Enhancement of prostate cancer diagnosis by machine learning techniques: an algorithm development and validation study

Chiu, Peter Ka‐Fung; Shen, Xiao; Wang, Guanjin; Ho, Cho-Lik; Leung, S. F.; Ng, Chi‐Fai; Choi, Kup‐Sze; Teoh, Jeremy Yuen‐Chun

doi:10.1038/s41391-021-00429-x

Cited by 25 publications

(21 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This finding was consistent with that of P. Chiu et al. ( 35 ). An RF algorithm ( 36 ) builds each tree independently in parallel and integrates the results at the end, contributing to noise reduction and prediction accuracy improvement.…”

Section: Discussionsupporting

confidence: 94%

Radiomics-Based Machine Learning Models for Predicting P504s/P63 Immunohistochemical Expression: A Noninvasive Diagnostic Tool for Prostate Cancer

et al. 2022

View full text Add to dashboard Cite

ObjectiveTo develop and validate a noninvasive radiomic-based machine learning (ML) model to identify P504s/P63 status and further achieve the diagnosis of prostate cancer (PCa).MethodsA retrospective dataset of patients with preoperative prostate MRI examination and P504s/P63 pathological immunohistochemical results between June 2016 and February 2021 was conducted. As indicated by P504s/P63 expression, the patients were divided into label 0 (atypical prostatic hyperplasia), label 1 (benign prostatic hyperplasia, BPH) and label 2 (PCa) groups. This study employed T2WI, DWI and ADC sequences to assess prostate diseases and manually segmented regions of interest (ROIs) with Artificial Intelligence Kit software for radiomics feature acquisition. Feature dimensionality reduction and selection were performed by using a mutual information algorithm. Based on screened features, P504s/P63 prediction models were established by random forest (RF), gradient boosting decision tree (GBDT), logistic regression (LR), adaptive boosting (AdaBoost) and k-nearest neighbor (KNN) algorithms. The performance was evaluated by the area under the ROC curve (AUC) and accuracy.ResultsA total of 315 patients were enrolled. Among the 851 radiomic features, the 32 top features were derived from T2WI, in which the gray-level run length matrix (GLRLM) and gray-level cooccurrence matrix (GLCM) features accounted for the largest proportion. Among the five models, the RF algorithm performed best in general evaluations (microaverage AUC=0.920, macroaverage AUC=0.870) and provided the most accurate result in further sublabel prediction (the accuracies of label 0, 1, and 2 were 0.831, 0.831, and 0.932, respectively). In comparative sequence analyses, T2WI was the best single-sequence candidate (microaverage AUC=0.94 and macroaverage AUC=0.78). The merged datasets of T2WI, DWI, and ADC yielded optimal AUCs (microaverage AUC=0.930 and macroaverage AUC=0.900).ConclusionsThe radiomic-based RF classifier has the potential to be used to evaluate the presurgical P504s/P63 status and further diagnose PCa noninvasively and accurately.

show abstract

Section: Discussionsupporting

confidence: 94%

Radiomics-Based Machine Learning Models for Predicting P504s/P63 Immunohistochemical Expression: A Noninvasive Diagnostic Tool for Prostate Cancer

et al. 2022

View full text Add to dashboard Cite

show abstract

“…More research is urgently needed to establish a reliable risk model for predicting EAOC in patients with endometriosis 27 . To our delight, machine learning has been widely used in clinical differentiation and early diagnosis, such as Parkinson's disease, 28 diabetic kidney disease, 29 non‐alcoholic fatty liver disease, 30 and prostate cancer diagnosis 31 . Compared with traditional methods, the use of machine learning algorithms has advantages for modeling and validation.…”

Section: Discussionmentioning

confidence: 99%

“… 27 To our delight, machine learning has been widely used in clinical differentiation and early diagnosis, such as Parkinson's disease, 28 diabetic kidney disease, 29 non‐alcoholic fatty liver disease, 30 and prostate cancer diagnosis. 31 Compared with traditional methods, the use of machine learning algorithms has advantages for modeling and validation. A random subsampling scheme was used to minimize the estimated bias.…”

Section: Discussionmentioning

confidence: 99%

The application of risk models based on machine learning to predict endometriosis‐associated ovarian cancer in patients with endometriosis

