Multiple instance learning combined with label invariant synthetic data for guiding systematic prostate biopsy: a feasibility study

被引:11
作者
Javadi, Golara [1 ]
Samadi, Samareh [1 ]
Bayat, Sharareh [1 ]
Pesteie, Mehran [1 ]
Jafari, Mohammad H. [1 ]
Sojoudi, Samira [1 ]
Kesch, Claudia [3 ]
Hurtado, Antonio [3 ]
Chang, Silvia [3 ]
Mousavi, Parvin [2 ]
Black, Peter [3 ]
Abolmaesumi, Purang [1 ]
机构
[1] Univ British Columbia, Vancouver, BC, Canada
[2] Queens Univ, Kingston, ON, Canada
[3] Vancouver Gen Hosp, Vancouver, BC, Canada
基金
加拿大自然科学与工程研究理事会; 加拿大健康研究院;
关键词
Temporal enhanced ultrasound; Deep neural networks; Augmentation; Multiple instance learning; Systematic biopsy; Prostate cancer; ULTRASOUND; CANCER; DIAGNOSIS; MRI;
D O I
10.1007/s11548-020-02168-1
中图分类号
R318 [生物医学工程];
学科分类号
0831 ;
摘要
Purpose Ultrasound imaging is routinely used in prostate biopsy, which involves obtaining prostate tissue samples using a systematic, yet, non-targeted approach. This approach is blinded to individual patient intraprostatic pathology, and unfortunately, has a high rate of false negatives. Methods In this paper, we propose a deep network for improved detection of prostate cancer in systematic biopsy. We address several challenges associated with training such network: (1) Statistical labels: Since biopsy core's pathology report only represents a statistical distribution of cancer within the core, we use multiple instance learning (MIL) networks to enable learning from ultrasound image regions associated with those data; (2) Limited labels: The number of biopsy cores are limited to at most 12 per patient. As a result, the number of samples available for training a deep network is limited. We alleviate this issue by effectively combining Independent Conditional Variational Auto Encoders (ICVAE) with MIL. We train ICVAE to learn label-invariant features of RF data, which is subsequently used to generate synthetic data for improved training of the MIL network. Results Our in vivo study includes data from 339 prostate biopsy cores of 70 patients. We achieve an area under the curve, sensitivity, specificity, and balanced accuracy of 0.68, 0.77, 0.55 and 0.66, respectively. Conclusion The proposed approach is generic and can be applied to several other scenarios where unlabeled data and noisy labels in training samples are present.
引用
收藏
页码:1023 / 1031
页数:9
相关论文
共 36 条
[1]   Diagnostic accuracy of multi-parametric MRI and TRUS biopsy in prostate cancer (PROMIS): a paired validating confirmatory study [J].
Ahmed, Hashim U. ;
Bosaily, Ahmed El-Shater ;
Brown, Louise C. ;
Gabe, Rhian ;
Kaplan, Richard ;
Parmar, Mahesh K. ;
Collaco-Moraes, Yolanda ;
Ward, Katie ;
Hindley, Richard G. ;
Freeman, Alex ;
Kirkham, Alex P. ;
Oldroyd, Robert ;
Parker, Chris ;
Emberton, Mark .
LANCET, 2017, 389 (10071) :815-822
[2]   Deep Recurrent Neural Networks for Prostate Cancer Detection: Analysis of Temporal Enhanced Ultrasound [J].
Azizi, Shekoofeh ;
Bayat, Sharareh ;
Yan, Pingkun ;
Tahmasebi, Amir ;
Kwak, Jin Tae ;
Xu, Sheng ;
Turkbey, Baris ;
Choyke, Peter ;
Pinto, Peter ;
Wood, Bradford ;
Mousavi, Parvin ;
Abolmaesumi, Purang .
IEEE TRANSACTIONS ON MEDICAL IMAGING, 2018, 37 (12) :2695-2703
[3]   Ultrasound-Based Detection of Prostate Cancer Using Automatic Feature Selection with Deep Belief Networks [J].
Azizi, Shekoofeh ;
Imani, Farhad ;
Zhuang, Bo ;
Tahmasebi, Amir ;
Kwak, Jin Tae ;
Xu, Sheng ;
Uniyal, Nishant ;
Turkbey, Baris ;
Choyke, Peter ;
Pinto, Peter ;
Wood, Bradford ;
Moradi, Mehdi ;
Mousavi, Parvin ;
Abolmaesumi, Purang .
MEDICAL IMAGE COMPUTING AND COMPUTER-ASSISTED INTERVENTION - MICCAI 2015, PT II, 2015, 9350 :70-77
[4]   Investigation of Physical Phenomena Underlying Temporal-Enhanced Ultrasound as a New Diagnostic Imaging Technique: Theory and Simulations [J].
Bayat, Sharareh ;
Azizi, Shekoofeh ;
Daoud, Mohammad I. ;
Nir, Guy ;
Imani, Farhad ;
Gerardo, Carlos D. ;
Yan, Pingkun ;
Tahmasebi, Amir ;
Vignon, Francois ;
Sojoudi, Samira ;
Wilson, Storey ;
Iczkowski, Kenneth A. ;
Lucia, M. Scott ;
Goldenberg, Larry ;
Salcudean, Septimiu E. ;
Abolmaesumi, Purang ;
Mousavi, Parvin .
IEEE TRANSACTIONS ON ULTRASONICS FERROELECTRICS AND FREQUENCY CONTROL, 2018, 65 (03) :400-410
[5]   Tissue mimicking simulations for temporal enhanced ultrasound-based tissue typing [J].
Bayat, Sharareh ;
Imani, Farhad ;
Gerardo, Carlos D. ;
Nir, Guy ;
Azizi, Shekoofeh ;
Yan, Pingkun ;
Tahmasebi, Amir ;
Wilson, Storey ;
Iczkowski, Kenneth A. ;
Luci, M. Scott ;
Goldenberg, Larry ;
Salcudean, Septimiu E. ;
Mousavi, Parvin ;
Abolmaesumi, Purang .
MEDICAL IMAGING 2017: ULTRASONIC IMAGING AND TOMOGRAPHY, 2017, 10139
[6]   Standards for prostate biopsy [J].
Bjurlin, Marc A. ;
Taneja, Samir S. .
CURRENT OPINION IN UROLOGY, 2014, 24 (02) :155-161
[7]   Deep, Big, Simple Neural Nets for Handwritten Digit Recognition [J].
Ciresan, Dan Claudiu ;
Meier, Ueli ;
Gambardella, Luca Maria ;
Schmidhuber, Juergen .
NEURAL COMPUTATION, 2010, 22 (12) :3207-3220
[8]   Ultrasound elastography of the prostate: state of the art [J].
Correas, J. -M. ;
Tissier, A. -M. ;
Khairoune, A. ;
Khoury, G. ;
Eiss, D. ;
Helenon, O. .
JOURNAL DE RADIOLOGIE DIAGNOSTIQUE ET INTERVENTIONNELLE, 2013, 94 (05) :567-576
[9]  
DeVries T., 2017, 5th Inter- national Conference on Learning Representations, ICLR
[10]   Solving the multiple instance problem with axis-parallel rectangles [J].
Dietterich, TG ;
Lathrop, RH ;
LozanoPerez, T .
ARTIFICIAL INTELLIGENCE, 1997, 89 (1-2) :31-71