scholarly journals Comparison of machine learning algorithms to predict clinically significant prostate cancer of the peripheral zone with multiparametric MRI using clinical assessment categories and radiomic features

2020 ◽  
Vol 30 (12) ◽  
pp. 6757-6769 ◽  
Author(s):  
Simon Bernatz ◽  
Jörg Ackermann ◽  
Philipp Mandel ◽  
Benjamin Kaltenbach ◽  
Yauheniya Zhdanovich ◽  
...  
Keyword(s):  
Prostate Cancer ◽  
Machine Learning ◽  
Clinical Significance ◽  
Clinical Assessment ◽  
Diagnostic Performance ◽  
Learning Algorithms ◽  
Peripheral Zone ◽  
Multiparametric Mri ◽  
Machine Learning Algorithms ◽  
Clinically Significant

Abstract Objectives To analyze the performance of radiological assessment categories and quantitative computational analysis of apparent diffusion coefficient (ADC) maps using variant machine learning algorithms to differentiate clinically significant versus insignificant prostate cancer (PCa). Methods Retrospectively, 73 patients were included in the study. The patients (mean age, 66.3 ± 7.6 years) were examined with multiparametric MRI (mpMRI) prior to radical prostatectomy (n = 33) or targeted biopsy (n = 40). The index lesion was annotated in MRI ADC and the equivalent histologic slides according to the highest Gleason Grade Group (GrG). Volumes of interest (VOIs) were determined for each lesion and normal-appearing peripheral zone. VOIs were processed by radiomic analysis. For the classification of lesions according to their clinical significance (GrG ≥ 3), principal component (PC) analysis, univariate analysis (UA) with consecutive support vector machines, neural networks, and random forest analysis were performed. Results PC analysis discriminated between benign and malignant prostate tissue. PC evaluation yielded no stratification of PCa lesions according to their clinical significance, but UA revealed differences in clinical assessment categories and radiomic features. We trained three classification models with fifteen feature subsets. We identified a subset of shape features which improved the diagnostic accuracy of the clinical assessment categories (maximum increase in diagnostic accuracy ΔAUC = + 0.05, p < 0.001) while also identifying combinations of features and models which reduced overall accuracy. Conclusions The impact of radiomic features to differentiate PCa lesions according to their clinical significance remains controversial. It depends on feature selection and the employed machine learning algorithms. It can result in improvement or reduction of diagnostic performance. Key Points • Quantitative imaging features differ between normal and malignant tissue of the peripheral zone in prostate cancer. • Radiomic feature analysis of clinical routine multiparametric MRI has the potential to improve the stratification of clinically significant versus insignificant prostate cancer lesions in the peripheral zone. • Certain combinations of standard multiparametric MRI reporting and assessment categories with feature subsets and machine learning algorithms reduced the diagnostic performance over standard clinical assessment categories alone.

scholarly journals Application of machine learning algorithms for multiparametric MRI-based evaluation of murine colitis

PLoS ONE ◽  
2018 ◽  
Vol 13 (10) ◽  
pp. e0206576 ◽  
Author(s):  
Stephan Ellmann ◽  
Victoria Langer ◽  
Nathalie Britzen-Laurent ◽  
Kai Hildner ◽  
Carina Huber ◽  
...  
Keyword(s):  
Machine Learning ◽  
Learning Algorithms ◽  
Multiparametric Mri ◽  
Machine Learning Algorithms ◽  
Murine Colitis

scholarly journals Prediction of Prostate Cancer using Machine Learning Algorithms

2020 ◽  
Vol 8 (5) ◽  
pp. 5353-5362
Keyword(s):  
Prostate Cancer ◽  
Machine Learning ◽  
Logistic Regression ◽  
Random Forest ◽  
Missing Values ◽  
Learning Algorithms ◽  
Machine Learning Algorithms ◽  
Support Vector ◽  
Cancer Disease ◽  
The Individual

Background/Aim: Prostate cancer is regarded as the most prevalent cancer in the word and the main cause of deaths worldwide. The early strategies for estimating the prostate cancer sicknesses helped in settling on choices about the progressions to have happened in high-chance patients which brought about the decrease of their dangers. Methods: In the proposed research, we have considered informational collection from kaggle and we have done pre-processing tasks for missing values .We have three missing data values in compactness attribute and two missing values in fractal dimension were replaced by mean of their column values .The performance of the diagnosis model is obtained by using methods like classification, accuracy, sensitivity and specificity analysis. This paper proposes a prediction model to predict whether a people have a prostate cancer disease or not and to provide an awareness or diagnosis on that. This is done by comparing the accuracies of applying rules to the individual results of Support Vector Machine, Random forest, Naive Bayes classifier and logistic regression on the dataset taken in a region to present an accurate model of predicting prostate cancer disease. Results: The machine learning algorithms under study were able to predict prostate cancer disease in patients with accuracy between 70% and 90%. Conclusions: It was shown that Logistic Regression and Random Forest both has better Accuracy (90%) when compared to different Machine-learning Algorithms.

