Automated Detection of Poor-Quality Scintigraphic Images Using Machine Learning

Objective In the present study, we have used machine learning algorithm to accomplish the task of automated detection of poor-quality scintigraphic images. We have validated the accuracy of our machine learning algorithm on 99mTc-methyl diphosphonate (99mTc-MDP) bone scan images. Material...

Full description

Saved in:
Bibliographic Details
Main Authors: Anil K. Pandey, Akshima Sharma, Param D. Sharma, Chandra S. Bal, Rakesh Kumar
Format: Article
Language:English
Published: Thieme Medical and Scientific Publishers Pvt. Ltd. 2022-12-01
Series:World Journal of Nuclear Medicine
Subjects:
Online Access:http://www.thieme-connect.de/DOI/DOI?10.1055/s-0042-1750436
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849702298184318976
author Anil K. Pandey
Akshima Sharma
Param D. Sharma
Chandra S. Bal
Rakesh Kumar
author_facet Anil K. Pandey
Akshima Sharma
Param D. Sharma
Chandra S. Bal
Rakesh Kumar
author_sort Anil K. Pandey
collection DOAJ
description Objective In the present study, we have used machine learning algorithm to accomplish the task of automated detection of poor-quality scintigraphic images. We have validated the accuracy of our machine learning algorithm on 99mTc-methyl diphosphonate (99mTc-MDP) bone scan images. Materials and Methods Ninety-nine patients underwent 99mTC-MDP bone scan acquisition twice at two different acquisition speeds, one at low speed and another at double the speed of the first scan, with patient lying in the same position on the scan table. The low-speed acquisition resulted in good-quality images and the high-speed acquisition resulted in poor-quality images. The principal component analysis (PCA) of all the images was performed and the first 32 principal components (PCs) were retained as feature vectors of the image. These 32 feature vectors of each image were used for the classification of images into poor or good quality using machine learning algorithm (multivariate adaptive regression splines [MARS]). The data were split into two sets, that is, training set and test set in the ratio of 60:40. Hyperparameter tuning of the model was done in which five-fold cross-validation was performed. Receiver operator characteristic (ROC) analysis was used to select the optimal model using the largest value of area under the ROC curve. Sensitivity, specificity, and accuracy for the classification of poor- and good-quality images were taken as metrics for the performance of the algorithm. Result Accuracy, sensitivity, and specificity of the model in classifying poor-quality and good-quality images were 93.22, 93.22, and 93.22%, respectively, for the training dataset and 86.88, 80, and 93.7%, respectively, for the test dataset. Conclusion Machine learning algorithms can be used to classify poor- and good-quality images with good accuracy (86.88%) using 32 PCs as the feature vector and MARS as the classification model.
format Article
id doaj-art-29b1d13033f043eebeefdf42ebde5bfd
institution DOAJ
issn 1450-1147
1607-3312
language English
publishDate 2022-12-01
publisher Thieme Medical and Scientific Publishers Pvt. Ltd.
record_format Article
series World Journal of Nuclear Medicine
spelling doaj-art-29b1d13033f043eebeefdf42ebde5bfd2025-08-20T03:17:42ZengThieme Medical and Scientific Publishers Pvt. Ltd.World Journal of Nuclear Medicine1450-11471607-33122022-12-01210427628210.1055/s-0042-1750436Automated Detection of Poor-Quality Scintigraphic Images Using Machine LearningAnil K. Pandey0Akshima Sharma1Param D. Sharma2Chandra S. Bal3Rakesh Kumar4Department of Nuclear Medicine, All India Institute of Medical Sciences, Ansari Nagar, New Delhi, IndiaDepartment of Urology, All India Institute of Medical Sciences, Ansari Nagar, New Delhi, IndiaDepartment of Computer Science, Sri Guru Tegh Bahadur Khalsa College, University of Delhi, Delhi, IndiaDepartment of Nuclear Medicine, All India Institute of Medical Sciences, Ansari Nagar, New Delhi, IndiaDepartment of Nuclear Medicine, All India Institute of Medical Sciences, Ansari Nagar, New Delhi, IndiaObjective In the present study, we have used machine learning algorithm to accomplish the task of automated detection of poor-quality scintigraphic images. We have validated the accuracy of our machine learning algorithm on 99mTc-methyl diphosphonate (99mTc-MDP) bone scan images. Materials and Methods Ninety-nine patients underwent 99mTC-MDP bone scan acquisition twice at two different acquisition speeds, one at low speed and another at double the speed of the first scan, with patient lying in the same position on the scan table. The low-speed acquisition resulted in good-quality images and the high-speed acquisition resulted in poor-quality images. The principal component analysis (PCA) of all the images was performed and the first 32 principal components (PCs) were retained as feature vectors of the image. These 32 feature vectors of each image were used for the classification of images into poor or good quality using machine learning algorithm (multivariate adaptive regression splines [MARS]). The data were split into two sets, that is, training set and test set in the ratio of 60:40. Hyperparameter tuning of the model was done in which five-fold cross-validation was performed. Receiver operator characteristic (ROC) analysis was used to select the optimal model using the largest value of area under the ROC curve. Sensitivity, specificity, and accuracy for the classification of poor- and good-quality images were taken as metrics for the performance of the algorithm. Result Accuracy, sensitivity, and specificity of the model in classifying poor-quality and good-quality images were 93.22, 93.22, and 93.22%, respectively, for the training dataset and 86.88, 80, and 93.7%, respectively, for the test dataset. Conclusion Machine learning algorithms can be used to classify poor- and good-quality images with good accuracy (86.88%) using 32 PCs as the feature vector and MARS as the classification model.http://www.thieme-connect.de/DOI/DOI?10.1055/s-0042-1750436multivariate adaptive regression splinesnumber of principal componentsclassification of imagesmachine learning
spellingShingle Anil K. Pandey
Akshima Sharma
Param D. Sharma
Chandra S. Bal
Rakesh Kumar
Automated Detection of Poor-Quality Scintigraphic Images Using Machine Learning
World Journal of Nuclear Medicine
multivariate adaptive regression splines
number of principal components
classification of images
machine learning
title Automated Detection of Poor-Quality Scintigraphic Images Using Machine Learning
title_full Automated Detection of Poor-Quality Scintigraphic Images Using Machine Learning
title_fullStr Automated Detection of Poor-Quality Scintigraphic Images Using Machine Learning
title_full_unstemmed Automated Detection of Poor-Quality Scintigraphic Images Using Machine Learning
title_short Automated Detection of Poor-Quality Scintigraphic Images Using Machine Learning
title_sort automated detection of poor quality scintigraphic images using machine learning
topic multivariate adaptive regression splines
number of principal components
classification of images
machine learning
url http://www.thieme-connect.de/DOI/DOI?10.1055/s-0042-1750436
work_keys_str_mv AT anilkpandey automateddetectionofpoorqualityscintigraphicimagesusingmachinelearning
AT akshimasharma automateddetectionofpoorqualityscintigraphicimagesusingmachinelearning
AT paramdsharma automateddetectionofpoorqualityscintigraphicimagesusingmachinelearning
AT chandrasbal automateddetectionofpoorqualityscintigraphicimagesusingmachinelearning
AT rakeshkumar automateddetectionofpoorqualityscintigraphicimagesusingmachinelearning