Deep transfer learning-based prostate cancer classification using 3 Tesla multi-parametric MRI

Zhong, Xinran; Cao, Ruiming; Shakeri, Sepideh; Scalzo, Fabien; Lee, Yeejin; Enzmann, Dieter R.; Wu, Holden H.; Raman, Steven S.; Sung, Kyunghyun

doi:10.1007/s00261-018-1824-5

Cited by 75 publications

(105 citation statements)

References 22 publications

Supporting

Mentioning

100

Contrasting

Unclassified

Order By: Relevance

“…The flow diagram is depicted in Figure 1 . In total, 27 articles were eligible for inclusion in this review [ 5 , 6 , 7 , 8 , 9 , 10 , 11 , 12 , 13 , 14 , 15 , 16 , 17 , 18 , 19 , 20 , 21 , 22 , 23 , 24 , 25 , 26 , 27 , 28 , 29 , 30 , 31 ]. From these, 13 studies reported enough information to perform a meta-analysis.…”

Section: Resultsmentioning

confidence: 99%

“…Nevertheless, the details regarding the selected populations were heterogenous between studies. Some papers limited their description to the number of positive and negative samples [27,31], while others mentioned the PCa lesion distribution per GS [11,23] and/or the volume distribution [16,21]. Both characteristics describe the type of population used and whether the PCa lesions are clinically relevant [2].…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Automated Classification of Significant Prostate Cancer on MRI: A Systematic Review on the Performance of Machine Learning Applications

Arif

Niessen

et al. 2020

Cancers

View full text Add to dashboard Cite

Significant prostate carcinoma (sPCa) classification based on MRI using radiomics or deep learning approaches has gained much interest, due to the potential application in assisting in clinical decision-making. Objective: To systematically review the literature (i) to determine which algorithms are most frequently used for sPCa classification, (ii) to investigate whether there exists a relation between the performance and the method or the MRI sequences used, (iii) to assess what study design factors affect the performance on sPCa classification, and (iv) to research whether performance had been evaluated in a clinical setting Methods: The databases Embase and Ovid MEDLINE were searched for studies describing machine learning or deep learning classification methods discriminating between significant and nonsignificant PCa on multiparametric MRI that performed a valid validation procedure. Quality was assessed by the modified radiomics quality score. We computed the median area under the receiver operating curve (AUC) from overall methods and the interquartile range. Results: From 2846 potentially relevant publications, 27 were included. The most frequent algorithms used in the literature for PCa classification are logistic regression (22%) and convolutional neural networks (CNNs) (22%). The median AUC was 0.79 (interquartile range: 0.77–0.87). No significant effect of number of included patients, image sequences, or reference standard on the reported performance was found. Three studies described an external validation and none of the papers described a validation in a prospective clinical trial. Conclusions: To unlock the promising potential of machine and deep learning approaches, validation studies and clinical prospective studies should be performed with an established protocol to assess the added value in decision-making.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Automated Classification of Significant Prostate Cancer on MRI: A Systematic Review on the Performance of Machine Learning Applications

Arif

Niessen

et al. 2020

Cancers

View full text Add to dashboard Cite

show abstract

“…For the task of distinguishing indolent PCas and csPCas, deep learning methods have shown some potential for future applications. Zhong et al [50] compared both a deep learning and a deep transfer-learning algorithm to the performance of PI-RADSv2. In their validation cohort of 30 patients with 47 lesions, it was comparable to the radiologist’s assessment (AUC 0.73 and 0.71, respectively), while outperforming deep learning alone (AUC 0.69).…”