Comparison of Machine Learning Algorithms for the Automated Detection of Functional Brain Networks in fMRI (Preprint)

2018 ◽  
Author(s):  
Jean-Michel Lem�e ◽  
Florian Bernard ◽  
Matthieu Labriffe ◽  
Philippe Menei ◽  
Aram Ter Minassian
Keyword(s):  
Machine Learning ◽  
Conjugate Gradient ◽  
Diagnostic Performance ◽  
Learning Algorithm ◽  
Learning Algorithms ◽  
Machine Learning Algorithms ◽  
Functional Networks ◽  
Machine Learning Algorithm ◽  
Functional Brain ◽  
Functional Brain Networks

BACKGROUND The functional MRI (fMRI) is an essential tool for the presurgical planning of brain tumor removal, allowing the identification of functional brain networks in order to preserve the patient’s neurological functions. One fMRI technique used to identify the functional brain network is the resting-state-fMRI (rsfMRI). However, this technique is not routinely used because of the necessity to have a expert reviewer to identify manually each functional networks. OBJECTIVE We aimed to automatize the detection of brain functional networks in rsfMRI data using machine learning algorithms. METHODS We used the rsfMRI data of 30 healthy patients to test the diagnostic performance of 10 machine learning algorithms compared to the reference functional networks identified manually by 2 expert reviewers. Then we selected the most fitted algorithm that we trained and tuned to optimize the diagnostic performance. RESULTS The comparison of the diagnostic performance of the machine learning algorithms identified the artificial neuron network using a scale conjugate gradient backpropagation as the most fitted algorithm. After training and fine tuning of the hyperparameters, the selected machine learning algorithm was able to identify correctly the different functional networks with an accuracy between 89 and 100%. CONCLUSIONS The artificial neural network using a scaled conjugate gradient backpropagation was the most performant machine learning algorithm. The use of this machine learning to automatize the functional networks detection in rsfMRI may allow to broaden the use of the rsfMRI, allowing the presurgical identification of these networks and thus help to preserve the patient’s neurological status.

scholarly journals Visibility of significant prostate cancer on multiparametric magnetic resonance imaging (MRI)—do we still need contrast media?

2020 ◽  
Author(s):  
Nicolai Alexander Huebner ◽  
Stephan Korn ◽  
Irene Resch ◽  
Bernhard Grubmüller ◽  
Tobias Gross ◽  
...  
Keyword(s):  
Prostate Cancer ◽  
Linear Regression ◽  
Clinical Decision Making ◽  
Peripheral Zone ◽  
Clinical Decision ◽  
Lesion Size ◽  
Multiparametric Mri ◽  
Characteristic Analysis ◽  
Visual Grading ◽  
Clinically Significant

Abstract Objectives To assess the visibility of clinically significant prostate cancer (PCA) lesions on the sequences multiparametric MRI of the prostate (mpMRI) and to evaluate whether the addition of dynamic contrast–enhanced imaging (DCE) improves the overall visibility. Methods We retrospectively evaluated multiparametric MRI images of 119 lesions in 111 patients with biopsy-proven clinically significant PCA. Three readers assigned visual grading scores for visibility on each sequence, and a visual grading characteristic analysis was performed. Linear regression was used to explore which factors contributed to visibility in individual sequences. Results The visibility of lesions was significantly better with mpMRI when compared to biparametric MRI in visual grading characteristic (VGC) analysis, with an AUCVGC of 0.62 (95% CI 0.55–0.69; p < 0.001). This benefit was seen across all readers. Multivariable linear regression revealed that a location in the peripheral zone was associated with better visibility on T2-weighted imaging (T2w). A higher Prostate Imaging-Reporting and Data System (PI-RADS) score was associated with better visibility on both diffusion-weighted imaging (DWI) and DCE. Increased lesion size was associated with better visibility on all sequences. Conclusions Visibility of clinically significant PCA is improved by using mpMRI. DCE and DWI images independently improve lesion visibility compared to T2w images alone. Further research into the potential of DCE to impact on clinical decision-making is suggested. Key Points • DCE and DWI images independently improve clinically significant prostate cancer lesion visibility compared to T2w images alone. • Multiparametric MRI (DCE, DWI, T2w) achieved significantly higher visibility scores than biparametric MRI (DWI, T2w). • Location in the transition zone is associated with poor visibility on T2w, while it did not affect visibility on DWI or DCE.

scholarly journals DWI of Prostate Cancer: Optimalb-Value in Clinical Practice

2014 ◽  
Vol 2014 ◽  
pp. 1-9 ◽  
Author(s):  
Guglielmo Manenti ◽  
Marco Nezzo ◽  
Fabrizio Chegai ◽  
Erald Vasili ◽  
Elena Bonanno ◽  
...  
Keyword(s):  
Prostate Cancer ◽  
Diffusion Weighted Imaging ◽  
Diagnostic Performance ◽  
Peripheral Zone ◽  
Transitional Zone ◽  
Young Reader ◽  
Mri Scan ◽  
Benign Tissue ◽  
Clinically Significant ◽  
Prostate Cancers

Aim. To compare the diagnostic performance of diffusion weighted imaging (DWI) usingb-values of 1000 s/mm2and 2000 s/mm2at 3 Tesla (T) for the evaluation of clinically significant prostate cancer.Matherials and Methods. Seventy-eight prostate cancer patients underwent a 3T MRI scan followed by radical prostatectomy. DWI was performed usingb-values of 0, 1000, and 2000 s/mm2and qualitatively analysed by two radiologists. ADC maps were obtained atb-values of 1000 and 2000 s/mm2and quantitatively analyzed in consensus.Results. For diagnosis of 78 prostate cancers the accuracy of DWI for the young reader was significantly greater atb= 2000 s/mm2for the peripheral zone (PZ) but not for the transitional zone (TZ). For the experienced reader, DWI did not show significant differences in accuracy betweenb-values of 1000 and 2000 s/mm2. The quantitative analysis in the PZ and TZ was substantially superimposable between the twob-values, albeit with a higher accuracy with ab-value of 2000 s/mm2.Conclusions. With ab-value of 2000 s/mm2at 3T both readers differentiated clinical significant cancer from benign tissue; higherb-values can be helpful for the less experienced readers.

Sign in / Sign up

Export Citation Format

Share Document