Automatic Characterization of Prostate Suspect Lesions on T2-Weighted Image Acquisitions Using Texture Features and Machine-Learning Methods: A Pilot Study
<b>Background</b>: Prostate cancer (PCa) is the most frequent neoplasia in the male population. According to the International Society of Urological Pathology (ISUP), PCa can be divided into two major groups, based on their prognosis and treatment options. Multiparametric magnetic resona...
Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-01-01
|
| Series: | Diagnostics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2075-4418/15/1/106 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841549257277964288 |
|---|---|
| author | Teodora Telecan Cosmin Caraiani Bianca Boca Roxana Sipos-Lascu Laura Diosan Zoltan Balint Raluca Maria Hendea Iulia Andras Nicolae Crisan Monica Lupsor-Platon |
| author_facet | Teodora Telecan Cosmin Caraiani Bianca Boca Roxana Sipos-Lascu Laura Diosan Zoltan Balint Raluca Maria Hendea Iulia Andras Nicolae Crisan Monica Lupsor-Platon |
| author_sort | Teodora Telecan |
| collection | DOAJ |
| description | <b>Background</b>: Prostate cancer (PCa) is the most frequent neoplasia in the male population. According to the International Society of Urological Pathology (ISUP), PCa can be divided into two major groups, based on their prognosis and treatment options. Multiparametric magnetic resonance imaging (mpMRI) holds a central role in PCa assessment; however, it does not have a one-to-one correspondence with the histopathological grading of tumors. Recently, artificial intelligence (AI)-based algorithms and textural analysis, a subdivision of radiomics, have shown potential in bridging this gap. <b>Objectives</b>: We aimed to develop a machine-learning algorithm that predicts the ISUP grade of manually contoured prostate nodules on T2-weighted images and classifies them into clinically significant and indolent ones. <b>Materials and Methods:</b> We included 55 patients with 76 lesions. All patients were examined on the same 1.5 Tesla mpMRI scanner. Each nodule was manually segmented using the open-source 3D Slicer platform, and textural features were extracted using the PyRadiomics (version 3.0.1) library. The software was based on machine-learning classifiers. The accuracy was calculated based on precision, recall, and F1 scores. <b>Results</b>: The median age of the study group was 64 years (IQR 61–68), and the mean PSA value was 11.14 ng/mL. A total of 85.52% of the nodules were graded PI-RADS 4 or higher. Overall, the algorithm classified indolent and clinically significant PCas with an accuracy of 87.2%. Further, when trained to differentiate each ISUP group, the accuracy was 80.3%. <b>Conclusions</b>: We developed an AI-based decision-support system that accurately differentiates between the two PCa prognostic groups using only T2 MRI acquisitions by employing radiomics with a robust machine-learning architecture. |
| format | Article |
| id | doaj-art-35076c0e3b5b45768294cbcd98afbf12 |
| institution | Kabale University |
| issn | 2075-4418 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Diagnostics |
| spelling | doaj-art-35076c0e3b5b45768294cbcd98afbf122025-01-10T13:16:44ZengMDPI AGDiagnostics2075-44182025-01-0115110610.3390/diagnostics15010106Automatic Characterization of Prostate Suspect Lesions on T2-Weighted Image Acquisitions Using Texture Features and Machine-Learning Methods: A Pilot StudyTeodora Telecan0Cosmin Caraiani1Bianca Boca2Roxana Sipos-Lascu3Laura Diosan4Zoltan Balint5Raluca Maria Hendea6Iulia Andras7Nicolae Crisan8Monica Lupsor-Platon9Department of Anatomy and Embryology, “Iuliu Hatieganu” University of Medicine and Pharmacy, 400012 Cluj-Napoca, RomaniaDepartment of Medical Imaging, “Iuliu Hatieganu” University of Medicine and Pharmacy, 400012 Cluj-Napoca, RomaniaDepartment of Medical Imaging, “Iuliu Hatieganu” University of Medicine and Pharmacy, 400012 Cluj-Napoca, RomaniaDepartment of Computer Science, Faculty of Mathematics and Computer Science, “Babes-Bolyai” University, 400157 Cluj-Napoca, RomaniaDepartment of Computer Science, Faculty of Mathematics and Computer Science, “Babes-Bolyai” University, 400157 Cluj-Napoca, RomaniaDepartment of Biomedical Physics, Faculty of Physics, “Babes-Bolyai” University, 400084 Cluj-Napoca, RomaniaDepartment of Pathology, “Iuliu Hatieganu” University of Medicine and Pharmacy, 400012 Cluj-Napoca, RomaniaDepartment of Urology, “Iuliu Hatieganu” University of Medicine and Pharmacy, 400012 Cluj-Napoca, RomaniaDepartment of Urology, “Iuliu Hatieganu” University of Medicine and Pharmacy, 400012 Cluj-Napoca, RomaniaDepartment of Medical Imaging, “Iuliu Hatieganu” University of Medicine and Pharmacy, 400012 Cluj-Napoca, Romania<b>Background</b>: Prostate cancer (PCa) is the most frequent neoplasia in the male population. According to the International Society of Urological Pathology (ISUP), PCa can be divided into two major