1
|
Alanezi ST, Kraśny MJ, Kleefeld C, Colgan N. Differential Diagnosis of Prostate Cancer Grade to Augment Clinical Diagnosis Based on Classifier Models with Tuned Hyperparameters. Cancers (Basel) 2024; 16:2163. [PMID: 38893281 PMCID: PMC11171700 DOI: 10.3390/cancers16112163] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/20/2024] [Revised: 05/25/2024] [Accepted: 05/30/2024] [Indexed: 06/21/2024] Open
Abstract
We developed a novel machine-learning algorithm to augment the clinical diagnosis of prostate cancer utilizing first and second-order texture analysis metrics in a novel application of machine-learning radiomics analysis. We successfully discriminated between significant prostate cancers versus non-tumor regions and provided accurate prediction between Gleason score cohorts with statistical sensitivity of 0.82, 0.81 and 0.91 in three separate pathology classifications. Tumor heterogeneity and prediction of the Gleason score were quantified using two feature selection approaches and two separate classifiers with tuned hyperparameters. There was a total of 71 patients analyzed in this study. Multiparametric MRI, incorporating T2WI and ADC maps, were used to derive radiomics features. Recursive feature elimination (RFE), the least absolute shrinkage and selection operator (LASSO), and two classification approaches, incorporating a support vector machine (SVM) (with randomized search) and random forest (RF) (with grid search), were utilized to differentiate between non-tumor regions and significant cancer while also predicting the Gleason score. In T2WI images, the RFE feature selection approach combined with RF and SVM classifiers outperformed LASSO with SVM and RF classifiers. The best performance was achieved by combining LASSO and SVM into a model that used both T2WI and ADC images. This model had an area under the curve (AUC) of 0.91. Radiomic features computed from ADC and T2WI images were used to predict three groups of Gleason score using two kinds of feature selection methods (RFE and LASSO), RF and SVM classifier models with tuned hyperparameters. Using combined sequences (T2WI and ADC map images) and combined radiomics (1st and GLCM features), LASSO, with a feature selection method with RF, was able to predict G3 with the highest sensitivity at a level AUC of 0.92. To predict G3 for single sequence (T2WI images) using GLCM features, LASSO with SVM achieved the highest sensitivity with an AUC of 0.92.
Collapse
Affiliation(s)
- Saleh T. Alanezi
- Department of Physics, College of Science, Northern Border University, Arar P.O. Box 1321, Saudi Arabia
- Department of Physics, School of Natural Sciences, College of Science and Engineering, University of Galway, H91 TK33 Galway, Ireland; (M.J.K.); (C.K.); (N.C.)
| | - Marcin Jan Kraśny
- Department of Physics, School of Natural Sciences, College of Science and Engineering, University of Galway, H91 TK33 Galway, Ireland; (M.J.K.); (C.K.); (N.C.)
- Translational Medical Device Lab (TMDLab), Lambe Institute for Translational Research, University of Galway, H91 V4AY Galway, Ireland
| | - Christoph Kleefeld
- Department of Physics, School of Natural Sciences, College of Science and Engineering, University of Galway, H91 TK33 Galway, Ireland; (M.J.K.); (C.K.); (N.C.)
| | - Niall Colgan
- Department of Physics, School of Natural Sciences, College of Science and Engineering, University of Galway, H91 TK33 Galway, Ireland; (M.J.K.); (C.K.); (N.C.)
- Faculty of Engineering & Informatics, Technological University of the Shannon, N37 HD68 Athlone, Ireland
| |
Collapse
|
2
|
Ma J, Kou W, Lin M, Cho CCM, Chiu B. Multimodal Image Classification by Multiview Latent Pattern Extraction, Selection, and Correlation. IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS 2024; 35:8134-8148. [PMID: 37015566 DOI: 10.1109/tnnls.2022.3224946] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 06/19/2023]
Abstract
The large amount of data available in the modern big data era opens new opportunities to expand our knowledge by integrating information from heterogeneous sources. Multiview learning has recently achieved tremendous success in deriving complementary information from multiple data modalities. This article proposes a framework called multiview latent space projection (MVLSP) to integrate features extracted from multiple sources in a discriminative way to facilitate binary and multiclass classifications. Our approach is associated with three innovations. First, most existing multiview learning algorithms promote pairwise consistency between two views and do not have a natural extension to applications with more than two views. MVLSP finds optimum mappings from a common latent space to match the feature space in each of the views. As the matching is performed on a view-by-view basis, the framework can be readily extended to multiview applications. Second, feature selection in the common latent space can be readily achieved by adding a class view, which matches the latent space representations of training samples with their corresponding labels. Then, high-order view correlations are extracted by considering feature-label correlations. Third, a technique is proposed to optimize the integration of different latent patterns based on their correlations. The experimental results on the prostate image dataset demonstrate the effectiveness of the proposed method.