Section: Applicationsmentioning

confidence: 99%

Machine learning applications in prostate cancer magnetic resonance imaging

et al. 2019

View full text Add to dashboard Cite

With this review, we aimed to provide a synopsis of recently proposed applications of machine learning (ML) in radiology focusing on prostate magnetic resonance imaging (MRI). After defining the difference between ML and classical rule-based algorithms and the distinction among supervised, unsupervised and reinforcement learning, we explain the characteristic of deep learning (DL), a particular new type of ML, including its structure mimicking human neural networks and its ‘black box’ nature. Differences in the pipeline for applying ML and DL to prostate MRI are highlighted. The following potential clinical applications in different settings are outlined, many of them based only on MRI-unenhanced sequences: gland segmentation; assessment of lesion aggressiveness to distinguish between clinically significant and indolent cancers, allowing for active surveillance; cancer detection/diagnosis and localisation (transition versus peripheral zone, use of prostate imaging reporting and data system (PI-RADS) version 2), reading reproducibility, differentiation of cancers from prostatitis benign hyperplasia; local staging and pre-treatment assessment (detection of extraprostatic disease extension, planning of radiation therapy); and prediction of biochemical recurrence. Results are promising, but clinical applicability still requires more robust validation across scanner vendors, field strengths and institutions.

show abstract

“…Our search retrieved 3107 potentially eligible articles. Of these, 58 were excluded because of duplication, and 3012 on the basis of title or abstract that was irrelevant to the topic, Of the remaining 37 articles, 22 were excluded after reading the full text, and finally 15 studies [ 4 , 11 – 14 , 26 – 35 ] were included in our review. Figure 1 provides an overview of the literature search and study selection.…”

Section: Resultsmentioning

confidence: 99%

“…The studies were published between 2007 and 2018. Four studies [ 13 , 28 , 33 , 35 ] were performed in the US, 3 in Germany, [ 11 , 14 , 29 ] 3 in China, [ 31 , 32 , 34 ] 2 in the Netherlands, [ 27 , 30 ] 1 in France, [ 4 ] and 2 in Italy. [ 12 , 26 ] The sample size ranged from 21 to 347.…”

Section: Resultsmentioning

confidence: 99%

Diagnostic accuracy of different computer-aided diagnostic systems for prostate cancer based on magnetic resonance imaging

et al. 2021

View full text Add to dashboard Cite

Background: Computer-aided detection (CAD) system for accurate and automated prostate cancer (PCa) diagnosis have been developed, however, the diagnostic test accuracy of different CAD systems is still controversial. This systematic review aimed to assess the diagnostic accuracy of CAD systems based on magnetic resonance imaging for PCa. Methods: Cochrane library, PubMed, EMBASE and China Biology Medicine disc were systematically searched until March 2019 for original diagnostic studies. Two independent reviewers selected studies on CAD based on magnetic resonance imaging diagnosis of PCa and extracted the requisite data. Pooled sensitivity, specificity, and the area under the summary receiver operating characteristic curve were calculated to estimate the diagnostic accuracy of CAD system. Results: Fifteen studies involving 1945 patients were included in our analysis. The diagnostic meta-analysis showed that overall sensitivity of CAD system ranged from 0.47 to 1.00 and, specificity from 0.47 to 0.89. The pooled sensitivity of CAD system was 0.87 (95% CI: 0.76–0.94), pooled specificity 0.76 (95% CI: 0.62–0.85), and the area under curve (AUC) 0.89 (95% CI: 0.86–0.91). Subgroup analysis showed that the support vector machines produced the best AUC among the CAD classifiers, with sensitivity ranging from 0.87 to 0.92, and specificity from 0.47 to 0.95. Among different zones of prostate, CAD system produced the best AUC in the transitional zone than the peripheral zone and central gland; sensitivity ranged from 0.89 to 1.00, and specificity from 0.38 to 0.85. Conclusions: CAD system can help improve the diagnostic accuracy of PCa especially using the support vector machines classifier. Whether the performance of the CAD system depends on the specific locations of the prostate needs further investigation.

show abstract

Deep transfer learning-based prostate cancer classification using 3 Tesla multi-parametric MRI

Cited by 75 publications

References 22 publications

Automated Classification of Significant Prostate Cancer on MRI: A Systematic Review on the Performance of Machine Learning Applications

Automated Classification of Significant Prostate Cancer on MRI: A Systematic Review on the Performance of Machine Learning Applications

Machine learning applications in prostate cancer magnetic resonance imaging

Diagnostic accuracy of different computer-aided diagnostic systems for prostate cancer based on magnetic resonance imaging

Contact Info

Product

Resources

About