Machine Learning-Based Computed Tomography Texture Analysis of Lytic Bone Lesions Needing Biopsy: A Preliminary Study
Introduction:Currently, medical imaging has a limited capacity to achieve a final histopathological diagnosis of bone lesions. This study aimed to evaluate the use of machine learning (ML)-based computed tomography (CT) texture analysis to determine benign and malignant behaviors of lytic bone lesio...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Galenos Yayinevi
2021-08-01
|
| Series: | İstanbul Medical Journal |
| Subjects: | |
| Online Access: |
http://istanbulmedicaljournal.org/archives/archive-detail/article-preview/machine-learning-based-computed-tomography-texture/48477
|
| _version_ | 1797914392318181376 |
|---|---|
| author | İlhan Nahit Mutlu Burak Koçak Ece Ateş Kuş Melis Baykara Ulusan Özgür Kılıçkesmez |
| author_facet | İlhan Nahit Mutlu Burak Koçak Ece Ateş Kuş Melis Baykara Ulusan Özgür Kılıçkesmez |
| author_sort | İlhan Nahit Mutlu |
| collection | DOAJ |
| description | Introduction:Currently, medical imaging has a limited capacity to achieve a final histopathological diagnosis of bone lesions. This study aimed to evaluate the use of machine learning (ML)-based computed tomography (CT) texture analysis to determine benign and malignant behaviors of lytic bone lesions needing a biopsy.Methods:This retrospective study included 58 patients with lytic bone lesions. Lesion segmentation was independently performed by two observers. After evaluating unenhanced CT images, a total of 744 texture features were obtained. Reproducibility analysis and feature selection were used for dimension reduction. A training data set with a nested cross-validation approach was used for feature selection, optimization, and validation. Testing was executed on the remaining unseen data set. Classifications were done using five base ML classifiers and three voting strategies.Results:The best predictive performance was achieved using the k-nearest neighbors algorithm with six features. The area under the curve, accuracy, sensitivity, and specificity of the best algorithm were, respectively, 0.774%, 78.1%, 78%, and 78.1% for the validation data set; and 0.861, 82.4%, 82.4%, and 81.5% for the unseen test data set.Conclusion:The ML-based CT texture analysis may be a promising non-invasive technique for determining benign and malignant behaviors of lytic bone lesions that need a biopsy. |
| first_indexed | 2024-04-10T12:27:06Z |
| format | Article |
| id | doaj.art-f1ead63a54ca4dd1b776632d27a5c7f3 |
| institution | Directory Open Access Journal |
| issn | 2619-9793 2148-094X |
| language | English |
| last_indexed | 2024-04-10T12:27:06Z |
| publishDate | 2021-08-01 |
| publisher | Galenos Yayinevi |
| record_format | Article |
| series | İstanbul Medical Journal |
| spelling | doaj.art-f1ead63a54ca4dd1b776632d27a5c7f32023-02-15T16:15:10ZengGalenos Yayineviİstanbul Medical Journal2619-97932148-094X2021-08-0122322323110.4274/imj.galenos.2021.8752813049054Machine Learning-Based Computed Tomography Texture Analysis of Lytic Bone Lesions Needing Biopsy: A Preliminary Studyİlhan Nahit Mutlu0Burak Koçak1Ece Ateş Kuş2Melis Baykara Ulusan3Özgür Kılıçkesmez4 University of Health Sciences Turkey, Başaksehir Çam ve Sakura City Hospital, Clinic of Radiology, İstanbul, Turkey University of Health Sciences Turkey, Başaksehir Çam ve Sakura City Hospital, Clinic of Radiology, İstanbul, Turkey University of Health Sciences Turkey, İstanbul Training and Research Hospital, Clinic of Radiology, İstanbul, Turkey University of Health Sciences Turkey, İstanbul Training and Research Hospital, Clinic of Radiology, İstanbul, Turkey University of Health Sciences Turkey, Başaksehir Çam ve Sakura City Hospital, Clinic of Radiology, İstanbul, Turkey Introduction:Currently, medical imaging has a limited capacity to achieve a final histopathological diagnosis of bone lesions. This study aimed to evaluate the use of machine learning (ML)-based computed tomography (CT) texture analysis to determine benign and malignant behaviors of lytic bone lesions needing a biopsy.Methods:This retrospective study included 58 patients with lytic bone lesions. Lesion segmentation was independently performed by two observers. After evaluating unenhanced CT images, a total of 744 texture features were obtained. Reproducibility analysis and feature selection were used for dimension reduction. A training data set with a nested cross-validation approach was used for feature selection, optimization, and validation. Testing was executed on the remaining unseen data set. Classifications were done using five base ML classifiers and three voting strategies.Results:The best predictive performance was achieved using the k-nearest neighbors algorithm with six features. The area under the curve, accuracy, sensitivity, and specificity of the best algorithm were, respectively, 0.774%, 78.1%, 78%, and 78.1% for the validation data set; and 0.861, 82.4%, 82.4%, and 81.5% for the unseen test data set.Conclusion:The ML-based CT texture analysis may be a promising non-invasive technique for determining benign and malignant behaviors of lytic bone lesions that need a biopsy. http://istanbulmedicaljournal.org/archives/archive-detail/article-preview/machine-learning-based-computed-tomography-texture/48477 bonetexture analysisradiomicsmachine learningartificial intelligence |
| spellingShingle | İlhan Nahit Mutlu Burak Koçak Ece Ateş Kuş Melis Baykara Ulusan Özgür Kılıçkesmez Machine Learning-Based Computed Tomography Texture Analysis of Lytic Bone Lesions Needing Biopsy: A Preliminary Study İstanbul Medical Journal bone texture analysis radiomics machine learning artificial intelligence |
| title | Machine Learning-Based Computed Tomography Texture Analysis of Lytic Bone Lesions Needing Biopsy: A Preliminary Study |
| title_full | Machine Learning-Based Computed Tomography Texture Analysis of Lytic Bone Lesions Needing Biopsy: A Preliminary Study |
| title_fullStr | Machine Learning-Based Computed Tomography Texture Analysis of Lytic Bone Lesions Needing Biopsy: A Preliminary Study |
| title_full_unstemmed | Machine Learning-Based Computed Tomography Texture Analysis of Lytic Bone Lesions Needing Biopsy: A Preliminary Study |
| title_short | Machine Learning-Based Computed Tomography Texture Analysis of Lytic Bone Lesions Needing Biopsy: A Preliminary Study |
| title_sort | machine learning based computed tomography texture analysis of lytic bone lesions needing biopsy a preliminary study |
| topic | bone texture analysis radiomics machine learning artificial intelligence |
| url |
http://istanbulmedicaljournal.org/archives/archive-detail/article-preview/machine-learning-based-computed-tomography-texture/48477
|
| work_keys_str_mv | AT ilhannahitmutlu machinelearningbasedcomputedtomographytextureanalysisoflyticbonelesionsneedingbiopsyapreliminarystudy AT burakkocak machinelearningbasedcomputedtomographytextureanalysisoflyticbonelesionsneedingbiopsyapreliminarystudy AT eceateskus machinelearningbasedcomputedtomographytextureanalysisoflyticbonelesionsneedingbiopsyapreliminarystudy AT melisbaykaraulusan machinelearningbasedcomputedtomographytextureanalysisoflyticbonelesionsneedingbiopsyapreliminarystudy AT ozgurkılıckesmez machinelearningbasedcomputedtomographytextureanalysisoflyticbonelesionsneedingbiopsyapreliminarystudy |