Collapse
|
3
|
Wang X, Li X, Chen H, Peng Y, Li Y. Pulmonary MRI Radiomics and Machine Learning: Effect of Intralesional Heterogeneity on Classification of Lesion. Acad Radiol 2022; 29 Suppl 2:S73-S81. [PMID: 33495072 DOI: 10.1016/j.acra.2020.12.020] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/04/2020] [Revised: 12/09/2020] [Accepted: 12/30/2020] [Indexed: 12/20/2022]
Abstract
RATIONALE AND OBJECTIVES To investigate the effect of intralesional heterogeneity on differentiating benign and malignant pulmonary lesions, quantitative magnetic resonance imaging (MRI) radiomics, and machine learning methods were adopted. MATERIALS AND METHODS A total of 176 patients with multiparametric MRI were involved in this exploratory study. To investigate the effect of intralesional heterogeneity on lesion classification, a radiomics model called tumor heterogeneity model was developed and compared to the conventional radiomics model based on the entire tumor. In tumor heterogeneity model, each lesion was divided into five sublesions depending on the spatial location through clustering algorithm. From the five sublesions in multi-parametric MRI sequences, 1100 radiomics features were extracted. The recursive feature elimination method was employed to select features and support vector machine classifier was used to distinguish benign and malignant lesion. The performance of classification was evaluated with the receiver operating characteristic curve and the area under the curve (AUC) was the figure of merit. The 3-fold cross-validation (CV) with and without nesting was used to validate the model, respectively. RESULTS The tumor heterogeneity model (AUC = 0.74 ± 0.04 and 0.90 ± 0.03, CV with and without nesting, respectively) outperforms conventional model (AUC = 0.68 ± 0.04 and 0.87 ± 0.03). The difference between the two models is statistically significant (p = 0.03) for lesions greater than 18.80 cm3. CONCLUSION Intralesional heterogeneity influences the classification of pulmonary lesions. The tumor heterogeneity model tends to perform better than conventional radiomics model.
Collapse
Affiliation(s)
- Xinhui Wang
- School of Electronic and Information Engineering, Beijing Jiaotong University, Shangyuan Village No 3 in Haidian, Beijing, China
| | - Xinchun Li
- Department of Radiology, The First Affiliated Hospital of Guangzhou Medical University, Guangzhou, China
| | - Houjin Chen
- School of Electronic and Information Engineering, Beijing Jiaotong University, Shangyuan Village No 3 in Haidian, Beijing, China.
| | - Yahui Peng
- School of Electronic and Information Engineering, Beijing Jiaotong University, Shangyuan Village No 3 in Haidian, Beijing, China
| | - Yanfeng Li
- School of Electronic and Information Engineering, Beijing Jiaotong University, Shangyuan Village No 3 in Haidian, Beijing, China
| |
Collapse
|
4
|
Lin M, Wynne JF, Zhou B, Wang T, Lei Y, Curran WJ, Liu T, Yang X. Artificial intelligence in tumor subregion analysis based on medical imaging: A review. J Appl Clin Med Phys 2021; 22:10-26. [PMID: 34164913 PMCID: PMC8292694 DOI: 10.1002/acm2.13321] [Citation(s) in RCA: 7] [Impact Index Per Article: 2.3] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/29/2020] [Revised: 04/17/2021] [Accepted: 05/22/2021] [Indexed: 12/20/2022] Open
Abstract
Medical imaging is widely used in the diagnosis and treatment of cancer, and artificial intelligence (AI) has achieved tremendous success in medical image analysis. This paper reviews AI-based tumor subregion analysis in medical imaging. We summarize the latest AI-based methods for tumor subregion analysis and their applications. Specifically, we categorize the AI-based methods by training strategy: supervised and unsupervised. A detailed review of each category is presented, highlighting important contributions and achievements. Specific challenges and potential applications of AI in tumor subregion analysis are discussed.
