Minireviews
Copyright ©The Author(s) 2022.
Artif Intell Med Imaging. Apr 28, 2022; 3(2): 21-32
Published online Apr 28, 2022. doi: 10.35711/aimi.v3.i2.21
Table 1 Studies on artificial intelligence using computed tomography or MRI imaging to diagnose pancreatic ductal adenocarcinoma
Ref.
Clinical question
Training set (number of subjects)
Validation set (number of subjects)
AI instrument
AUROC
Accuracy
Sensitivity
Specificity
Watson et al[66], 2021Detection of pancreatic cystic neoplasms (including PDAC) vs benign cysts189CNNNANANANA
Si et al[65], 2021Detection of pancreatic cancer (including PDAC, IPMN, PNET)319347DL0.87187.6% for PDAC86.8% for pancreatic cancer 69.5% for pancreatic cancer
Park et al[64], 2020Distinguishing pancreatic cancer tissue from autoimmune pancreatitis12062Random forest machine learning0.97595.2%89.7%100%
Ma et al[63], 2020Differentiate pancreatic cancer from benign tissue33041CNN0.9653 (plain scan)95.47% (plain scan),95.76% (arterial scan), 95.15% (venous phase)91.58% (plain scan), 94.08% (arterial scan), 92.28% (venous phase)98.3% (plain scan), 97.6% (arterial scan), 97.9% (venous phase)
Zhang et al[67], 2020Detection of pancreatic cancer 2650 images240 imagesCNN0.945590.2%83.8%91.8%
Liu et al[69], 2020Differentiating pancreatic cancer tissue from non-cancerous pancreatic tissue412139CNN0.9283.2%79.0%97.6%
Gao et al[71], 2020To differentiate pancreatic diseases in pancreatic lesions 398106CNN0.9035 (includes PDAC, adenosquamous carcinoma, acinar cell carcinoma, colloid carcinoma, myoepithelial carcinoma, undifferentiated carcinoma with osteoclast-like giant cells, mucinous cystadenocarcinoma, pancreatoblastoma, pancreatic neuroendocrine carcinoma and metastatic carcinoma)NANA NA
Chu et al[70], 2019Differentiating PDAC from normal pancreas255125Random forestNA93.6%95%92.3%
Zhu et al[72], 2019Detecting PDAC from normal pancreas205234CNNNA57.3%94.1%98.5%
Liu et al[73], 2019Diagnosis of pancreatic cancer238100CNN0.9632NANANA
Corral et al[21], 2019Identify and stratify IPMN lesions 139DL0.783NA75% (for PDAC or high grade dysplasia)78% (for PDAC or high grade dysplasia)
Chu et al[74], 2019Differentiating PDAC from normal pancreas456DLNANA94.1%98.5%
Fu et al[75], 2018Pancreas segmentation (including PDAC, IPMN, Pancreatic Neuroendocrine Tumors, Serous Cyst Adenoma, and Solid Pseudopapillary Tumour of the pancreas)59CNNNANA82.5%76.22 (PPV)
Table 2 Studies on artificial intelligence using endoscopic ultrasound to diagnose pancreatic ductal adenocarcinoma
Ref.Clinical questionTraining set (number of subjects)Validation set (number of subjects)AI instrument AUROCAccuracySensitivity Specificity
Udristoiu et al[84], 2021Detecting focal pancreatic masses in four EUS imaging modalities65CNN and Long Short-term Memory models0.9797.6%98.1%96.7%
Tonozuka et al[83], 2021Detecting PDAC in patients with normal pancreas/Chronic pancreatitis92CNN0.924NA90.2%74.9%
Marya et al[78], 2021Differentiate AIP from PDAC, chronic pancreatitis and other pancreatic diseases336124CNN0.976NA95%90%
Kuwahara et al[77], 2019Predicting malignancy in IPMN50CNN0.9894%95.7%92.6%
Ozkan et al[80], 2016Differentiating pancreatic cancer from healthy pancreas260 images 72 images ANNNA87.5%83.3%93.3%
Saftoiu et al[81], 2015Differentiate pancreatic cancer from chronic pancreatitis11725ANNNANA94.6%94.4%
Zhu et al[86], 2013Differentiating pancreatic cancer from chronic pancreatitis. 194194SVMNA94.2%96.3%93.4%
Saftoiu et al[82], 2012Diagnosis of focal pancreatic lesions 258 patients ANN0.9484.27%87.59%82.94%
Zhang et al[85], 2010Differentiate pancreatic cancer from non-tumorous tissue108108SVMNA97.98%94.3%99.45%
Saftoiu et al[20], 2008 cancerDifferentiate normal pancreas, chronic pancreatitis, pancreatic cancer, and neuroendocrine tumors68Neural network0.847 (for PDAC vs chronic pan-creatitis)86.1% (for PDAC vs chronic pan-creatitis)93.8% (for PDAC vs chronic pan-creatitis)63.6% (for PDAC vs chronic pan-creatitis)
Das et al[19], 2008Differentiating pancreatic adenocarcinoma from non-neoplastic tissue (includes normal pancreas and chronic pancreatitis)160159ANN0.93NA93%92%
Norton et al[79], 2001Differentiate malignancy from pancreatitis35MLNA80%100%50%