The Primacy of High B-Value 3T-DWI Radiomics in the Prediction of Clinically Significant Prostate Cancer
Predicting clinically significant prostate cancer (csPCa) is crucial in PCa management. 3T-magnetic resonance (MR) systems may have a novel role in quantitative imaging and early csPCa prediction, accordingly. In this study, we develop a radiomic model for predicting csPCa based solely on native b20...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-04-01
|
| Series: | Diagnostics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2075-4418/11/5/739 |
| _version_ | 1797536836107632640 |
|---|---|
| author | Alessandro Bevilacqua Margherita Mottola Fabio Ferroni Alice Rossi Giampaolo Gavelli Domenico Barone |
| author_facet | Alessandro Bevilacqua Margherita Mottola Fabio Ferroni Alice Rossi Giampaolo Gavelli Domenico Barone |
| author_sort | Alessandro Bevilacqua |
| collection | DOAJ |
| description | Predicting clinically significant prostate cancer (csPCa) is crucial in PCa management. 3T-magnetic resonance (MR) systems may have a novel role in quantitative imaging and early csPCa prediction, accordingly. In this study, we develop a radiomic model for predicting csPCa based solely on native b2000 diffusion weighted imaging (DWI<sub>b2000</sub>) and debate the effectiveness of apparent diffusion coefficient (ADC) in the same task. In total, 105 patients were retrospectively enrolled between January–November 2020, with confirmed csPCa or ncsPCa based on biopsy. DWI<sub>b2000</sub> and ADC images acquired with a 3T-MRI were analyzed by computing 84 local first-order radiomic features (RFs). Two predictive models were built based on DWI<sub>b2000</sub> and ADC, separately. Relevant RFs were selected through LASSO, a support vector machine (SVM) classifier was trained using repeated 3-fold cross validation (CV) and validated on a holdout set. The SVM models rely on a single couple of uncorrelated RFs (ρ < 0.15) selected through Wilcoxon rank-sum test (<i>p</i> ≤ 0.05) with Holm–Bonferroni correction. On the holdout set, while the ADC model yielded AUC = 0.76 (95% CI, 0.63–0.96), the DWI<sub>b2000</sub> model reached AUC = 0.84 (95% CI, 0.63–0.90), with specificity = 75%, sensitivity = 90%, and informedness = 0.65. This study establishes the primary role of 3T-DWI<sub>b2000</sub> in PCa quantitative analyses, whilst ADC can remain the leading sequence for detection. |
| first_indexed | 2024-03-10T12:06:28Z |
| format | Article |
| id | doaj.art-2a66c88695b642b3a0fdff7a1e7981c9 |
| institution | Directory Open Access Journal |
| issn | 2075-4418 |
| language | English |
| last_indexed | 2024-03-10T12:06:28Z |
| publishDate | 2021-04-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Diagnostics |
| spelling | doaj.art-2a66c88695b642b3a0fdff7a1e7981c92023-11-21T16:31:36ZengMDPI AGDiagnostics2075-44182021-04-0111573910.3390/diagnostics11050739The Primacy of High B-Value 3T-DWI Radiomics in the Prediction of Clinically Significant Prostate CancerAlessandro Bevilacqua0Margherita Mottola1Fabio Ferroni2Alice Rossi3Giampaolo Gavelli4Domenico Barone5Department of Computer Science and Engineering (DISI), University of Bologna, Viale Risorgimento 2, I-40136 Bologna, ItalyAdvanced Research Center on Electronic Systems (ARCES), University of Bologna, Via Toffano 2/2, I-40125 Bologna, ItalyIRCCS Istituto Romagnolo per lo Studio dei Tumori (IRST) “Dino Amadori”, Via Piero Maroncelli 40, I-47014 Meldola, ItalyIRCCS Istituto Romagnolo per lo Studio dei Tumori (IRST) “Dino Amadori”, Via Piero Maroncelli 40, I-47014 Meldola, ItalyIRCCS Istituto Romagnolo per lo Studio dei Tumori (IRST) “Dino Amadori”, Via Piero Maroncelli 40, I-47014 Meldola, ItalyIRCCS Istituto Romagnolo per lo Studio dei Tumori (IRST) “Dino Amadori”, Via Piero Maroncelli 40, I-47014 Meldola, ItalyPredicting clinically significant prostate cancer (csPCa) is crucial in PCa management. 