Collapse
Affiliation(s)
- Mingquan Lin
- Department of Radiation Oncology and Winship Cancer InstituteEmory UniversityAtlantaGeorgiaUSA
| | - Jacob F. Wynne
- Department of Radiation Oncology and Winship Cancer InstituteEmory UniversityAtlantaGeorgiaUSA
| | - Boran Zhou
- Department of Radiation Oncology and Winship Cancer InstituteEmory UniversityAtlantaGeorgiaUSA
| | - Tonghe Wang
- Department of Radiation Oncology and Winship Cancer InstituteEmory UniversityAtlantaGeorgiaUSA
| | - Yang Lei
- Department of Radiation Oncology and Winship Cancer InstituteEmory UniversityAtlantaGeorgiaUSA
| | - Walter J. Curran
- Department of Radiation Oncology and Winship Cancer InstituteEmory UniversityAtlantaGeorgiaUSA
| | - Tian Liu
- Department of Radiation Oncology and Winship Cancer InstituteEmory UniversityAtlantaGeorgiaUSA
| | - Xiaofeng Yang
- Department of Radiation Oncology and Winship Cancer InstituteEmory UniversityAtlantaGeorgiaUSA
| |
Collapse
|
5
|
Zhou Y, Mai Z, Yan W, Chen Y, Zhou Z, Xiao Y, Wang W, Shang Z, Yuan R, Ji Z, Li H. The characteristics and spatial distributions of prostate cancer in autopsy specimens. Prostate 2021; 81:135-141. [PMID: 33306857 DOI: 10.1002/pros.24091] [Citation(s) in RCA: 5] [Impact Index Per Article: 1.7] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 10/08/2020] [Revised: 11/23/2020] [Accepted: 11/25/2020] [Indexed: 12/24/2022]
Abstract
BACKGROUND The characteristics of prostate cancer on autopsy and early-stage prostate cancer are identical. Using autopsy specimens, we analysed prostate cancer characteristics and clarified the spatial distributions of lesions. METHOD We obtained prostate specimens from Chinese donors without a prostate cancer diagnosis and analyzed prostate cancer pathological characteristics on autopsy by whole-mount sampling. We determined the distributions of lesions in horizontal and vertical dimensions. The horizontal dimension included four horizontal quadrants (left-anterior, left-posterior, right-anterior, and right-posterior quadrants), the peripheral zone, and the transition zone. RESULT The overall positive rate of prostate cancer among 113 specimens was 35.4%. There were 73 lesions in 40 prostates with prostate cancer. The positive rates of lesions in the left-anterior, left-posterior, right-anterior, and right-posterior quadrants were 24.7% (18/73), 27.4% (20/73), 26.0% (19/73), and 21.9% (16/73), respectively. The positive rate of prostate cancer was 74% in the areas between the apex above 0.5-0.8 cm and the middle slice. There were 22 (30.1%) and 51 (69.9%) lesions in the superior and inferior half of the prostate. There were no significant differences in the median volume and Gleason grade group between the superior and inferior half (p = .876 and p = .228). CONCLUSION In the horizontal dimension, the positive rate of prostate cancer was consistent in the four quadrants. Prostate cancer mainly originated from the areas between the apex above 0.5-0.8 cm and the middle slice. Compared with the superior half, the inferior half of the prostate had a higher positive rate but the same lesion characteristics.