Chao

Wang

Lang

et al. 2022

Acta Obstet Gynecol Scand

View full text Add to dashboard Cite

Introduction:There is currently no satisfactory model for predicting malignant transformation of endometriosis. The aim of this study was to construct and evaluate a risk model incorporating noninvasive clinical parameters to predict endometriosisassociated ovarian cancer (EAOC) in patients with endometriosis. Material and Methods:We enrolled 6809 patients with endometriosis confirmed by pathology, and randomly allocated them to training (n = 4766) and testing cohorts (n = 2043). The proportion of patients with EAOC in each cohort was similar. We extracted a total of 94 demographic and clinicopathologic features from the medical records using natural language processing. We used a machine learning methodgradient-boosting decision tree -to construct a predictive model for EAOC and to evaluate the accuracy of the model. We also constructed a multivariate logistic regression model inclusive of the EAOC-associated risk factors using a back stepwise procedure. Then we compared the performance of the two risk-predicting models using DeLong's test. Results:The occurrence of EAOC was 1.84% in this study. The logistic regression model comprised 10 selected features and demonstrated good discrimination in the testing cohort, with an area under the curve (AUC) of 0.891 (95% confidence interval[CI] 0.821-0.960), sensitivity of 88.9%, and specificity of 76.7%. The risk model based on machine learning had an AUC of 0.942 (95% CI 0.914-0.969), sensitivity of 86.8%, and specificity of 86.7%. The machine learning-based risk model performed better than the logistic regression model in DeLong's test (p = 0.036). Furthermore, in a prospective dataset, the machine learning-based risk model had an AUC of 0.8758, a sensitivity of 94.4%, and a specificity of 73.8%. Conclusions:The machine learning-based risk model was constructed to predict EAOC and had high sensitivity and specificity. This model could be of considerable use in helping reduce medical costs and designing follow-up schedules.

show abstract

“…In the study of Peter Ka-Fung Chiu et al., four variables, PSA, DRE, PV and transrectal ultrasound findings, were included, and SVM, LR, and RF models were constructed. All models were shown to have better prediction for PCa and clinically significant PCa than PSA and PSAD alone ( 19 ). Similarly, Nitta et al.…”

Section: Discussionmentioning

confidence: 99%

Machine Learning-Based Models Enhance the Prediction of Prostate Cancer

et al. 2022

View full text Add to dashboard Cite

PurposePSA is currently the most commonly used screening indicator for prostate cancer. However, it has limited specificity for the diagnosis of prostate cancer. We aim to construct machine learning-based models and enhance the prediction of prostate cancer.MethodsThe data of 551 patients who underwent prostate biopsy were retrospectively retrieved and divided into training and test datasets in a 3:1 ratio. We constructed five PCa prediction models with four supervised machine learning algorithms, including tPSA univariate logistic regression (LR), multivariate LR, decision tree (DT), random forest (RF), and support vector machine (SVM). The five prediction models were compared based on model performance metrics, such as the area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, specificity, calibration curve, and clinical decision curve analysis (DCA).ResultsAll five models had good calibration in the training dataset. In the training dataset, the RF, DT, and multivariate LR models showed better discrimination, with AUCs of 1.0, 0.922 and 0.91, respectively, than the tPSA univariate LR and SVM models. In the test dataset, the multivariate LR model exhibited the best discrimination (AUC=0.918). The multivariate LR model and SVM model had better extrapolation and generalizability, with little change in performance between the training and test datasets. Compared with the DCA curves of the tPSA LR model, the other four models exhibited better net clinical benefits.ConclusionThe results of the current retrospective study suggest that machine learning techniques can predict prostate cancer with significantly better AUC, accuracy, and net clinical benefits.

show abstract

Enhancement of prostate cancer diagnosis by machine learning techniques: an algorithm development and validation study

Cited by 25 publications

References 20 publications

Radiomics-Based Machine Learning Models for Predicting P504s/P63 Immunohistochemical Expression: A Noninvasive Diagnostic Tool for Prostate Cancer

Radiomics-Based Machine Learning Models for Predicting P504s/P63 Immunohistochemical Expression: A Noninvasive Diagnostic Tool for Prostate Cancer

The application of risk models based on machine learning to predict endometriosis‐associated ovarian cancer in patients with endometriosis

Machine Learning-Based Models Enhance the Prediction of Prostate Cancer

Contact Info

Product

Resources

About