3T-magnetic resonance (MR) systems may have a novel role in quantitative imaging and early csPCa prediction, accordingly. In this study, we develop a radiomic model for predicting csPCa based solely on native b2000 diffusion weighted imaging (DWI<sub>b2000</sub>) and debate the effectiveness of apparent diffusion coefficient (ADC) in the same task. In total, 105 patients were retrospectively enrolled between January–November 2020, with confirmed csPCa or ncsPCa based on biopsy. DWI<sub>b2000</sub> and ADC images acquired with a 3T-MRI were analyzed by computing 84 local first-order radiomic features (RFs). Two predictive models were built based on DWI<sub>b2000</sub> and ADC, separately. Relevant RFs were selected through LASSO, a support vector machine (SVM) classifier was trained using repeated 3-fold cross validation (CV) and validated on a holdout set. The SVM models rely on a single couple of uncorrelated RFs (ρ < 0.15) selected through Wilcoxon rank-sum test (<i>p</i> ≤ 0.05) with Holm–Bonferroni correction. On the holdout set, while the ADC model yielded AUC = 0.76 (95% CI, 0.63–0.96), the DWI<sub>b2000</sub> model reached AUC = 0.84 (95% CI, 0.63–0.90), with specificity = 75%, sensitivity = 90%, and informedness = 0.65. This study establishes the primary role of 3T-DWI<sub>b2000</sub> in PCa quantitative analyses, whilst ADC can remain the leading sequence for detection.https://www.mdpi.com/2075-4418/11/5/739prostate cancerradiomicsmachine learningtumor stagingcancer heterogeneityimage processing |
| spellingShingle | Alessandro Bevilacqua Margherita Mottola Fabio Ferroni Alice Rossi Giampaolo Gavelli Domenico Barone The Primacy of High B-Value 3T-DWI Radiomics in the Prediction of Clinically Significant Prostate Cancer Diagnostics prostate cancer radiomics machine learning tumor staging cancer heterogeneity image processing |
| title | The Primacy of High B-Value 3T-DWI Radiomics in the Prediction of Clinically Significant Prostate Cancer |
| title_full | The Primacy of High B-Value 3T-DWI Radiomics in the Prediction of Clinically Significant Prostate Cancer |
| title_fullStr | The Primacy of High B-Value 3T-DWI Radiomics in the Prediction of Clinically Significant Prostate Cancer |
| title_full_unstemmed | The Primacy of High B-Value 3T-DWI Radiomics in the Prediction of Clinically Significant Prostate Cancer |
| title_short | The Primacy of High B-Value 3T-DWI Radiomics in the Prediction of Clinically Significant Prostate Cancer |
| title_sort | primacy of high b value 3t dwi radiomics in the prediction of clinically significant prostate cancer |
| topic | prostate cancer radiomics machine learning tumor staging cancer heterogeneity image processing |
| url | https://www.mdpi.com/2075-4418/11/5/739 |
| work_keys_str_mv | AT alessandrobevilacqua theprimacyofhighbvalue3tdwiradiomicsinthepredictionofclinicallysignificantprostatecancer AT margheritamottola theprimacyofhighbvalue3tdwiradiomicsinthepredictionofclinicallysignificantprostatecancer AT fabioferroni theprimacyofhighbvalue3tdwiradiomicsinthepredictionofclinicallysignificantprostatecancer AT alicerossi theprimacyofhighbvalue3tdwiradiomicsinthepredictionofclinicallysignificantprostatecancer AT giampaologavelli theprimacyofhighbvalue3tdwiradiomicsinthepredictionofclinicallysignificantprostatecancer AT domenicobarone theprimacyofhighbvalue3tdwiradiomicsinthepredictionofclinicallysignificantprostatecancer AT alessandrobevilacqua primacyofhighbvalue3tdwiradiomicsinthepredictionofclinicallysignificantprostatecancer AT margheritamottola primacyofhighbvalue3tdwiradiomicsinthepredictionofclinicallysignificantprostatecancer AT fabioferroni primacyofhighbvalue3tdwiradiomicsinthepredictionofclinicallysignificantprostatecancer AT alicerossi primacyofhighbvalue3tdwiradiomicsinthepredictionofclinicallysignificantprostatecancer AT giampaologavelli primacyofhighbvalue3tdwiradiomicsinthepredictionofclinicallysignificantprostatecancer AT domenicobarone primacyofhighbvalue3tdwiradiomicsinthepredictionofclinicallysignificantprostatecancer |