Collapse
Affiliation(s)
- Yi Zhou
- Department of Urology, Chinese Academy of Medical Sciences, Peking Union Medical College Hospital, Beijing, China
| | - Zhipeng Mai
- Department of Urology, Zhongshan City People's Hospital, Zhongshan, Guangdong, China
| | - Weigang Yan
- Department of Urology, Chinese Academy of Medical Sciences, Peking Union Medical College Hospital, Beijing, China
| | - Yuliang Chen
- Department of Urology, Chinese Academy of Medical Sciences, Peking Union Medical College Hospital, Beijing, China
| | - Zhien Zhou
- Department of Urology, Chinese Academy of Medical Sciences, Peking Union Medical College Hospital, Beijing, China
| | - Yu Xiao
- Department of Pathology, Chinese Academy of Medical Sciences, Peking Union Medical College Hospital, Beijing, China
| | - Wenze Wang
- Department of Pathology, Chinese Academy of Medical Sciences, Peking Union Medical College Hospital, Beijing, China
| | - Zhiyuan Shang
- Department of Urology, Chinese Academy of Medical Sciences, Peking Union Medical College Hospital, Beijing, China
| | - Runqiang Yuan
- Department of Urology, Zhongshan City People's Hospital, Zhongshan, Guangdong, China
| | - Zhigang Ji
- Department of Urology, Chinese Academy of Medical Sciences, Peking Union Medical College Hospital, Beijing, China
| | - Hanzhong Li
- Department of Urology, Chinese Academy of Medical Sciences, Peking Union Medical College Hospital, Beijing, China
| |
Collapse
|
6
|
Wang X, Wan Q, Chen H, Li Y, Li X. Classification of pulmonary lesion based on multiparametric MRI: utility of radiomics and comparison of machine learning methods. Eur Radiol 2020; 30:4595-4605. [PMID: 32222795 DOI: 10.1007/s00330-020-06768-y] [Citation(s) in RCA: 22] [Impact Index Per Article: 5.5] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/07/2019] [Revised: 01/23/2020] [Accepted: 02/20/2020] [Indexed: 11/26/2022]
Abstract
OBJECTIVES We develop and validate a radiomics model based on multiparametric magnetic resonance imaging (MRI) in the classification of the pulmonary lesion and identify optimal machine learning methods. MATERIALS AND METHODS This retrospective analysis included 201 patients (143 malignancies, 58 benign lesions). Radiomics features were extracted from multiparametric MRI, including T2-weighted imaging (T2WI), T1-weighted imaging (TIWI), and apparent diffusion coefficient (ADC) map. Three feature selection methods, including recursive feature elimination (RFE), t test, and least absolute shrinkage and selection operator (LASSO), and three classification methods, including linear discriminate analysis (LDA), support vector machine (SVM), and random forest (RF) were used to distinguish benign and malignant pulmonary lesions. Performance was compared by AUC, sensitivity, accuracy, precision, and specificity. Analysis of performance differences in three randomly drawn cross-validation sets verified the stability of the results. RESULTS For most single MR sequences or combinations of multiple MR sequences, RFE feature selection method with SVM classifier had the best performance, followed by RFE with RF. The radiomics model based on multiple sequences showed a higher diagnostic accuracy than single sequence for every machine learning method. Using RFE with SVM, the joint model of T1WI, T2WI, and ADC showed the highest performance with AUC = 0.88 ± 0.02 (sensitivity 83%; accuracy 82%; precision 91%; specificity 79%) in test set. CONCLUSION Quantitative radiomics features based on multiparametric MRI have good performance in differentiating lung malignancies and benign lesions. The machine learning method of RFE with SVM is superior to the combination of other feature selection and classifier methods. KEY POINTS • Radiomics approach has the potential to distinguish between benign and malignant pulmonary lesions. • Radiomics model based on multiparametric MRI has better performance than single-sequence models. • The machine learning methods RFE with SVM perform best in the current cohort.
Collapse
Affiliation(s)
- Xinhui Wang
- School of Electronic and Information Engineering, Beijing Jiaotong University, Shangyuan Village No 3 in Haidian, Beijing, China
| | - Qi Wan
- Department of Radiology, The First Affiliated Hospital of Guangzhou Medical University, Yanjiangxilu No 151 in Yuexiu, Guangzhou, China
| | - Houjin Chen
- School of Electronic and Information Engineering, Beijing Jiaotong University, Shangyuan Village No 3 in Haidian, Beijing, China.
| | - Yanfeng Li
- School of Electronic and Information Engineering, Beijing Jiaotong University, Shangyuan Village No 3 in Haidian, Beijing, China
| | - Xinchun Li
- Department of Radiology, The First Affiliated Hospital of Guangzhou Medical University, Yanjiangxilu No 151 in Yuexiu, Guangzhou, China.
