title_short Kidney Boundary Detection Algorithm Based on Extended Maxima Transformations for Computed Tomography Diagnosis
title_full Kidney Boundary Detection Algorithm Based on Extended Maxima Transformations for Computed Tomography Diagnosis
title_fullStr Kidney Boundary Detection Algorithm Based on Extended Maxima Transformations for Computed Tomography Diagnosis
title_full_unstemmed Kidney Boundary Detection Algorithm Based on Extended Maxima Transformations for Computed Tomography Diagnosis
title_sort kidney boundary detection algorithm based on extended maxima transformations for computed tomography diagnosis
publisher MDPI AG
series Applied Sciences
issn 2076-3417
publishDate 2020-10-01
description This article describes the automated computed tomography (CT) image processing technique supporting kidney detection. The main goal of the study is a fully automatic generation of a kidney boundary for each slice in the set of slices obtained in the computed tomography examination. This work describes three main tasks in the process of automatic kidney identification: the initial location of the kidneys using the U-Net convolutional neural network, the generation of an accurate kidney boundary using extended maxima transformation, and the application of the slice scanning algorithm supporting the process of generating the result for the next slice, using the result of the previous one. To assess the quality of the proposed technique of medical image analysis, automatic numerical tests were performed. In the test section, we presented numerical results, calculating the F1-score of kidney boundary detection by an automatic system, compared to the kidneys boundaries manually generated by a human expert from a medical center. The influence of the use of U-Net support in the initial detection of the kidney on the final F1-score of generating the kidney outline was also evaluated. The F1-score achieved by the automated system is <inline-formula><math display="inline"><semantics><mrow><mn>84</mn><mo>%</mo><mo> </mo><mo>±</mo><mo> </mo><mn>10</mn><mo>%</mo></mrow></semantics></math></inline-formula> for the system without U-Net support and <inline-formula><math display="inline"><semantics><mrow><mn>89</mn><mo>%</mo><mo> </mo><mo>±</mo><mo> </mo><mn>9</mn><mo>%</mo></mrow></semantics></math></inline-formula> for the system with U-Net support. Performance tests show that the presented technique can generate the kidney boundary up to 3 times faster than raw U-Net-based network. The proposed kidney recognition system can be successfully used in systems that require a very fast image processing time. The measurable effect of the developed techniques is a practical help for doctors, specialists from medical centers dealing with the analysis and description of medical image data.
topic computer-aided diagnosis
image segmentation
artificial intelligence
kidney disease diagnosis
url https://www.mdpi.com/2076-3417/10/21/7512
work_keys_str_mv AT tomaszles kidneyboundarydetectionalgorithmbasedonextendedmaximatransformationsforcomputedtomographydiagnosis
AT tomaszmarkiewicz kidneyboundarydetectionalgorithmbasedonextendedmaximatransformationsforcomputedtomographydiagnosis
AT miroslawdziekiewicz kidneyboundarydetectionalgorithmbasedonextendedmaximatransformationsforcomputedtomographydiagnosis
AT malgorzatalorent kidneyboundarydetectionalgorithmbasedonextendedmaximatransformationsforcomputedtomographydiagnosis
_version_ 1724527919824371712