groups, based on their prognosis and treatment options. Multiparametric magnetic resonance imaging (mpMRI) holds a central role in PCa assessment; however, it does not have a one-to-one correspondence with the histopathological grading of tumors. Recently, artificial intelligence (AI)-based algorithms and textural analysis, a subdivision of radiomics, have shown potential in bridging this gap. <b>Objectives</b>: We aimed to develop a machine-learning algorithm that predicts the ISUP grade of manually contoured prostate nodules on T2-weighted images and classifies them into clinically significant and indolent ones. <b>Materials and Methods:</b> We included 55 patients with 76 lesions. All patients were examined on the same 1.5 Tesla mpMRI scanner. Each nodule was manually segmented using the open-source 3D Slicer platform, and textural features were extracted using the PyRadiomics (version 3.0.1) library. The software was based on machine-learning classifiers. The accuracy was calculated based on precision, recall, and F1 scores. <b>Results</b>: The median age of the study group was 64 years (IQR 61–68), and the mean PSA value was 11.14 ng/mL. A total of 85.52% of the nodules were graded PI-RADS 4 or higher. Overall, the algorithm classified indolent and clinically significant PCas with an accuracy of 87.2%. Further, when trained to differentiate each ISUP group, the accuracy was 80.3%. <b>Conclusions</b>: We developed an AI-based decision-support system that accurately differentiates between the two PCa prognostic groups using only T2 MRI acquisitions by employing radiomics with a robust machine-learning architecture.https://www.mdpi.com/2075-4418/15/1/106radiomicstextural analysisprostate cancermpMRIartificial intelligencemachine learning |
| spellingShingle | Teodora Telecan Cosmin Caraiani Bianca Boca Roxana Sipos-Lascu Laura Diosan Zoltan Balint Raluca Maria Hendea Iulia Andras Nicolae Crisan Monica Lupsor-Platon Automatic Characterization of Prostate Suspect Lesions on T2-Weighted Image Acquisitions Using Texture Features and Machine-Learning Methods: A Pilot Study Diagnostics radiomics textural analysis prostate cancer mpMRI artificial intelligence machine learning |
| title | Automatic Characterization of Prostate Suspect Lesions on T2-Weighted Image Acquisitions Using Texture Features and Machine-Learning Methods: A Pilot Study |
| title_full | Automatic Characterization of Prostate Suspect Lesions on T2-Weighted Image Acquisitions Using Texture Features and Machine-Learning Methods: A Pilot Study |
| title_fullStr | Automatic Characterization of Prostate Suspect Lesions on T2-Weighted Image Acquisitions Using Texture Features and Machine-Learning Methods: A Pilot Study |
| title_full_unstemmed | Automatic Characterization of Prostate Suspect Lesions on T2-Weighted Image Acquisitions Using Texture Features and Machine-Learning Methods: A Pilot Study |
| title_short | Automatic Characterization of Prostate Suspect Lesions on T2-Weighted Image Acquisitions Using Texture Features and Machine-Learning Methods: A Pilot Study |
| title_sort | automatic characterization of prostate suspect lesions on t2 weighted image acquisitions using texture features and machine learning methods a pilot study |
| topic | radiomics textural analysis prostate cancer mpMRI artificial intelligence machine learning |
| url | https://www.mdpi.com/2075-4418/15/1/106 |
| work_keys_str_mv | AT teodoratelecan automaticcharacterizationofprostatesuspectlesionsont2weightedimageacquisitionsusingtexturefeaturesandmachinelearningmethodsapilotstudy AT cosmincaraiani automaticcharacterizationofprostatesuspectlesionsont2weightedimageacquisitionsusingtexturefeaturesandmachinelearningmethodsapilotstudy AT biancaboca automaticcharacterizationofprostatesuspectlesionsont2weightedimageacquisitionsusingtexturefeaturesandmachinelearningmethodsapilotstudy AT roxanasiposlascu automaticcharacterizationofprostatesuspectlesionsont2weightedimageacquisitionsusingtexturefeaturesandmachinelearningmethodsapilotstudy AT lauradiosan automaticcharacterizationofprostatesuspectlesionsont2weightedimageacquisitionsusingtexturefeaturesandmachinelearningmethodsapilotstudy AT zoltanbalint automaticcharacterizationofprostatesuspectlesionsont2weightedimageacquisitionsusingtexturefeaturesandmachinelearningmethodsapilotstudy AT ralucamariahendea automaticcharacterizationofprostatesuspectlesionsont2weightedimageacquisitionsusingtexturefeaturesandmachinelearningmethodsapilotstudy AT iuliaandras automaticcharacterizationofprostatesuspectlesionsont2weightedimageacquisitionsusingtexturefeaturesandmachinelearningmethodsapilotstudy AT nicolaecrisan automaticcharacterizationofprostatesuspectlesionsont2weightedimageacquisitionsusingtexturefeaturesandmachinelearningmethodsapilotstudy AT monicalupsorplaton automaticcharacterizationofprostatesuspectlesionsont2weightedimageacquisitionsusingtexturefeaturesandmachinelearningmethodsapilotstudy |