Application of Deep Learning to Pancreatic Cancer Detection: Lessons Learned From Our Initial Experience

Chu, Linda C.; Park, Seyoun; Kawamoto, Satomi; Wang, Yan; Zhou, Yuyin; Shen, Wei; Zhu, Zhuotun; Xia, Yingda; Xie, Lingxi; Liu, Fengze; Yu, Qihang; Fouladi, Daniel Fadaei; Shayesteh, Shahab; Zinreich, Eva S.; Graves, Jefferson S.; Horton, Karen M.; Yuille, Alan; Hruban, Ralph H.; Kinzler, Kenneth W.; Vogelstein, Bert; Fishman, Elliot K.

doi:10.1016/j.jacr.2019.05.034

Cited by 75 publications

(38 citation statements)

References 9 publications

(12 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Third, deep learning requires a large number of labelled images, but labelling of the pancreas and tumour is cumbersome and time consuming because of the organ's anatomical complexity. 27 To tackle these difficulties, we preprocessed CT images into patches and used tech niques including moving window and flipping as a means of data augmentation. Our patchbased analytical approach also had the advantage of allowing the CNN to interrogate the tumour multiple times, rather than once only if the pancreas and tumour was analysed as a whole.…”

Section: Discussionmentioning

confidence: 99%

“…In a preliminary report, deep learning differentiated between 156 patients with pancreatic cancer and 300 healthy participants with 94•1% sensitivity and 98•5% specificity, but all images were obtained using a standardised protocol on CT scanners from a single vendor and carefully matched on the basis of imaging parameters to minimise technical variations across images. 27 Therefore, the wide variations resulting from equipment, scanner, and various patient factors that radiologists inevitably encounter in clinical practice were excluded in that study, with unclear generalisability to other datasets. By contrast, this study included images obtained using six CT scanners from four major CT vendors, without selection based on imaging parameters.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Deep learning to distinguish pancreatic cancer tissue from non-cancerous pancreatic tissue: a retrospective study with cross-racial external validation

Liu

Chen

et al. 2020

The Lancet Digital Health

147

View full text Add to dashboard Cite

Background The diagnostic performance of CT for pancreatic cancer is interpreter-dependent, and approximately 40% of tumours smaller than 2 cm evade detection. Convolutional neural networks (CNNs) have shown promise in image analysis, but the networks' potential for pancreatic cancer detection and diagnosis is unclear. We aimed to investigate whether CNN could distinguish individuals with and without pancreatic cancer on CT, compared with radiologist interpretation. MethodsIn this retrospective, diagnostic study, contrast-enhanced CT images of 370 patients with pancreatic cancer and 320 controls from a Taiwanese centre were manually labelled and randomly divided for training and validation (295 patients with pancreatic cancer and 256 controls) and testing (75 patients with pancreatic cancer and 64 controls; local test set 1). Images were preprocessed into patches, and a CNN was trained to classify patches as cancerous or non-cancerous. Individuals were classified as with or without pancreatic cancer on the basis of the proportion of patches diagnosed as cancerous by the CNN, using a cutoff determined using the training and validation set. The CNN was further tested with another local test set (101 patients with pancreatic cancers and 88 controls; local test set 2) and a US dataset (281 pancreatic cancers and 82 controls). Radiologist reports of pancreatic cancer images in the local test sets were retrieved for comparison.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Deep learning to distinguish pancreatic cancer tissue from non-cancerous pancreatic tissue: a retrospective study with cross-racial external validation

Liu

Chen

et al. 2020

The Lancet Digital Health

147

View full text Add to dashboard Cite

show abstract

“…For example, Koay and colleagues (MD Anderson Cancer Center, Houston, TX) have shown that quantitative analysis of the tumor-stromal interface was a significant prognostic indicator in 3 different cohorts (N ¼ 303 total) of patients with PDAC treated with surgery or chemotherapy (76). Chu and colleagues (Felix project at Johns Hopkins, Baltimore, MD) have related their initial investigations using deep CNNs to identify normal pancreas (and other visceral tissues) as well as classifying PDAC as a case study (77). They trained the network on 575 control subjects and were able to achieve excellent segmentations of most major organs with accuracies >88% (compared with manual segmentations).…”

Section: Pancreatic Cancer (Ct Mri Pet/ct)mentioning

confidence: 99%

Radiomics Improves Cancer Screening and Early Detection

Gillies

Schabath

2020

Cancer Epidemiology, Biomarkers &Amp; Prevention

View full text Add to dashboard Cite

Imaging is a key technology in the early detection of cancers, including X-ray mammography, low-dose CT for lung cancer, or optical imaging for skin, esophageal, or colorectal cancers. Historically, imaging information in early detection schema was assessed qualitatively. However, the last decade has seen increased development of computerized tools that convert images into quantitative mineable data (radiomics), and their subsequent analyses with artificial intelligence (AI). These tools are improving diagnostic accuracy of early lesions to define risk and classify malignant/ aggressive from benign/indolent disease. The first section of this review will briefly describe the various imaging modalities and their use as primary or secondary screens in an early detection pipeline. The second section will describe specific use cases to illustrate the breadth of imaging modalities as well as the benefits of quantitative image analytics. These will include optical (skin cancer), X-ray CT (pancreatic and lung cancer), X-ray mammography (breast cancer), multiparametric MRI (breast and prostate cancer), PET (pancreatic cancer), and ultrasound elastography (liver cancer). Finally, we will discuss the inexorable improvements in radiomics to build more robust classifier models and the significant limitations to this development, including access to well-annotated databases, and biological descriptors of the imaged feature data. See all articles in this CEBP Focus section, "NCI Early Detection Research Network: Making Cancer Detection Possible."

show abstract

“…Reactions to the emergence of artificial intelligence (AI) in medicine, particularly in radiology, have ranged from excitement to fear. We have suggested in several places that radiologists who embrace novel deep learning and convolutional neural network applications should thrive despite some results suggesting that AI can occasionally outperform radiologists [1][2][3]. What does the radiology data suggest so far?…”

Section: Introductionmentioning

confidence: 98%

Man Versus Machine? Radiologists and Artificial Intelligence Work Better Together

Weisberg

Fishman

Chu

et al. 2021

Journal of the American College of Radiology

Self Cite

View full text Add to dashboard Cite

Application of Deep Learning to Pancreatic Cancer Detection: Lessons Learned From Our Initial Experience

Cited by 75 publications

References 9 publications

Deep learning to distinguish pancreatic cancer tissue from non-cancerous pancreatic tissue: a retrospective study with cross-racial external validation

Deep learning to distinguish pancreatic cancer tissue from non-cancerous pancreatic tissue: a retrospective study with cross-racial external validation

Radiomics Improves Cancer Screening and Early Detection

Man Versus Machine? Radiologists and Artificial Intelligence Work Better Together

Contact Info

Product

Resources

About