| |
Collapse
|
7
|
Chen X, Lin M, Cui H, Chen Y, van Engelen A, de Bruijne M, Azarpazhooh MR, Sohrevardi SM, Chow TWS, Spence JD, Chiu B. Three-dimensional ultrasound evaluation of the effects of pomegranate therapy on carotid plaque texture using locality preserving projection. COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE 2020; 184:105276. [PMID: 31887617 DOI: 10.1016/j.cmpb.2019.105276] [Citation(s) in RCA: 3] [Impact Index Per Article: 0.8] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Received: 02/14/2019] [Revised: 11/19/2019] [Accepted: 12/11/2019] [Indexed: 06/10/2023]
Abstract
BACKGROUND AND OBJECTIVE Dietary supplements are expected to confer a smaller beneficial effect than medical treatments. Therefore, there is a need to develop cost-effective biomarkers that can demonstrate the efficacy of such supplements for carotid atherosclerosis. The aim of this study is to develop such a biomarker based on the changes of 376 plaque textural features measured from 3D ultrasound images. METHODS Since the number of features (376) was greater than the number of subjects (171) in this study, principal component analysis (PCA) was applied to reduce the dimensionality of feature vectors. To generate a scalar biomarker for each subject, elements in the reduced feature vectors produced by PCA were weighted using locality preserving projections (LPP) to capture essential patterns exhibited locally in the feature space. 96 subjects treated by pomegranate juice and tablets, and 75 subjects receiving placebo-matching juice and tablets were evaluated in this study. The discriminative power of the proposed biomarker was evaluated and compared with existing biomarkers using t-tests. As the cost of a clinical trial is directly related to the number of subjects enrolled, the cost-effectiveness of the proposed biomarker was evaluated by sample size estimation. RESULTS The proposed biomarker was more able to discriminate plaque changes exhibited by the pomegranate and placebo groups than total plaque volume (TPV) according to the result of t-tests (TPV: p=0.34, Proposed biomarker: p=1.5×10-5). The sample size required by the new biomarker to detect a significant effect was 20 times smaller than that required by TPV. CONCLUSION With the increase in cost-effectiveness afforded by the proposed biomarker, more proof-of-principle studies for novel treatment options could be performed.
Collapse
Affiliation(s)
- Xueli Chen
- Department of Electrical Engineering, City University of Hong Kong, 83 Tat Chee Avenue, Kowloon, Hong Kong
| | - Mingquan Lin
- Department of Electrical Engineering, City University of Hong Kong, 83 Tat Chee Avenue, Kowloon, Hong Kong
| | - He Cui
- Department of Electrical Engineering, City University of Hong Kong, 83 Tat Chee Avenue, Kowloon, Hong Kong
| | - Yimin Chen
- Department of Electrical Engineering, City University of Hong Kong, 83 Tat Chee Avenue, Kowloon, Hong Kong
| | - Arna van Engelen
- Biomedical Imaging Group Rotterdam, Department of Radiology and Nuclear Medicine, Erasmus MC, Rotterdam, the Netherlands
| | - Marleen de Bruijne
- Biomedical Imaging Group Rotterdam, Department of Radiology and Nuclear Medicine, Erasmus MC, Rotterdam, the Netherlands; Machine Learning Section, Department of Computer Science, University of Copenhagen, Denmark
| | - M Reza Azarpazhooh
- Stroke Prevention & Atherosclerosis Research Centre, Robarts Research Institute, London, Ontario, Canada
| | - Seyed Mojtaba Sohrevardi
- Stroke Prevention & Atherosclerosis Research Centre, Robarts Research Institute, London, Ontario, Canada; Pharmaceutical Sciences Research Center, Faculty of Pharmacy, Shahid Sadoughi University of Medical Sciences, Yazd, Iran
| | - Tommy W S Chow
- Department of Electrical Engineering, City University of Hong Kong, 83 Tat Chee Avenue, Kowloon, Hong Kong
| | - J David Spence
- Stroke Prevention & Atherosclerosis Research Centre, Robarts Research Institute, London, Ontario, Canada
| | - Bernard Chiu
- Department of Electrical Engineering, City University of Hong Kong, 83 Tat Chee Avenue, Kowloon, Hong Kong.
| |
Collapse
|
8
|
Lin M, Cui H, Chen W, van Engelen A, de Bruijne M, Azarpazhooh MR, Sohrevardi SM, Spence JD, Chiu B. Longitudinal assessment of carotid plaque texture in three-dimensional ultrasound images based on semi-supervised graph-based dimensionality reduction and feature selection. Comput Biol Med 2020; 116:103586. [DOI: 10.1016/j.compbiomed.2019.103586] [Citation(s) in RCA: 6] [Impact Index Per Article: 1.5] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/30/2019] [Revised: 11/25/2019] [Accepted: 12/13/2019] [Indexed: 11/28/2022]
|